capabilities and competitive advantage in the forest industry

CAPABILITIES AND COMPETITIVE ADVANTAGE IN THE FOREST INDUSTRY

Edited by

JUHA-ANTTI LAMBERG

Aalto University School of Science and Technology

MIRVA PELTONIEMI

Aalto University School of Science and Technology

AALTO UNIVERSITY SCHOOL OF SCIENCE AND TECHNOLOGY

FACULTY OF INFORMATION AND NATURAL SCIENCES

DEPARTMENT OF INDUSTRIAL ENGINEERING AND MANAGEMENT

1

FOREWORD

The research reported in this publication was conducted in "Capabilities and competitive

advantage in the forest industry" project funded by ForestCluster Ltd and Finnish Funding

Agency for Technology and Innovation (Tekes). The project was a part of Radical Market

Innovations Program managed by ForestCluster Ltd. The research was undertaken between

June 2009 and May 2010.

We thank Kaija Pehu-Lehtonen (MetsäBotnia), Esa Torniainen (VTT), and Suvi Nenonen

(Vectia) for their insightful comments and professionalism during the research process.

Moreover, we gratefully acknowledge support received from Rector Matti Pursula (TKK),

Christine Hagström-Näsi (ForestCluster Ltd), Lars Gädda (ForestCluster Ltd), and Markku

Leskelä prior, during, and after the project. Finally, the research work benefited from active

communications with dozens of forest industry specialists. We have an impossible mission to

credit each of them individually. Yet, their willingness to help hugely benefited our research

efforts.

Editors

2

LIST OF CONTRIBUTORS

ARNE KÖHLER, M.Sc. (Tech.), is Researcher at Aalto University School of Science and

Technology, at Institute of Strategy. He recently graduated from Department of Industrial

Engineering and Management, majoring in strategy and international business while attaining

a minor in applied mathematics. His major academic interests include competitive dynamics

and organizational ecology.

B.Sc. (Tech.) OLA LAAKSONEN studies towards a Master of Science in Technology degree at

Aalto University School of Science and Technology. He is interested in the dynamics of the

forest industry, social network analysis and institutional theory.

DR. JUHA-ANTTI LAMBERG is Professor of Strategic Management at Aalto University School

of Science and Technology. His research interests include continuity and change in strategy;

especially in the contexts of retail industry and paper industry. He has published articles in, for

example, Strategic Management Journal, Industrial and Corporate Change, Journal of

Management Studies, and Organization Studies. Prof. Lamberg won the Carolyn Dexter Award

for best international paper in the 2009 Academy of Management conference and the Sloan

Foundation’s Industry Studies Best Paper Prize in the 2008 Academy of Management

conference.

M.Sc. (Tech.) JAAKKO LINNAKANGAS is a graduate from the Department of Industrial

Engineering and Management, Aalto University School of Science and Technology. He majored

in strategic management with computer sciences as his minor. His main interests are related to

technology management and strategic marketing.

DR. MIRVA PELTONIEMI works as Postdoctoral Researcher at Institute of Strategy, Aalto

University School of Science and Technology. She has Doctor of Science in Technology degree

from Tampere University of Technology. Her thesis research focused on the evolution of the

games industry. Currently she works on industry evolution in several empirical settings,

including games, retail and forest industries.

3

M.Sc. (Business) ANTTI SIHVONEN works as Researcher at Department of Marketing and

Management, Aalto University School of Economics. His research interests include strategic

marketing, dynamic capabilities and resource-based competition. He is associated with

StratMark and GloStra research groups.

M.Sc. (Business) ULRIIKKA TIKKANEN is a graduate form Turku School of Economics.

Currently she works at Institute of Strategy, Aalto University School of Science and

Technology. She is interested in knowledge management, innovation dynamics and

stakeholder theory.

4

CONTENTS

FOREWORD 1

LIST OF CONTRIBUTORS 2

CONTENTS 4

CHAPTER 1

EXECUTIVE SUMMARY 6

JUHA-ANTTI LAMBERG AND MIRVA PELTONIEMI

CHAPTER 2

INTRODUCTION 12

MIRVA PELTONIEMI AND ARNE KÖHLER

CHAPTER 3

THE STRUCTURE AND EVOLUTION OF CAPABILITIES RESEARCH IN 1986–2009: A BIBLIOMETRIC STUDY 25

OLA LAAKSONEN

CHAPTER 4

OPERATIONALIZATION OF CAPABILITIES: A REVIEW 67

MIRVA PELTONIEMI, OLA LAAKSONEN AND ULRIIKKA TIKKANEN

5

CHAPTER 5

CAPABILITY EVOLUTION IN THE FINNISH FOREST CLUSTER: A QUALITATIVE APPROACH 81

ULRIIKKA TIKKANEN

CHAPTER 6

EVOLUTION OF CAPABILITIES IN THE FINNISH FOREST INDUSTRY: A QUANTITATIVE APPROACH 135

JAAKKO LINNAKANGAS

CHAPTER 7

DYNAMIC CAPABILITIES OF EXPLOITATION AND EXPLORATION IN COMPETITIVE SETTINGS 208

ARNE KÖHLER

CHAPTER 8

SUCCESS STRATEGIES IN DECLINING INDUSTRIES: A CASE SURVEY 267

ANTTI SIHVONEN

6

CHAPTER 1

CAPABILITIES AND COMPETITIVE ADVANTAGE IN THE FOREST

INDUSTRY: EXECUTIVE SUMMARY

JUHA-ANTTI LAMBERG

Aalto University School of Science and Technology [email protected]

MIRVA PELTONIEMI


Capabilities are those knowledge, skills, and competencies that allow firms to operate in the

market place. What makes capabilities a challenge is the fact that they materialize only in

realized actions (patents, investments, sold products etc.). Otherwise, they are embedded in

processes, and routine-like interactions. From this vantage point, our research project focused

on the following questions:

1. What capabilities does the Forest Cluster possess?

2. What are the strengths and weaknesses compared to international competitors?

3. How and why capabilities interact with cluster evolution?

4. How capabilities enable (and hinder) innovation and renewal?

Regarding the first research question, the Finnish Forest Cluster demonstrates strong

knowledge and skills in all areas of pulp and paper industry. Equally, as a result of decades’ long

7

experience in international business firms posses deep managerial knowledge and routines

related to supply chain management on a global scale. However, when we compare Finnish

firms to other industries or even to other firms in the same industry yet from a different

geographic area, we may witness alarming narrowness in the expressed set of capabilities.

Regarding the second research question, the comparative analysis reported in this book,

supported by earlier similar studies, shows that the Finnish / Scandinavian cluster is clearly the

most viable actor in the global paper industry. A worrying example, however, is the rapid

erosion of the North American cluster which culminated in the demise of Beloit in the late

1990s. At the moment, Scandinavian cluster is still the dynamic ‘engine’ of the whole industry

being forerunner in new product launches, innovative activities, and a vivid network of a

variety of organizations. Relative to the North American story, heterogeneity of organizations

and activities can be seen as trademarks of healthy development for the cluster as an entity.

However, as a variety of existing cluster evolution studies exhibits: competitive advantage at

cluster level is extremely vulnerable when firms turn from development mode to low-cost

strategy.

Regarding the third research question, our theoretical work in conjunction with the results

from our international comparative work, demonstrate that variation in capability portfolio is a

necessary antecedent of successful cluster evolution. However, it is both conceptually, and

from the point of view of cluster policy, problematic to derive cluster level ‘capability portfolio’

from the capabilities identified in the firm level studies. That is, firm-specific capabilities do not

necessarily contribute to cluster level vitality. Vice versa, cluster level capabilities may not

contribute to firm level competitive success. Logically, the link exist yet (a) it does not neatly

materialize as empirical observations (the idea in capabilities really is that they are causally

ambiguous); and (b) the interests of different actors may collide (and often do) in ways that

prevent the use of shared knowledge and skills. Building on existing studies in cluster literature

(especially Menzel and Fornahl, 2010), we may reason that the fundamental measure of cluster

level capabilities is the number of employees in the focal object of research. That is, individual

employees (from paper engineers to sales personnel) are the repository of capabilities. The

problem is that all these capabilities are tied to firm level activities, processes, and structures

not easily accessible when the interests of the whole cluster are at stake. This means that

cluster level policy making (strategy in other words) needs to address the links between

existing and new organizations as a way to keep the cluster vital.

8

Regarding the fourth research question, existing capabilities are the stock of innovations,

renewal, and inertia. New capabilities emerge as we write yet capability evolution takes time,

capabilities are always embedded in existing architectures, and are difficult to identify even by

the actors themselves (i.e. it is relatively easy to acquire knowledge about an individual’s

knowledge and skills yet no one literally knows what an organization can do expect focusing on

the realized activities). Thus, any viable strategy for the future must be based on the existing

capabilities simultaneously understanding the dynamics of the market place. The few existing

studies on cluster renewal (especially studies focusing on the rejuvenation of the Ruhr are

metal industry) demonstrate that (a) birth of new firms and organizations is the way to

facilitate cluster renewal as new organizational actors create the new business models which

change the ways business is conducted, and raise the overall level of activity at the cluster level;

and (b) changes in the cluster and firm architectures (i.e. structures and incentives that drive

activities to certain directions) allow the full use of existing capabilities or even awakening of

‘dormant capabilities’. Such architectural innovations, could be, for example, organized

markets for ideas (allowing transactions between organizations that have ideas but no

capabilities with organizations that posses capabilities but no ideas); massive emergence of

spin-off companies; reduction or lowering boundaries between different clusters (e.g. between

SHOKs in the Finnish context); and active creation of inter-organizational projects (‘virtual

firms’) along shared business ideas. Finally, a powerful yet hardly controllable mechanism for

renewal is increasing competition. On global scale, we can already witness how price

competition drives companies to explore new business opportunities. Brought to the Finnish

context, this further underlines the need to have new firms even with interests overlapping

with those of the incumbent organizations.

Our widespread research also resulted in more material based findings which also catalyzed

preliminary problem solving discussion among the research group. These more emergent issues

are listed below.

9

Empirical findings Comment / suggestion

Compared to the less efficient and less

focused firms of the 1970s, recent firms

are more efficient yet also narrowly

focused. This is a problem when the

market dynamics change.

A simple evolutionary logic suggests that more

variation increases probabilities of successful

selection (i.e. emergence of new successful products

and services). This would mean allowance of slightly

less focused innovation processes. Or the firms with

long history could simply conduct archeological

studies of their past and once abandoned business

ideas and technologies.

Overall depression regarding the

future of the forest cluster.

The paper market growth gradually slows down, and

may even decline. However, there are no signs that

even this traditional market would disappear – not

even mentioning the possible new business ideas.

Forest cluster is still a global player that offers career

and investor opportunities even after this particular

cycle turns upward. Our suggestion is that the forest

cluster should take an active role in managing the

image and public perception of the field. This is not

only a case of promotional activities yet requires more

fine-grained mechanisms acknowledged in the

institutional entrepreneurship literature.

Also, growth may be seen as an obsession rather than

a fact. It may be that some (if not all) business

activities are not eternally viable or may lose their

viability over time. A truly radical business innovation

would be to identify mechanisms that allow ‘managed

decline’ instead of violent collapse of closing of

activities.

10

Overall confusion about the possible

new product areas regarding especially

the bulk (e.g. bio-fuels) versus niche

product continuum.

Most probably, there is a market for different wood-

based business activities. However, especially the

more niche market businesses most probably require

a massive number of new firms. Like one the

informants stated: if new businesses are seen as

attractive, firms and actors need to make crisp

decisions, and activate market and research functions

accordingly – walk the walk.

One of the advantages of the forest

industry firms is the capability to

manage fully integrated mills.

Especially, when the energy question

comes more important this knowledge

may prove to be unique even in global

scale.

All existing capabilities, even if seen as supporting

activities, should be used in new business creation.

Structural changes in the raw material

market have changed the logic of how

wood material is bought.

Contrary to seeing this change (ownership transferred

from rural areas to urban households) as a hurdle, the

cluster should analyze and deeply the logics of how

these new types of actors want to deal with their

forest areas. Customer is always right – even in

resource markets.

New technologies are increasingly

invented and commercialized outside

Finnish borders. A crucial question is

if Finnish actors should focus on some

specific technologies instead of

offering the ‘whole package’.

Even conceptually, evolution of technology and

business is almost impossible (at least very difficult to

foresee). Thus, allowing variation in research and

development is vital for the forest cluster

development vis-à-vis planned order.

11

The logic how new machinery and

other equipment is acquired drives the

whole industry towards isomorphism.

That is, the use of the dominant

consulting houses, and machine

producers makes radical development

virtually impossible. More specifically,

the way in which consultants are hired

by customers to turn competing offers

technologically similar in order to

make prices comparable lowers the

incentives for radical innovation.

If this mechanism, that clearly drives towards mimic

outcomes, is recognized it would also allow the

invention of business practices. Referring to our

overall research findings and suggestion, this is a

specific case in which more competition (i.e. new

firms or existing firms offering new services) would

benefit the entire cluster.

Most of the individuals working in the

cluster firms share a very similar

educational and professional

background. What is more, this

homogeneity is strengthened by

following the traditional ways to

recruit new people from the same

institutions that have fed the industry

for decades.

A widespread consensus among our informants was

that the cluster needs a radical shift towards

heterogeneity in the educational and professional

background of key employees. This is also the

mechanism that may start the process of new

capability emergence.

12

CHAPTER 2

INTRODUCTION

MIRVA PELTONIEMI


ARNE KÖHLER


1 BACKGROUND AND OBJECTIVES

The main objective of the research reported in the current publication is to assess the

capabilities of the shareholders of ForestCluster Ltd (1) in relation to international competitors

and (2) as enablers of new business models and technological innovations. Here capabilities are

understood as a collection of routines and knowhow that combined enable firms to conduct

their business and to learn new capabilities as environmental change demands.

Capabilities are situated between intension and action, and they enable the firm to produce

intended outcomes (Dosi et al., 2000). Capabilities are higher-level routines that enable the

firm to produce significant outputs of a particular type (Winter, 2003). Whether an

organization has a certain capability is often a matter of degree (Winter, 2000). In literature

capabilities are often divided into technological, organizational and market capabilities, and

most studies aim at finding out the amount of a specific type of capability a particular firm

possesses (see Section 4 in the present publication).

13

Capability is an organization-level concept. Capabilities are built from lower-level routines,

individual skills, technological equipment and other assets that the firm possesses (Dosi et al.,

2000). Dynamic capabilities allow the firm to change in order to address rapidly changing

environments (Teece et al., 1997). This takes place through reconfiguring their resources to

respond to changing markets (Galunic and Eisenhardt, 2001, Brown and Eisenhardt, 1997).

Organizational architecture defines how capabilities and other resources are organized in the

firm (Jacobides, 2006, Galunic and Eisenhardt, 2001). Thus existing capabilities may be

reorganized to achieve a desired change in the firm’s operations. Dynamic capabilities are

learned and stable patterns of collective activity, that are used to continuously improve the

effectiveness of the organization (Zollo and Winter, 2002).

Here capabilities are approached from the viewpoint of cluster evolution. This choice has been

made because (1) the forest industries form a strong cluster in the Finnish economy and (2)

they have entered the mature phase of industry evolution. The challenge is to find plausible

routes to cluster renewal. Clusters tend to follow a predictable path or emergence, growth,

sustainment and decline (Menzel and Fornahl, 2010). The key variables that change during

such a process are technological heterogeneity of firms and the size of the cluster measured as

the number of employees (see Figure 1). As a new cluster emerges, the heterogeneity of

accessible knowledge increases rapidly. Thereafter the incentives to increase efficiency drive

technological convergence. As the cluster faces the threat of decline, the firms may respond

through accessing and developing new knowledge that increases the heterogeneity of the

cluster. In this way new links may be built among cluster members, but also with outsiders, and

new businesses may be created.

Figure 1 - Cluster life cycle (Menzel and Fornahl, 2010).

14

In the research reported here, the main puzzle is to understand how and why capabilities are

related to cluster renewal. Capabilities emerge slowly as a product of history. This means that

competitive advantage must be built on the existing set of capabilities as the development of

new capabilities takes time. Moreover, not having a clear picture of the capabilities that exist is

an important hurdle in finding new paths for capability development. So far capabilities

research has concentrated on the firm level. For example, of the five most highly cited

capabilities studies none go beyond the firm level (Teece et al., 1997, Kogut and Zander, 1992,

Leonard-Barton, 1992, Grant, 1996, Eisenhardt and Martin, 2000). Thus there is both an

academic and a practical need to look at capabilities at the cluster level.

The above reasoning leads us to pose four research questions:

5. What capabilities does the Forest Cluster possess?

6. What are the strengths and weaknesses compared to international competitors?

7. How and why capabilities interact with cluster evolution?

8. How capabilities enable (and hinder) innovation and renewal?

To answer the questions above, we first conduct an international comparison of the capability

accumulation of forest industry firms. This forms a diagnosis that will guide further empirical

work reported in the remainder of the report.

2 DIAGNOSIS – INTERNATIONAL COMPARISON OF CAPABILITY ACCUMULATION IN FOREST INDUSTRIES

Taking a quantitative perspective, capabilities may be best observed in the activities of

organizations. Actions are manifestations of internal organizational capabilities in the external

product-market environment. As Kiesler and Sproull (1982) first contended the capability to

take action always precedes action. Moreover, managers, and hence firms, are more likely to

engage in activities in which they possess some comparative advantage to competitors

(Levinthal and March, 1993, March, 1991). On the other hand, as organizations take action, they

garner experience – constantly moving down a learning curve – and process knowledge that

over time becomes solidified in routines (Nelson and Winter, 1982). The sum of routines, both

tacit and explicit, in turn constitutes what is generally known as organizational capabilities

(Teece et al., 1997). Organizations sequence of prior moves hence reveals organizations

15

competitive strengths and indicate capabilities possessed and gained over the course of the

organizations observed history.

Action data has been sourced using quantitative content analysis methods (Krippendorff, 1980)

of newsfeed data. The newsfeeds relating to the 200 largest paper & pulp industry firms during

1989-2009 was collected. This was done with an automated categorizing process based on

finding key terms in each newsfeed. The resulting codification was of astoundingly high

validity. However, we recognize that the cluster includes firms from a broader range of

industries. Still, the paper and pulp companies remain at the heart of the industry and though

the sample cannot give absolute values for any action it succeeds in presenting a comparative

view – a sufficient method for finding strengths and weaknesses. For a more detailed

description of the content analysis method refer to the methodology section in chapter 7.

Looking at the prior actions of the Finnish forest cluster (here analyzed as a part of a broader

Nordic cluster1) it portrays a vibrant and highly experienced image of the Nordic companies.

This bird’s-eye view suggests that the Finnish forest cluster owns competitive capabilities in

comparison to other regional groupings. Nonetheless, the overview of industry activity fails to

paint the whole picture; in the light of the findings, how and why do certain patterns of

activities emerge? Four interesting aspects rise from analyzing the history of actions taken

across regions in the forest industry during the past two decades. These relate to (1) the

intensity of capability accumulation, (2) innovation capabilities, (3) collaboration capabilities

and (4) restructuring capabilities.

2.1 Intensity of capability accumulation

First, analyzing the development actions taken by regional clusters reveals that the Nordic

firms have garnered more experience than firms in other regions. Figure 2 presents the amount

of actions firms have taken in aggregate from 1989 to 2009 focusing on developing the

efficiency of current operations, whereas Figure 3 includes actions that aim at creating value

through innovation. Furthermore, the sum of both exploitation and exploration activity of

Nordic firms has been rising between each five-year period except in 2004-2009. In contrast,

1 The Finnish and Swedish forest clusters are highly interlinked and over the period of study a large part of the firms have merged or acquired across the border of Finland and Sweden. Other Nordic companies are scarce in the sample (n=1). Note, that the sample is constructed in an illustrative manner and contains paper & pulp firms as well as machinery producers, but no other parts of the cluster (e.g. chemicals). Hence, findings are not absolute but provide a basis for comparison.

16

the reverse can be seen in North America and Asia. Based on these figures the Finnish forest

cluster should have superior skills in production and be relatively innovative. This should in

turn be seen in production efficiency, the number of new products launched and profitability

levels across the firms included in the cluster.

Figure 2 – Number of actions to improve current processes by exploiting existing capabilities 1989-2009 (e.g. new equipment, new plant, and process innovation).

Figure 3 – Number of actions taken to explore new possibilities 1989-2009 (e.g. reported innovations and launched R&D projects).

0

100

200

300

400

500

600

700

800

Africa America Asia Australasia Europe Scandinavia South America

Sum

of

acti

on

s

1989-1994

1994-1999

1999-2004

2004-2009

0

50

100

150

200

250

300

350

400


Sum

of

acti

on

s 1989-1994

1994-1999

1999-2004

2004-2009

Nordic North America


17

However, the large number of actions to improve efficiency and innovativeness by North

American firms in the beginning of the 1990’s did not save the North American forest industry

from becoming both inefficient and unsuccessful by the late 1990’s (see Ghosal and Nair-

Reichert, 2009). The findings indicate that high level of activity in exploitation or exploration

does not guarantee competitive advantage. This gives rise to two questions:

(I) Why are current capabilities in existing production processes or a high

innovation rate not enough?

(II) Is the Finnish cluster going to follow the declining path of North American

firms?

2.2 Innovation capabilities

Based on the absolute values in Figures 2 and 3, it is evident that the shares of exploitation and

exploration activities vary regionally. In the larger paper regions (i.e. North America, Nordic

countries and the rest of Europe), actions to improve efficiency by far outweigh innovative

exploration efforts. Further dissecting the efforts taken by firms to become competitive in

current domains (see Figure 4), it is apparent that most actions pertain to installing new

(enhanced) equipment, upgrading existing equipment and, to a lesser degree, to expanding to

new facilities. In comparison, process innovations have received far less attention. Hence,

investment in new equipment is the most popular development action. However, a small

number of distributors stand for the global supply of machinery and technology in the entire

industry. Any advantage gained in efficiency by installing machinery with new technology is

thus available with similar cost to competitors.

18

Figure 4 - Share of exploitation development actions by type. Allocation of all exploitation actions during 1989-2009.

Imitable incremental improvements cannot be the source of long term advantage as competitor

response will nullify such aspiration for competitive advantage (D'aveni, 1994, Kim and

Mauborgne, 2005). As stated above, the Finnish forest cluster has recently been one of the most

active regions to explore new product-market possibilities. Still, the industry offering remains a

commodity with little premiums earned and few new niches of growth have been developed

(Diesen, 2007). Not just innovation efforts, but radical innovations are needed to break vicious

cycles of increasing competition and decreasing demand (Schumpeter, 1934, Porter, 1980,

McGahan and Silverman, 2001). Thus we pose the following questions:

(III) Why has then the material part of development actions in the industry

focused on production efficiency through the installation of new equipment?

(IV) Why is radical innovation so difficult in the forest industry even though

efforts to innovate are numerous?

2.3 Collaboration capabilities

Activities taken in collaboration with fellow firms (e.g. joint ventures and partnerships) hint to

acquired skills in sourcing and sharing knowledge. The ability to discover, extend and utilize

knowledge beyond the single firm is the key strength in clusters (Menzel and Fornahl, 2010). As

seen in Figure 5, the firms in the Nordic cluster have on average been active collaborators

0 %

10 %

20 %

30 %

40 %

50 %

60 %

70 %

80 %

90 %

100 %


PROCESS INNOVATION/SYSTEM SOLUTIONS

UPGRADING

EXPANSION

NEW EQUIPMENT


19

through forming numerous joint ventures and through non-asset sharing collaborations and

partnerships.

Figure 5 – Initiated co-operation actions, categorized as asset sharing in joint ventures and non-asset sharing more loosely coupled collaborations and partnerships, made by regions firms 1994-2009. The plot shows the aggregate of actions weighed by the amount of companies in the cluster in five year periods.

The intensive co-operation enables Finnish firms to effectively share knowledge, and

experience in partnerships hint to the existence of established routines for cooperating flexibly.

Benefits from the interconnectedness of the firms could be manifold. Still, there is evidence

suggesting that strong network connections make cluster firms inert to change (Grabher, 1993)

and that too little heterogeneity hinders the emergence of new ideas (Jacobs, 1969). The above

reasoning leads us to ask:

(V) What factors enable or limit the benefits of collaboration in Finnish forest

cluster?

2.4 Restructuring capabilities

The global distribution of paper and pulp consumption is shifting from Western industrialized

continents towards growing third world nations, especially booming Asian economies.

Moreover, demand fluctuates following major economic trends. These circumstances have

increased the importance of the ability to flexibly restructure around capabilities (Lamberg,

0

1

2

3

4

5

6

7

8

9

0 1 2 3 4 5 6

Init

iate

d c

olla

bo

rati

on

s an

d

par

tne

rsh

ips

Initiated joint ventures

Africa

America

Asia

Australasia

Europe

ScandinaviaNordic

North America

20

2005, Diesen, 2007). The changing environment requires the firms to change, but also to

actively reshape the industry landscape through consolidating or restructuring internal

capabilities to serve market demand effectively. Figure 6 shows the aggregate of

transformational actions taken from 1989 to 2009 by region.

Figure 6 – Aggregate regional (A) acquisition actions and (B) restructuring actions.

Both merger and acquisition activity as well as restructuring activity have been intense in the

Nordic cluster compared to rival clusters. This indicates that the Nordic firms have perceived

such actions as either beneficial or absolutely necessary, and also that they have gained

experience of conducting and benefiting such actions. Nonetheless, it is unclear whether such

experience can be turned into competitive advantage. This gives rise to the following question:

(VI) In what ways has the Finnish forest cluster gained or failed to realize gains from

transformative action?

0

100

200

300

400

500

600

700

800

900

M&A Acquisitions

M&A Mergers

0

50

100

150

200

250

300

350

400

450

Restructuring Permanent

Restructuring Partial/Temporary

A. B.

21

3 RESEARCH PROCESS

The project begun with the four (1-4) research questions posed in Section 1. On the other hand,

the findings from the international comparison presented in Section 2 give rise to several

interesting questions (I-VI). The logic of the study is that the findings of the international

comparison guide the search for answers for the four research questions.

Figure 7 – The logic of the study.

To tackle the research task, the project was divided into seven sub-projects (Figure 8). First of

all, to ground the study on capabilities literature (A) the evolution of capabilities research was

studied with a bibliometrical method and (B) the methods used to operationalize capabilities

were studied through a review of 174 empirical capabilities studies published in top journals.

The sub-projects (A) and (B) forms Chapters 3 and 4.

Second, to answer the research question (1) “What capabilities does the Forest Cluster

possess?”, two empirical studies concerning the Finnish forest cluster were conducted.

(C) Interviews with 30 experts serve a source of qualitative understanding of the issue.

(D) Annual reports were used for constructing self-organizing maps to gain a quantitative view.

The sub-projects (C) and (D) form Chapters 3 and 4.

Third, to answer the question (2) “What are the strengths and weaknesses compared to

international competitors?”, international newsfeed data was used. This allows the comparison

of different geographical areas. Findings from sub-project (E) are presented in the previous

section.

22

Figue 8 – Research tasks and methodological approaches.

Fourth, to answer the questions (3) “How and why capabilities interact with cluster evolution?”

and (4) “How capabilities enable (and hinder) innovation and renewal?”, to sub-projects were

undertaken. Sub-project (F) analyses the relationship of competitive intensity and firms’

tendency to undertake exploitative and explorative actions. This Sub-project (G) looks at

capabilities in declining industries through a case survey. This allows us to assess strategies that

are successful in decline conditions and to identify interesting cases. These sub-projects are

reported in Chapters 5 and 6.

Findings, answers to the research questions and conclusions are summarized in Chapter 1.

23

REFERENCES

BROWN, S. L. & EISENHARDT, K. M. 1997. The art of continuous change: Linking complexity

theory and time-paced evolution in relentlessly shifting organizations. Administrative

Science Quarterly, 42, 1-34.

D'AVENI, R. 1994. Hypercompetition: Managing the dynamics of strategic maneuvering, New

York, NY, Free Press.

DIESEN, M. 2007. Economics of pulp and paper industry. Paper making science and

technology. Jyväskylä, FI: Gummerus.

DOSI, G., NELSON, R. & WINTER, S. 2000. The nature and dynamics of organizational

capabilities, Oxford University Press.

EISENHARDT, K. M. & MARTIN, J. A. 2000. Dynamic capabilities: What are they? Strategic

Management Journal, 21, 1105-1121.

GALUNIC, D. C. & EISENHARDT, K. M. 2001. Architectural innovation and modular corporate

forms. Academy of Management Journal, 44, 1229-1249.

GHOSAL, V. & NAIR-REICHERT, U. 2009. Investments in modernization, innovation and gains

in productivity: Evidence from firms in the global paper industry. Research Policy, 38,

536-547.

GRABHER, G. 1993. The weakness of strong ties. the lock-in of regional development in the Ruhr

area, London, Routledge.

GRANT, R. M. 1996. Prospering in dynamically-competitive environments: Organizational

capability as knowledge integration. Organization Science, 7, 375-387.

JACOBIDES, M. G. 2006. The architecture and design of organizational capabilities. Industrial

and Corporate Change, 15, 151-171.

JACOBS, J. 1969. The Economy of Cities, New York, NY, Vintage books.

KIESLER, S. & SPROULL, L. 1982. Managerial response to changing environments - Pespectives

on problem sensing from social cognition. Administrative Science Quarterly, 27, 548-

570.

KIM, W. & MAUBORGNE, R. 2005. Blue Ocean Strategy, Boston, MA, Harvard Business School

Press.

KOGUT, B. & ZANDER, U. 1992. Knowledge of the firm, combinative capabilities, and the

replication of technology. Organization Science, 3, 383-397.

KRIPPENDORFF, K. 1980. Content Analysis. An Introduction to its Methodology, London, UK,

The Sage Commtext Series, Sage Publications Ltd.

24

LAMBERG, J. A. 2005. Strategic actions and path dependence: Profiles and archetypes of

competitive behavior in a global industry, Helsinki, FI, Helsinki University of

Technology.

LEONARD-BARTON, D. 1992. Core capabilities and core rigidities - A paradox in managing

new product development. Strategic Management Journal, 13, 111-125.

LEVINTHAL, D. A. & MARCH, J. G. 1993. The myopia of learning. Strategic Management

Journal, 14, 95-112.

MARCH, J. G. 1991. Exploration and exploitation in organizational learning. Organization

Science, 2, 71-87.

MCGAHAN, A. M. & SILVERMAN, B. S. 2001. How does innovative activity change as industries

mature? International Journal of Industrial Organization, 19, 1141-1160.

MENZEL, M. P. & FORNAHL, D. 2010. Cluster life cycles-dimensions and rationales of cluster

evolution. Industrial and Corporate Change, 19, 205-238.

NELSON, R. R. & WINTER, S. G. 1982. An evolutionary theory of economic change, Cambridge,

Belknap Press of of Harvard University Press.

PORTER, M. E. 1980. Competitive Strategy: Techniques for analyzing Industries and

Competitors, New York, NY, Free Press.

SCHUMPETER, J. A. 1934. The theory of economic development : an inquiry into profits, capital,

credit, interest, and the business cycle, Cambridge, Harvard University Press.

TEECE, D. J., PISANO, G. & SHUEN, A. 1997. Dynamic capabilities and strategic management.

Strategic Management Journal, 18, 509-533.

WINTER, S. G. 2000. The satisficing principle in capability learning. Strategic Management

Journal, 21, 981-996.

WINTER, S. G. 2003. Understanding dynamic capabilities. Strategic Management Journal, 24,

991-995.

ZOLLO, M. & WINTER, S. G. 2002. Deliberate learning and the evolution of dynamic

capabilities. Organization Science, 13, 339-351.

25

CHAPTER 3

THE STRUCTURE AND EVOLUTION OF CAPABILITIES RESEARCH IN 1986–2009:

A BIBLIOMETRIC STUDY

OLA LAAKSONEN


1 INTRODUCTION

The term ‘capabilities’ floats in the literature like an iceberg in a foggy Arctic sea, one iceberg among many, not easily recognized as different from several icebergs near by (Dosi, Nelson & Winter 2000, p. 3).

Since the ideas proposed by, for example, Penrose (1959), Wernerfelt (1984), and Barney (1991),

and the subsequent emergence of the ‘resource-based view of the firm’, the concept of

capabilities has become a focal area in business strategy research. In contrast to the industrial-

organization tradition that emphasizes the analysis of external forces, the resource-based view

(RBV) has an internal focus (Armstrong & Shimizu 2007) since it considers a firm’s resources

the key to its competitiveness. In order to gain advantage related to its competitors, a firm has

to build resources that are valuable, rare, and hard to imitate and substitute (Barney 1991) or an

‘asset stock’ that is hard to replicate (Dierickx & Cool 1989).2

2 There is no consensus on whether to use the term ‘resource-based theory’ or ‘resource-based view’ when referring

to this stream of literature (Acedo, Barroso & Galan 2006). For a debate on whether the resource-based view is a theory, see Priem and Butler (2001a, 2001b) and Barney (2001). The term ‘resource-based view’ is used consistently throughout this study.

26

Capabilities are one building block of this asset stock. According to Barney (1991; building on

Daft 1983), a firm’s resources include “all assets, capabilities, organizational processes, firm

attributes, information etc. controlled by a firm that enable the firm to conceive of and

implement strategies that improve its efficiency and effectiveness.” (p. 101.) In other words,

capabilities are a central part of the resource base of an organization and thus, according to the

resource-based view, an important factor in defining a firm’s competitiveness.

There are quite many literature reviews on the resource-based view and related concepts3.

Usually, however, the concept of capabilities is not explicitly discussed in these reviews and the

understanding of capabilities remains fragmented4. This is why it is justified to conduct a

structured review on capabilities research in particular. Firstly, this approach has the potential

of revealing the most important works within capabilities research. Secondly, it can shed light

on the use of the concept of capabilities and its relatedness to other concepts such as

organizational routines (Nelson & Winter 1982), dynamic capabilities (Teece, Pisano & Shuen

1997), combinative capabilities (Kogut & Zander 1992), and absorptive capacity (Cohen &

Levinthal 1990) that are also important to the resource-based view.

Hence, this study is a bibliometric review of the structure and evolution of academic

capabilities research between 1986 and 2009. A bibliometric analysis is a structured and a

relatively objective way to study a field of research (Nerur, Rasheed & Natarajan 2008). The

citation data of more than 2700 articles from 39 academic journals is analyzed with citation

analysis, co-citation analysis, and keyword analysis techniques. This is also a longitudinal study:

the time period is divided into four sub-periods in order to shed light on the evolution of

capabilities research.

To conclude, this study seeks answers to the following research questions:

i) What are the most important academic publications in capabilities research?

ii) How has the focus of capabilities research evolved between 1986 and 2009?

The first research question is answered by conducting a citation analysis. The second research

question is answered by co-citation and keyword analysis techniques.

3 See, for example, Becker (2004) on organizational routines, Wang & Ahmed (2007), Arendt & Bromiley (2009) or

Ambrosini & Bowman (2009) on dynamic capabilities, and Acedo, Barroso & Galan (2006) or Newbert (2007) on the resource-based view. 4

Peng, Schroeder, and Shah (2008) make an exception since they review the concept of capabilities separately, too, but from an operations management rather than a strategic management perspective.

27

The term ’capabilities’ is not defined here since this study adopts a wide view of this construct.

Capabilities research is simply considered to consist of research in business and management

that deals with the concept of capabilities – regardless of how the authors define this concept.

Hence, the aim of this study is to shed light on the use of the concept of capabilities within the

fields of management and business rather than to delve into one specific definition.

2 METHOD AND DATA

This section first introduces the umbrella term ‘bibliometrics’ and then explains the data

collection process. This is followed by the description of the three specific techniques – citation

analysis, co-citation analysis, and keyword analysis – employed in conducting the bibliometric

analyses.

2.1 Bibliometrics

Pritchard (1969) offered one of the first definitions of the term ‘bibliometrics’ (Broadus 1987).

According to him, bibliometrics is “the application of mathematics and statistical methods to

books and other media of communication” (p. 349) in order to “shed light on the processes of

written communication” (p. 348). Having criticized Pritchard’s definition as too vague and after

reviewing various prior definitions, Broadus (1987) concludes that “bibliometrics is the

quantitative study of physical published units, or of bibliographic units, or of the surrogates for

either” (p. 376). These bibliographic units are, for example, citations or keywords in academic

articles. Generally speaking, the term bibliometrics thus covers a set of methods for analyzing

the structure of scholarly communication, its task being to “provide evolutionary models of

science, technology, and scholarship” (White & McCain 1989, p. 119).

The bibliometric methods employed in this study are citation analysis, co-citation analysis, and

keyword analysis. Similar techniques have been used, for example, in the fields of management

information systems (Culnan & Swanson 1986), strategic management (Martinsons, Everett &

Chan 2001, Ramos-Rodríguez & Ruíz-Navarro 2004, Nerur, Rasheed & Natarajan 2008),

knowledge management (Ponzi 2002), marketing (Hoffman & Holbrook 1993, Baumgartner &

Pieters 2003), operations management (Pilkington & Liston-Heyes 1999, Pilkington & Meredith

2009), entrepreneurship (Schildt, Zahra & Sillanpää 2006), and economics (Cahlik 2000).

The most interesting bibliometric review from the viewpoint of this study is the one of Acedo,

Barroso, and Galan (2006) on the resource-based view. Even though their study makes

28

interesting remarks on the concept of capabilities, it does not discuss this concept separately.

Thus, it seems that the structure and evolution of capabilities research in particular has not

previously been analyzed with bibliometric techniques.

2.2 Data

A citation is the basic unit of analysis in bibliometric research. In this study, this citation data

was retrieved from articles in academic journals. There are two reasons for this. Firstly, articles

published in journals can be considered ‘certified knowledge’ since they have been reviewed

and accepted by other researchers (Ramos-Rodríquez & Ruíz-Navarro 2004). Secondly, the

citation data of these journals is easily accessed through ISI Web of Science

(www.isiknowledge.com). The citation data used in this study was retrieved from the Social

Sciences Citation Index (SSCI) in the ISI Web of Science. This database indexes over 2000

journals from various social science disciplines.

The first step in gathering bibliometric data was the selection of journals from which the

citation data were retrieved. This is an important step in the data gathering process since the

selection of journals affects both the number and quality of the articles included in the analysis.

The aim in this study was to choose journals that are of high quality and that are associated

with capabilities research from the viewpoint of management and business.

There are three approaches to choosing the journals. The first approach is to rely on the

researcher’s or others experts’ judgement in choosing the most prominent journals (e.g.,

Pilkington & Meredith 2009). This approach is problematic since it violates the objectivity

principle of a bibliometric study (Acedo, Barroso & Galan 2006). The second alternative is to

choose the journals based on their impact factors (e.g., Järvinen & Sillanpää 2007) or the

number of citations they have received (e.g., Cahlik 2000). The impact factor is based on the

number of citations that articles in a journal have received and it can be used in evaluating the

journal’s quality in comparison to other journals in the same field (ISI Journal Citation Report,

Social Science 2008 Edition). The third method is to choose journals that are publishing most

actively in the specific field of research under consideration (e.g., Schildt, Zahra & Sillanpää

2006) – in this case, in capabilities research.

In this study, a combination of the two latter methods was used. Firstly, based on impact

factors, the top 40 journals from the areas of business and management were identified from

the ISI Journal Citation Report. Secondly, the 40 journals that included the largest number of

29

capabilities-related articles were identified (using the search word capabilit* in the SSCI). The

first list has a relative emphasis on strategic and general management journals, while the

second list includes more journals from the fields of operations and technology management.

Finally, the top 25 journals from each of these two rankings were combined into a single list,

which, after deleting the overlaps, comprised 39 journals. This list is shown in Appendix 1.

Next, a search from these 39 journals in the SSCI was conducted through ISI Web of Science.

The records in the SSCI were available from 1 January, 1986, and thus the investigated time

period was set to 1986–June 2009. In order to achieve a longitudinal view on capabilities

research, the time period was divided into four six-year sub-periods: 1986–1991, 1992–1997,

1998–2003, and 2004–2009. The search word used was capabilit*, but the search word capability

OR capabilities would have yielded the same number of results. The searches yielded a total of

2,753 articles with words beginning with “capabilit” in their titles, abstracts, or keywords. These

articles are called citing articles. These citing articles cited a total of 79,406 articles that are

called cited articles. Table 1 shows the number of articles retrieved for each period.

Table 1: Number of citing and cited articles for each sub-period

The citation data of these articles were then imported into four Microsoft Access databases

(one for each sub-period) using the bibliometric software tool Sitkis (Schildt 2005). This

database contains, for example, information about the number of citations a cited article, an

author, or a journal has received, or about the frequency with which a keyword has appeared in

the citing articles. The data in these databases act as the basis for further bibliometric analyses.

As this data often contains some misspellings of authors’ or journals’ names, volume numbers,

or years of publication, several corrections were made. According to Schildt (2004), it is

sufficient to identify 20–50 top authors and make sure that their entries are correct. In this

study, the entries of the 55 most cited authors in each sub-period were checked and corrected.

Following Järvinen and Sillanpää (2007), different editions of the same book were also recorded

as the original edition, even though there was a possibility that the book’s content had changed

between the editions.

Years # citing articles # cited articles

1986–1991 37 724

1992–1997 412 10 731

1998–2003 868 25 194

2004–2009 1 436 42 757

1986–2009 2 753 79 406

30

Also, the keywords that had appeared most frequently in the articles were identified and their

spelling was checked. In addition, different conjugations were unified. For example, the

keywords R-and-D and research-and-development were changed into R&D, and the keyword

Resource-based theory was changed into Resource-based view.

2.3 Citation Analysis

Citation analysis is the systematic analysis of citation patterns in a scientific discipline. Citation

analysis is based on the assumption that when authors cite a document in their work, that

document is somehow important for their research. That is why frequently cited articles can be

regarded as more important building blocks of a research stream than articles with fewer

citations. (Ramos-Rodríquez & Ruíz-Navarro 2004.)

In this study, a citation analysis was conducted in order to find the most cited documents

within capabilities research. In addition, the changes in the relative citation frequencies of the

most cited documents were investigated. The relative citation frequency of a document in a

sub-period is calculated by dividing the number of citations the document has received by the

total number of citing articles in the sub-period. Thus, the relative frequency is the percentage

of citing articles that have cited the document in question.

2.4 Co-citation Analysis and Cluster Formation

In his seminal article, Small (1973) defines co-citation as the frequency with which two

documents are cited jointly by a third document. The number of co-citations can be

interpreted as a measure for the similarity of content of the two documents (Ramos-Rodríquez

& Ruíz-Navarro 2004). Even though an author may cite one document for the purpose of

criticizing it and another document for giving credit to it, the popularity of the co-citation

method indicates that high co-citation strength is generally considered to indicate high

similarity of the two cited articles. Thus, co-citation analysis is a method for separating more

homogeneous research streams from a larger field of study (ibid.).

In this study, co-citation analysis is used to form separate clusters of highly cited articles in the

capabilities research stream. The aim is to separate groups of important articles with high

mutual co-citation strength from each other, in order to see in what kinds of different ways the

concept of capabilities is dealt with in the literature. Thus, this study has a ‘micro’ approach to

co-citation analysis since it aims to describe the structure and historical development of an

31

individual research stream; whereas a ‘macro’ approach would focus on the overall structure of

scientific disciplines (Gmür 2003).

A normalized value of co-citation strength was used in order to emphasize proximity between

similar references that are cited less often than the most common references (Schildt &

Mattsson 2006). The Jaccard coefficient of similarity (S) was used as the normalization method.

If a and b represent the number of citations to documents A and B and a b is the number of

common citations to both A and B, the Jaccard coefficient S can be expressed as:

.

The Jaccard coefficient takes values from the range 0–1, where 0 represents no co-citations

between the two articles and 1 indicates that the two articles are always cited together (Schildt,

Zahra & Sillanpää 2006). The Jaccard normalization was conducted for the most cited articles in

each sub-period by the bibliometric software tool Sitkis (Schildt 2005). The normalized co-

citations strengths were imported to UCINET software (Borgatti, Everett & Freeman 2002)

which, in turn, produced a matrix showing the normalized co-citation strength for each pair of

articles.

The grouping of the articles into more homogeneous clusters was performed using a dense sub-

network grouping algorithm that is implemented in the bibliometric software tool Sitkis

(Schildt & Mattsson 2006). The algorithm starts forming a group from the two articles that

have the highest normalized co-citation value. Next, it adds to the group the article which has

the strongest co-citation link weight to existing group members. This is repeated until the co-

citation link value is lower than a chosen cutoff value. The group is then separated from the

remaining network and the algorithm starts again from the most strongly connected two

articles. Consequently, the iteration yields a number of densely connected groups of articles.

(ibid.)

Following Schildt, Zahra, and Sillanpää (2006), several different cutoff values for the

normalized co-citation strength were tested. As there are no established criteria for choosing

the cutoff value (ibid.), a value that produced a reasonable number of clusters (5–7) was

chosen. Additionally, as Schildt, Zahra, and Sillanpää (2006) point out, the general results of

the analyses do not depend largely on the chosen cutoff value.

baba

baBAS ),(

32

After identifying the separate clusters of articles for each sub-period, a picture of the clusters

was drawn with Pajek software (Batagelj & Mrvar 2009). In these network pictures, the distance

and the width of the line between two nodes (articles) represent the normalized co-citation

strength between the two articles. The distances between the articles are determined by a non-

metric multidimensional scaling (MDS) algorithm (Scott 2000).

The number of the most frequently cited documents that were included in the analyses, the

minimum number of references a document must have had received in order to be included,

the cutoff values used in forming the clusters and the number of separate clusters retrieved for

each sub-period is shown in Table 2.

Table 2: Number of documents analyzed, minimum citation counts for these documents, cutoff values of normalized co-citation strengths, and the number of clusters in each sub-period

2.5 Keyword Analysis

The keyword analysis technique analyzes the frequency with which each keyword has appeared

in the citing articles during each sub-period. These keyword statistics, which can be derived

from the Access databases, help to map the most important themes and possible shifts of focus

in capabilities research. By analyzing the citing articles this technique thus complements the

citation and co-citation analyses that analyze the cited articles.

Years

# of documents

included

Min. # of

citations

Cutoff

value

# of

clusters

1986–1991 18 2 0,35 5

1992–1997 33 17 0,18 7

1998–2003 30 57 0,16 5

2004–2009 33 94 0,14 7

33

3 RESULTS

The following sections present the most interesting findings from the citation analyses, co-

citation analyses, and keyword analyses. The aim is to highlight the most important areas of

inquiry and shifts in focus of capabilities research.

3.1 Citation Analyses

Appendix 2 presents the 30 most cited documents in the overall period between 1986 and 2009

and their raw and relative citation frequencies in each sub-period. The documents are ranked

based on the number of citations they have received.

One of the most important remarks that can be made from these citation statistics is that very

few of the 30 most cited articles have been cited during the first sub-period 1986–1991. For

example, the book The Theory of the Growth of the Firm by Penrose (1959), which is nowadays

considered one of the most important building blocks of the resource-based view (Acedo,

Barroso & Galan 2006), was not cited by the citing articles during the first sub-period. This

suggests that capabilities research as a separate field of inquiry had not yet emerged during this

sub-period, even though many of the most important documents had been published in the

1980s or before.

In addition, the citation counts in the sub-period 1986–1991 are generally very low: the article

Visual Interactive Modelling in Operational Research: Successes and Opportunities by Bell (1985)

was the only document that gained more than two citations. This is why the first sub-period

has been excluded from the following co-citation and keyword analyses: the results would be

neither interesting nor reliable.

Nelson and Winter’s (1982) book An Evolutionary Theory of Economic Change seems to have

maintained the most stable popularity throughout the overall period. In addition to Barney’s

(1991) article on the resource-based view, it is the only document that has been in the top five

in each of the three latter sub-periods. Especially the chapters 3–5 in An Evolutionary Theory of

Economic Change5 are considered important for the capabilities discussion (Foss 2003).

According to Foss (2003), “re-reading the chapters makes one realize that perhaps not so much

5 These chapters are named The Foundations of Contemporary Orthodoxy (Ch. 3), Skills (Ch. 4), and Organizational

Capabilities and Behavior (Ch. 5).

34

essential has happened in two succeeding decades of work on capabilities, competence,

evolutionary, etc., theories of the firm that goes beyond Nelson and Winter’s treatment.” (p.

186.)

Even though Nelson and Winter’s (1982) book seems to have initially been the most influential

work on capabilities, the article Dynamic Capabilities and Strategic Management by Teece,

Pisano, and Shuen (1997) is the most cited document in the overall period of 1986–2009. It is

often cited as the seminal work on dynamic capabilities (Easterby-Smith, Lyles & Peteraf 2009),

even though already Williamson (1991) noted the emergence of this concept. The dynamic

capabilities perspective builds on the traditional resource-based view, but regards it as static

and considers that it has neglected the influence of market dynamism. Thus, the construct of

dynamic capabilities aims at explaining how certain firms succeed in gaining competitive

advantage in situations of rapid change (Eisenhardt & Martin 2000). Teece and colleagues

(1997) define dynamic capabilities as “the firm’s ability to integrate, build, and reconfigure

internal and external competences to address rapidly changing environments” (p. 516).

The third most cited document is Barney’s (1991) article Firm Resources and Sustained

Competitive Advantage. Barney builds his quest for sustained competitive advantage on the

assumption that resources are heterogeneously distributed across firms and that these resource

differences are somewhat stable. Barney (1991) views capabilities as one type of these firm

resources. According to him, these resources can be a source of sustained competitive

advantage only if they are valuable, rare, imperfectly imitable, and non-substitutable (this has

later been called the ‘VRIN’ framework).

The fourth most cited document is Cohen and Levinthal’s (1990) article Absorptive Capacity: A

New Perspective on Learning and Innovation. Absorptive capacity is defined as “the ability of a

firm to recognize the value of new, external information, assimilate it, and apply it to

commercial ends” (p. 128). Thus, absorptive capacity is a kind of a capability. Absorptive

capacity is especially important in product development, where both prior knowledge and the

ability to exploit external knowledge are essential. One important remark that Cohen and

Levinthal make is the path-dependence of absorptive capacity: prior knowledge affects the

firm’s ability to learn new things in the future. This is true with capabilities as well: early

investments in certain capabilities may to some extent limit the capability development

alternatives the firm has in the future.

The fifth most cited document is Kogut and Zander’s (1992) article Knowledge of the Firm,

Combinative Capabilities, and the Replication of Technology. The concept of combinative

35

capabilities introduced in the article is in many ways similar to Cohen and Levinthal’s (1990)

absorptive capacity. Combinative capability is defined as the ability to exploit and recombine

both internal and external knowledge and capabilities in order to learn new things. It, too, is a

path-dependent phenomenon since it resides in the firm’s organizing structure and social

relationships that are not easy to change. This is why learning by reconfiguring current

capabilities is often simpler than learning by creating totally new capabilities. Kogut and

Zander’s view that firms learn new skills by combining their current capabilities is thus

somewhat similar to the view Teece, Pisano, and Shuen (1997) would adopt five years later.

The diverse nature of capabilities research is quite clearly illustrated by the range of topics that

the most cited documents cover. Not surprisingly, the most cited documents include the

seminal works on capabilities and routines (Nelson & Winter 1982) and the resource-based

view (Barney 1991). The three other documents can be characterized as more recent

modifications to these initial ideas. Even though these modifications – dynamic capabilities,

absorptive capacity, and combinative capabilities – are named differently, they all share a

common theme: the reconfiguration of current capabilities.

The five most cited documents also illustrate the fragmented nature of capabilities research:

only the work of Nelson and Winter (1982) deals specifically with capabilities. For the other of

the five most cited works, the concept of capabilities is rather a building block for further

consideration than a separate field of study on its own.

In addition, the most cited works are not explicit in how the concept of capabilities is related to

the concepts they deal with, i.e. dynamic capabilities, absorptive capacity, resources,

knowledge, and combinative capabilities. For example, as Wang and Ahmed (2007) note,

Teece, Pisano and Shuen’s (1997) definition of capabilities is hardly different from their

definition of dynamic capabilities.6 Absorptive capacity, on the other hand, is defined as a type

of a capability (Cohen & Levinthal 1990) without addressing the question of what capabilities

are.

Of the 25 most cited documents, Figure 1 presents the four fastest growing and the four fastest

decreasing documents in relative frequency from the third to the fourth sub-period.

6 According to Teece, Pisano, and Shuen (1997), “The term ‘capabilities’ emphasizes the key role of strategic

management in appropriately adapting, integrating, and reconfiguring internal and external organizational skills, resources, and functional competences to match the requirements of a changing environment.” (p. 515.) They define dynamic capabilities as “the firm’s ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments” (p. 516).

36

Figure 1: The four fastest growing and the four fastest decreasing documents in relative frequency from the third to the fourth sub-period

Figure 1 shows that, in addition to being the most popular document, the article by Teece and

colleagues (1997) has gained increasing popularity at a significant pace. The other three

documents with the fastest growing relative frequency are the articles by Eisenhardt and

Martin (2000) on dynamic capabilities, March (1991) on organizational learning, and Grant

(1996b) on the ‘knowledge-based view’, which considers knowledge the most important

strategic resource for an organization. In contrast, the four documents with the fastest

decreasing relative frequency are Porter’s (1980) industrial-organization classic Competitive

Strategy, Barney’s (1986) article on strategic factor markets, Prahalad and Hamel’s (1990)

popularization of the term ‘core competence’ and Wernerfelt’s (1984) article on the resource-

based view.

These recent changes in the relative frequencies indicate a shift in focus from the traditional

resource-based view to the dynamic capabilities view, the knowledge-based view, and learning

approaches. These findings are somewhat similar to those of Acedo, Barroso, and Galan (2006),

who note a division between the classic works from the resource-based view and its more

recent applications. This shift in focus of capabilities research will be further discussed in the

following sections.

3.2 Co-citation Analyses

This section presents the groups of the most cited documents for each sub-period. The lists of

the articles included in each cluster and the cluster network pictures are presented in

0 %

5 %

10 %

15 %

20 %

25 %

30 %

35 %

Eisenhardt

& Martin

(2000)

Grant

(1996b)

Teece,

Pisano &

Shuen

(1997)

March

(1991)

Porter

(1980)

Barney

(1986)

Prahalad &

Hamel

(1990)

Wernerfelt

(1984)

1998–2003

2004–2009

37

Appendices 3–6. Following Gmür (2003), only the clusters that include at least three documents

that are all connected to each other by a node are discussed here. This leaves out all clusters

with less than three documents and cluster S in which the three works are not all linked to

each other. In addition, articles outside of the clusters (i.e. articles in cluster 0) are not

discussed here.

3.2.1 Clusters in 1986–1991

As noted before, the citation counts for the cited articles in this sub-period are very small.

Hence, the only reliable conclusion that can be made is that capabilities research was not yet

established during this sub-period, even though the theoretical foundations of the more recent

capabilities discussion were laid at that time. This is why the results of the co-citation analyses

for this sub-period are not further discussed here. However, the results of the co-citation

analyses for this sub-period are presented in Appendix 3. These results include a list and a

network picture of the five clusters (clusters A–E) that emerged at a normalized cut-off value of

0.35.

3.2.2 Clusters in 1992–1997

In this sub-period, seven clusters of articles emerged at a cut-off value of 0.18. These clusters

(clusters F–L) are listed and depicted as a network picture in Appendix 4. Clusters F, G, H, and

I include at least three documents and are discussed here.

Cluster F: The Resource-Based View. This cluster consists of works on the resource-based view

(Penrose 1959, Wernerfelt 1984, Barney 1991, Dierickx & Cool 1989) and on some of its

applications. These include, for example, Prahalad and Hamel’s (1990) popularization of the

term ‘core competence’, Leonard-Barton’s (1992) article on capabilities in product

development, and Amit and Schoemaker’s (1993) work on the role of capabilities and other

factors in generating economic rent. The works by Porter (1980, 1985) are not generally

considered a building block of the capabilities discussion since they focus on the impact of

external forces on competitiveness. However, Porter’s 1980 book Competitive Strategy is placed

in this cluster, which indicates its influence on the whole field of strategic management (Nerur,

Rasheed & Natarajan 2008).

Cluster G: Organization Theory and Organizational Learning. This cluster contains two works

on both organization theory and on organizational learning. James March being a co-author in

38

one work on both of these topics is probably one reason why these two research streams are

situated in the same cluster. The two organization theory classics in this cluster are March and

Simon’s (1958) book Organizations and Thompson’s (1967) book Organizations in Action. Both

of these works are rather all-inclusive studies on how organizations function. Issues covered in

these two books include, for example, control, incentives, structures, and rationality in

organizations.

The two other works in this cluster are Argyris and Schön’s (1978) book and Levitt and March’s

(1988) article that are both titled Organizational Learning. Argyris and Schön (1978)

concentrate more on individual learning within single organizations (Levitt & March 1988),

whereas Levitt and March (ibid.) examine the process of organizational learning. They view

learning as a routine-based, history-dependent, and target-oriented phenomenon.

Organizational learning thus has features quite similar to those later attributed to

organizational capabilities (e.g., Dosi, Nelson & Winter 2000).

Cluster H: Capabilities and Environmental Change. Studies in this cluster focus on economic

and technological change and their relatedness to organizational capabilities. In addition to

Nelson and Winter’s (1982) classic An Evolutionary Theory of Economic Change this cluster

consists of two works that concentrate on technological change (Tushman & Anderson 1986)

and on different kinds of innovations (Henderson & Clark 1990). Tushman and Anderson

(1986) view technological changes as either competence enhancing (that build on existing

capabilities) or competence destroying (that require new kinds of capabilities). Henderson and

Clark (1990), on the other hand, investigate the interplay between innovations and different

organizational capabilities.

Cluster I: Early Knowledge-Based Approaches. This rather fragmented cluster contains Lippman

and Rumelt’s (1982) article on efficiency differences between firms under competition and two

articles on the early knowledge-based view. These two works are Winter’s (1987) article

Knowledge and Competence as Strategic Assets and Kogut and Zander’s (1992) article that

introduces the concept of combinative capabilities.

3.2.3 Clusters in 1998–2003

In this sub-period, five clusters of articles emerged at a cut-off value of 0.16. The articles in

these clusters (clusters M–Q) are listed and depicted as a network picture in Appendix 5.

Clusters M, N, and O include at least three documents and are discussed here.

39

Cluster M: The Resource-Based View and Dynamic Capabilities. This cluster in very similar to

cluster F in 1992–1997: it contains the seminal works on the RBV and some of its applications.

In addition to the applications in cluster F, cluster M introduces a new one: the construct of

dynamic capabilities (Teece, Pisano & Shuen 1997).

Cluster N: Knowledge-Based Approaches. This cluster consists of works that focus on knowledge

as an important strategic resource for organizations. Except for Winter’s (1987) article (that

appeared already in cluster I), all of the works in this cluster have been published in the 1990s.

This is a sign of the relative newness of this “knowledge-based view” of the firm. Even though

all of the works in this cluster have this unifying theme, they also differ in how they approach

it. Zander and Kogut (1995) and Szulanski (1996), for example, examine the transfer of

capabilities (such as best practice) within organizations. Cohen and Levinthal’s (1990) article on

absorptive capacity has already been discussed in the previous section. Nonaka (1994) and

Nonaka and Takeuchi (1995), on the other hand, concentrate on the dynamic processes related

to the creation and transformation of tacit and explicit knowledge within organizations.

Cluster O: Organizational Learning. This cluster is very similar to cluster G in 1992–1997.

Thompson’s (1967) book has been replaced by March’s (1991) article on explorative and

exploitative actions in organizational learning. This framework is somewhat similar to the

division of capabilities into productive and dynamic types (see, for example, Jacobides 2006).

Applying this analogy, dynamic capabilities are directed towards exploring new resource

configurations (Teece, Pisano & Shuen 1997, Eisenhardt & Martin 2000) whereas productive

(e.g., Jacobides & Winter 2005) or ‘zero-level’ capabilities (Winter 2003) aim at making the

most out of the firm’s current resources.

3.2.4 Clusters in 2004–2009

In this sub-period, six clusters of articles emerged at a cut-off value of 0.13. The articles in these

clusters (clusters R–W) are listed and depicted as a network picture in Appendix 6. Clusters R,

T, V, and W include at least three documents and are discussed here.

Cluster R: The Resource-Based View and Dynamic Capabilities. This cluster is quite similar to

clusters F and M in the previous sub-periods. In addition to Teece, Pisano, and Shuen’s (1997)

article, also the two other works on dynamic capabilities (Eisenhardt & Martin 2000, Zollo &

Winter 2002) are placed in this cluster. It is interesting that the knowledge-based view and

learning approaches have formed separate clusters but the dynamic capabilities view is very

40

tightly connected to the representative works of the resource-based view. As Wang and Ahmed

(2007) and Acedo, Barroso and Galan (2006) note in their reviews, the construct of dynamic

capabilities is complementary to the resource-based view. The findings in this cluster analysis

indicate the same. It thus seems that Williamson (1991) has been on the right track when

pondering on where the concept of dynamic capabilities would be placed in regard to the

resource-based view:

The leading efficiency approaches to business strategy are the resource-based and the dynamic capabilities approach … It is not obvious to me how these two literatures will play out – either individually or in combination. Plainly, they deal with core issues. Possibly they will be joined. (p. 76.)

Cluster R also includes three works on the knowledge-based view (Cohen & Levinthal 1990,

Kogut & Zander 1992, Grant 1996b). Most probably, this results from the choice of the cut-off

value for the normalized co-citation strength and does not indicate any significant change in

the structure of capabilities research.

Cluster T: Organizational Learning. This cluster is very similar to clusters G and O in the

previous sub-periods. The works on organization theory have dropped out and this cluster

consists of works on organizational learning. In addition to Levitt and March’s (1988) and

March’s (1991) articles that had appeared in the previous clusters, too, this cluster includes

Levinthal and March’s (1993) article The Myopia of Learning. This article is a further

investigation of exploration and exploitation in organizational learning and “the learning

capabilities of organizations” (ibid.).

Cluster V: Transaction Cost Economics. The most important works in this cluster are

Williamson’s (1975, 1985) books on transaction cost economics. Transaction cost economics

offers a “governance perspective” to business strategy as opposed to the resource-based views

that Williamson (1999) calls the “competence perspective”. At first sight, Williamson’s ideas

seem to be cited as an antithesis to the capabilities view. For example, Williamson (1975, 1985)

views the firm as a governance structure whereas the competence perspective considers the

firm a bundle of related resources (Williamson 1999). Researchers within the competence

perspective, on the other hand, have criticized transaction cost economics as static (ibid.).

More recently, however, there has been interest in how capabilities and transaction costs

overlap and complement each other (ibid.) and co-evolve (Jacobides & Winter 2005). Teece,

Pisano and Shuen (1997), too, build their theory of dynamic capabilities partly on Williamson’s

works.

41

Cluster W: Knowledge-Based Approaches. This cluster in very similar to cluster N in 1998–2003.

A new work in this cluster is the article by Grant (1996a), which presents a very clear view of

the connectedness of knowledge and capabilities. According to Grant (ibid.), “… the essence of

organizational capability is the integration of individuals’ specialized knowledge.” (p. 375.)

3.2.5 Changes in the Clusters’ Relative Frequencies

Table 3 presents the relative frequency of each cluster discussed above. It is calculated by

dividing the total number of citations that documents in each cluster have received by the

number of citing articles. This relative frequency is the average number of citations that each

citing article has made to cited articles in a cluster. A relative frequency of 1.5 of a cluster would

thus indicate that each citing article has, on average, cited 1.5 articles of that cluster. It must be

noted that because of the different number of articles in each cluster the relative frequencies of

clusters within different sub-periods are not perfectly comparable. However, this analysis

reveals some interesting findings about the evolution of capabilities research.

Table 3: The relative frequencies of the most important clusters

The most important observation that can be made from table 3 is that the most central cluster

(clusters F, M, and R) gained significant popularity at the end of the 1990s. In 1992–1997, cluster

F gained an average of 0.74 citations from each citing article whereas in 1998–2003 each citing

article cited an average of 2.03 articles in the M-cluster. This observation can also be made by

having a look at the network pictures presented in the Appendices. In 1992–1997 the articles in

cluster F are still relatively far apart from each other, whereas in 2003–2009 the central articles

in cluster R are very close to each other.

Sub-period Cluster Rel. freq.

F 0.74

G 0.20

H 0.26

I 0.14

M 2.03

N 0.61

O 0.22

R 1.96

T 0.26

V 0.23

W 0.33

1992–1997

1998–2003

2004–2009

42

These findings indicate that in 1998–2003 capabilities research reached a relative consensus on

which works are the most important building blocks of the research stream. It can also be seen

that the extensions to the traditional resource-based view (i.e. the knowledge and learning

approaches etc.) seem to be almost equally popular. In addition, they all are far behind the

central cluster in popularity. Only the dynamic capabilities approach has gained a more

established position at the core of capabilities research.

3.3 Keyword Analyses

This section presents the most interesting findings from different keyword statistics in order to

further highlight the shifts in focus of capabilities research.

In the sub-period 1986–1991, the keywords capability or capabilities do not appear in the citing

articles. This, too, indicates that during this period, the concept of capabilities had not yet

emerged as a separate field of study. In addition, the frequencies of the keywords during this

sub-period are very small: the keywords simulation, attitudes, and evaluation appear in two

citing articles during the sub-period, whereas all other keywords appear only once. This is why

the first sub-period has been excluded also from the following keyword analyses: because of the

small number of keywords, analyzing them would not yield any reliable or interesting results.

Table 4 lists the most common keywords in the citing articles.

Table 4: The 18 most common keywords in 1992–2009

Having excluded such general keywords as firm, management, strategy, perspective, and models

from this list, we are left with some important (although not very surprising) findings. The

frequent occurrence of the keywords innovation, R&D, and product development, for example,

Keyword Count Keyword Count

1 Innovation 663 10 Strategy 321

2 Capabilities 642 11 Technology 290

3 Firm 614 12 Industry 275

4 Performance 556 13 Perspective 235

5 Competitive advantage 465 14 Models 233

6 Resource-based view 375 15 R&D 228

7 Knowledge 363 16 Product development 195

8 Management 351 17 Absorptive capacity 180

9 Dynamic capabilities 346 18 Networks 176

43

can be interpreted as one implication of the somewhat popular approach in empirical research

to use product development performance as a measure of capabilities (Wang & Ahmed 2007).

One interesting remark that can be made from Table 4 is the absence of the keywords

(organizational) routine or routines (Nelson & Winter 1982). Actually, these words have

appeared as keywords in no more than 22 articles in 1992–2009. However, many prominent

authors have considered routines the building blocks of capabilities (Dosi, Nelson & Winter

2000) or capabilities as a type of a routine (Teece, Pisano & Shuen 1997, Eisenhardt & Martin

2000, Winter 2003). On the other hand, Becker (2004) notes that the concept of routines has

gained increasing popularity in the past years.

Felin and Foss (2009) offer an explanation for this phenomenon. They note that there has been

a lack of interest in the origins of capabilities and that “…the dynamic or organizational

capabilities literature … has prematurely moved to higher level or higher order constructs,

without first being clear about the underlying routines construct itself” (p. 162). Thus, routines

might be so implicitly attributed to the concept of capabilities that they have not received

much attention on their own right. In addition, defining capabilities as routines is often

criticized for being tautological and for creating an ‘infinite regress’: capabilities come from

routines, routines come from previous routines etc. (Collis 1994, Felin & Foss 2009).

Figure 2 on the next page presents the changes in the relative frequency of the most common

keywords during the three latter sub-periods. Relative frequency is calculated by dividing the

number of times a keyword has appeared in the citing articles during a sub-period by the total

number citing articles in that sub-period. This ratio shows the percentage of citing articles in

which the keyword has appeared. The keyword dynamic capabilities has gained the most

notable increase in relative frequency. This can be interpreted as another indication of the shift

from the traditional resource-based view to the dynamic capabilities view at the end of the

1990s.

44

Figure 2: Changes in the relative frequency of the most common keywords in 1992–2009

The changes in the keywords’ relative frequencies are further highlighted in Figure 3, which

shows the largest percentage changes in relative frequency from the second to the third and

from the third to the fourth sub-period.

Figure 3: The four keywords with the fastest changing relative frequency from the second to the third and from the third to the fourth sub-period

Figure 3 indicates, above all, that the concept of dynamic capabilities has gained popularity at

the fastest pace. In addition to the concept of dynamic capabilities, the use of the concept of

absorptive capacity seems to have increased significantly. Zahra and George (2002) have

reviewed the discussion around the concept of absorptive capacity and they, too, have noted

the growing use of the construct. According to Zahra and George (ibid.), the concept of

0,0 %

5,0 %

10,0 %

15,0 %

20,0 %

25,0 %

30,0 %

Inno

vatio

n —

Cap

abilitie

s —

Firm —

Perfo

rman

ce —

Com

petitive

adva

ntag

e —

—

Res

ource-

base

d view

——

Knowledg

e —

Man

agem

ent —

Dyn

amic cap

abilit

ies —

—

Strategies

—

Techn

olog

y —

Indu

stry

—

Persp

ectiv

e —

Mod

els —

R&D

—

Produ

ct d

evelop

men

t ——

Absor

ptive

capa

city —

—

Net

wor

ks —

1992–1997

1998–2003

2004–2009

10,0 %

12,3 %

20,5 %

10,1 %

3,8 %5,8 %

10,0 %

8,2 %

0 %

5 %

10 %

15 %

20 %

Dynamic

capabilities

Absorptive

capacity

Models Industry

1998–2003

2004–2009

5,8 % 6,1 %

3,8 %

0,7 %0,5 % 0,7 %1,2 %

7,7 %

0 %

2 %

4 %

6 %

8 %

Dynamic

capabilities

Product

development

Networks Absorptive

capacity

1992–1997

1998–2003

45

absorptive capacity has been attributed to a variety of theoretical lenses, such as organizational

leaning (e.g., Szulanski 1996) and dynamic capabilities (Mowery, Oxley & Silverman 1996),

which has led to diversity and ambiguity in its definitions. Having reviewed past definitions,

Zahra and George (2002) define absorptive capacity as a dynamic capability. This connection

between the concepts of absorptive capacity and dynamic capabilities seems to have gained

considerable popularity: in 1998–2003, three citing articles had both concepts as keywords,

while in 2004–2009 the number of such articles had increased to 60.

This finding is in line with the cluster analyses presented in Chapter 3.2. In 1992–1997, Cohen

and Levinthal’s (1990) article on absorptive capacity was placed in the 0-cluster, which

indicates that the concept’s relatedness to the capabilities discussion was recognized but that

its relatedness to other concepts remained unclear. In 1998–2003, however, the article is placed

in cluster N with representative works from the knowledge-based view. Finally, in the sub-

period 2004–2009, it is separated from the “Knowledge-Based Approaches”-cluster to cluster R,

which includes the seminal works on the resource-based view and on dynamic capabilities.

Thus, it might be that the concept of absorptive capacity has gained increasing popularity ever

since it has been attributed to the concept of dynamic capabilities.

Other interesting remarks can be made from Table 5, which lists the most common keywords

that include the word capability or capabilities in some form.

Table 5: The 18 most common keywords that include the word capability or capabilities in 1992–2009

Table 5 shows that the most popular term attributed to the concept of capabilities is dynamic.

It can also be seen that the concept of capabilities is used in a variety of contexts: from firm-

level analysis to country-level analysis and from IT and technology to marketing and

innovations. Process capability indices, on the other hand, are used in statistical process

Keyword Count Keyword Count

1 Capabilities 642 10 Competitive capabilities 11

2 Dynamic capabilities 346 11 Marketing capabilities 10

3 Organizational capabilities 70 12 Innovation capability 8

4 Technological capabilities 53 13 Core capabilities 7

5 Firm capabilities 27 14 Country capabilities 6

6 IT capabilities 21 15 Resources and capabilities 6

7 Process capability indices 17 16 Capability-accumulation 5

8 Combinative capabilities 16 17 Innovative capability 5

9 Capability maturity model 15 18 Strategic capabilities 5

46

control in manufacturing (for a bibliography, see Spiring et al. 2003) and the capability

maturity model is used in software development (Paulik et al. 1995).

4 DISCUSSION

In this section the research questions are answered by highlighting the key findings from the

citation, co-citation, and keyword analyses.

4.1 The Structure of Capabilities Research

First of all, the analyses in this study have not revealed an individual stream of research that

could be titled “capabilities research”. Rather, this study has revealed the streams of research

that build on the concept of capabilities. The variety of issues attributed to the concept of

capabilities suggests that a view similar to the one of Dosi, Nelson, and Winter (2000) is quite

widely adopted. According to them, “To be capable of some thing is to have a generally reliable

capacity to bring that thing about as a result of intended action.” (p. 2.) Having adopted this

view, almost everything a firm purposefully and repeatedly does can be called a capability: e.g.,

if a firm is able to execute a marketing campaign, it has a marketing capability. This all-

inclusive view of capabilities and the “near tautology of defining capability as an ability” (Zollo

& Winter 2002, p. 340) could explain the multitude of different issues that have been attributed

to this concept.

There are many similarities between the co-citation networks in this study and the ones in the

bibliometric study of Acedo, Barroso and Galan (2006) on the resource-based view. It thus

seems that there are no significant differences between the concepts of capabilities and

resources. Similarly, Barney’s (1991) widely adopted definition of resources as “all assets,

capabilities, organizational processes, firm attributes, information etc. controlled by a firm that

enable the firm to conceive of and implement strategies that improve its efficiency and

effectiveness” (p. 101) makes no distinction between capabilities and resources.

In addition, Priem and Butler (2001a) note that almost everything associated with a firm can be

called its resource. Felin and Foss (2009), on the other hand, note that “…when scholars try to

proffer definitions, they often pack so much into routines and capabilities that they effectively

become identical to the organization itself…” (p. 159.) These ideas are somewhat similar to the

findings of this study. It seems that almost everything a firm possesses can be called its

resource, whereas almost everything it does can be called its capability. Foss (2003) uses the

47

term ‘the organizational capabilities approach’ as an umbrella term covering capabilities,

dynamic capabilities, competence approaches, and the evolutionary theory of the firm. In this

study, capabilities research thus consists of works that apply the organizational capabilities

approach.

It thus can be concluded that, from the viewpoint of this study, the capabilities discussion is

almost identical to the discussion around the resource-based view. Capabilities research has its

roots in the seminal works on the resource-based view and on dynamic capabilities. This

theoretical foundation of capabilities research is quite firmly established. Influential extensions

to this basic core have also emerged. The most important extensions are the knowledge-based

and the learning-based approaches – however, neither one of these has such an established

position as the dynamic capabilities view.

4.2 The Evolution of Capabilities Research

Two conclusions can be made on the evolution of capabilities research. First of all, the low

citation counts in the first sub-period indicate that capabilities research emerged as a field of

study first at the beginning of the 1990s. However, the theoretical foundation of capabilities

research had already been laid by Penrose in 1959, Nelson and Winter in 1982 and Wernerfelt in

1986.

Secondly, the most significant shift in focus of capabilities research is the shift towards the

dynamic capabilities view at the end of the 1990s. The article by Teece, Pisano, and Shuen

(1997) seems to have generated great interest towards this concept. However, only two other

works on dynamic capabilities (Eisenhardt & Martin 2000, Zollo & Winter 2002) have been able

to receive enough citations to be included in this study. The dynamic capabilities view is also

placed in the same cluster with the works on the resource-based view. This, too, indicates that

this stream of research has not yet formed its own, established literature apart from the three

works mentioned above.

In addition to the dynamic capabilities approach, also the learning and knowledge-based

approaches have become more popular in comparison with the traditional resource-based view.

The learning approach dates back to the organization theory classics of March and Simon

(1958) and Thompson (1967), and Argyris and Schön’s (1978) book on organizational learning.

More recently, the learning approach has been attributed to dynamic capabilities (Zollo &

48

Winter 2002). The knowledge-based approach has emerged more recently and all of its most

representative works have been published in the 1990s.

5 LIMITATIONS

The main limitation of a co-citation analysis is that an author may cite a document for multiple

reasons. Co-citation analysis predicts that most citations are made in order to agree with a

work and thus high co-citation strength would indicate high similarity of content between two

documents. However, an author may cite a document also in order to show disagreement with

its ideas. In this study this is partly the case with the works by Williamson (1975, 1985), which

are probably quite often cited as an antithesis for the resource-based view. Porter’s (1980, 1985)

industrial-organization books, on the other hand, seem to be among the most cited works

because of their vast influence on all strategic management research (Nerur, Rasheed &

Natarajan 2008).

Another limitation of bibliometric techniques is that they only focus on the most cited

documents. Because older documents have been exposed to the scientific community for a

longer time they are more likely to be cited. Thus, the most recent developments of capabilities

research are not covered in this study. However, as Ramos-Rodríguez and Ruíz-Navarro (2004)

note, citation count can be considered rather a sign of influence than of quality. Thus, an

analysis based on citation counts is a good way to reveal the intellectual bases of a discipline.

Some limitations also result from the research design of this study. The results have been

influenced by, for example, the journals chosen to be included in the study, the division of the

overall time period into four sub-periods, the algorithms used for generating the clusters of

articles and their network pictures, the chosen cut-off values for the normalized co-citation

strengths, and the search word capabilit* used in searching for the articles in ISI Web of

Science. Due to an easy access through ISI Web of Science, only citations in journal articles

have been analyzed – thus leaving out citations made in books, congress proceedings, and

doctoral theses, for example.

All of these subjective choices also violate the objectivity principle of a bibliometric study.

Some of the limitations mentioned above have no solution (Ramos-Rodríguez & Ruíz-Navarro

2004). The other limitations have been approached by following earlier studies. No conclusions

have been made based on unreliable findings such as the co-citation network of the first sub-

period (1986–1991).

49

6 CONCLUSIONS

Because of the wide range of topics that are placed within capabilities research, the results

obtained in this study are rather general. Now that a wide picture of the use of the concept of

capabilities has been drawn, it would be of value to focus more specifically on some of the

topics within capabilities research. These include, for example, the concepts of dynamic

capabilities, absorptive capacity, and combinative capabilities. Additionally, it would be

interesting to further delve into different forms of capabilities such as marketing, innovation,

or product development capabilities. A bibliometric analysis concentrating specifically on

dynamic capabilities, for example, would be suitable for identifying the origins and more recent

developments of this rapidly growing research stream. A large share of the most cited works in

capabilities research are conceptual and a review of empirical capabilities research would be of

value, too.

The research questions in this study were: “What are the most important academic

publications in capabilities research?” and “How has the focus of capabilities research evolved

between 1986 and 2009?” Looking back at the results of this study, it could be argued that the

bibliometric techniques used may not have been the most appropriate for answering these

questions. If the term ‘capabilities research’ is used to refer to conceptual and empirical

research on capabilities per se, this research is not very well covered in this study. Since this

kind of research has not been very popular (Felin & Foss 2009), such works are not among the

most cited documents that have been analyzed here.

In contrast, this study has investigated to what kinds of topics the concept of capabilities has

been attributed in the past. This study has thus adopted a wide view on capabilities and the

term ‘capabilities research’ is used to refer to research that builds on the concept of capabilities.

This stream of research covers such topics as dynamic capabilities, organizational learning,

knowledge, and technological change. The analyses conducted in this study offer a rather

comprehensive view on the structure and evolution of this kind of research on capabilities.

50

REFERENCES

ACEDO, F. J., BARROSO, C. & GALAN, J. L. (2006). The resource-based theory: dissemination

and main trends. Strategic Management Journal, 27, 621–636.

AMBROSINI, V. & BOWMAN, C. (2009). What are dynamic capabilities and are they a useful

construct in strategic management? International Journal of Management Reviews, 11(1), 29–

49.

AMIT, R. & SCHOEMAKER, P. J. H. (1993). Strategic assets and organizational rent. Strategic

Management Journal, 14, 33–46.

ARENDT, R. A. & BROMILEY, P. (2009). Assessing the dynamic capabilities view: Spare change,

everyone? Strategic Organization, 7(1), 75–90.

ARGYRIS, C. & SCHÖN, D. A. (1978). Organizational Learning. Addison-Wesley: Reading, MA.

ARMSTRONG, C. E. & SHIMIZU, K. (2007). A review of approaches to empirical research on

the resource-based view of the firm. Journal of Management, 33(6), 959–986.

BARNEY, J. B. (1986). Strategic factor markets: expectations, luck, and business strategy.

Management Science, 32(10), 1231–1241.

BARNEY, J. B. (1991). Firm resources and sustained competitive advantage. Journal of

Management, 17(1), 99–120.

BARNEY, J. B. (2001). Is the resource-based “view” a useful perspective for strategic

management? Yes. Academy of Management Review, 26(1), 41–56.

BATAGELJ, V. & MRVAR, A. (2009). Pajek 1.24. [Available at http://vlado.fmf.uni-

lj.si/pub/networks/pajek/]

BAUMGARTNER, H. & PIETERS, R. (2003). The structural influence of marketing journals: A

citation analysis of the discipline and its subareas over time. Journal of Marketing, 67, 123–

139.

BECKER, M. C. (2004). Organizational routines: a review of the literature. Industrial and

Corporate Change, 13(4), 643–677.

BELL, P. C. (1985). Visual Interactive Modelling in Operational Research: Successes and

Opportunities. The Journal of the Operational Research Society, 36(11), 975–982.

BORGATTI, S. P., EVERETT, M. G. & FREEMAN, L. C. (2002). Ucinet for Windows: Software for

Social Network Analysis. Harvard, MA: Analytic Technologies.

BROADUS, R. N. (1987). Toward a definition of “bibliometrics.” Scientometrics, 12(5–6), 373–

379.

51

CAHLIK, T. (2000). Search for fundamental articles in economics. Scientometrics, 49(3), 389–

402.

CLARK, K. B. & FUJIMOTO, T. (1991). Product Development Performance: Strategy,

Organization, and Management in the World Auto Industry. Harvard Business School Press:

Boston, MA.

COHEN, W. M. & LEVINTHAL, D. A. (1990). Absorptive capacity: A new perspective on

learning and innovation. Administrative Science Quarterly, 35, 128–152.

COLLIS, D. J. (1994). Research note: How valuable are organizational capabilities? Strategic


CULNAN, M. J. & SWANSON, E. B. (1986). Research in management information systems,

1980-1984: Points of work and reference. MIS Quarterly, 10(3), 289-302.

CYERT, R. M. & MARCH, J. G. (1963). A Behavioral Theory of the Firm. Prentice-Hall:

Englewood Cliffs, NJ.

DAFT, R. (1983). Organization theory and design. West: New York.

DAY, G. S. (1994). The capabilities of market-driven organizations. Journal of Marketing, 58(4),

37–52.

DIERICKX, I. & COOL, K. (1989). Asset stock accumulation and sustainability of competitive

advantage. Management Science, 35(12), 1504–1511.

DOSI, G. R., NELSON, R. R. & WINTER, S. G. (2000). Introduction: The nature and dynamics of

organizational capabilities. In G. R. Dosi, R. R. Nelson & S. G. Winter (Eds.), The Nature and

Dynamics of Organizational Capabilities. Oxford University Press: Oxford.

DYER, J. H. & SINGH, H. (1998). The relational view: Cooperative strategy and sources of

interorganizational competitive advantage. Academy of Management Review, 23(4), 660–

679.

EASTERBY-SMITH, M., LYLES, M. A. & PETERAF, M. A. (2009). Dynamic capabilities: Current

debates and future directions. British Journal of Management, 20, S1–S8.

EISENHARDT, K. M. & MARTIN, J. A. (2000). Dynamic capabilities: What are they? Strategic


EISENHARDT, K. M. (1989). Building theories from case-study research. Academy of

Management Review, 14(4), 532–550.

FELIN, T. & FOSS, N. J. (2009). Organizational routines and capabilities: Historical drift and a

course-correction toward microfoundations. Scandinavian Journal of Management, 25, 157–

167.

52

FOSS, N. J. (2003). Bounded rationality and tacit knowledge in the organizational capabilities

approach: An evaluation and a stock-taking. Industrial and Corporate Change, 12, 185–201.

GMÜR, M. (2003). Co-citation analysis and the search for invisible colleges: A methodological

evaluation. Scientometrics, 57(1), 27–57.

GRANT, R. M. (1991). The resource-based view theory of competitive advantage: implications

for strategy formulation. California Management Review, 33(3), 114–135.

GRANT, R. M. (1996a). Prospering in dynamically-competitive environments: Organizational

capability as knowledge integration. Organization Science, 7(4), 375–387.

GRANT, R. M. (1996b). Toward a knowledge-based theory of the firm. Strategic Management

Journal, 17(Winter Special Issue), 109–122.

HAYES, R. H., WHEELWRIGHT, S. C. (1984). Restoring Our Competitive Edge: Competing

Through Manufacturing. Wiley: New York.

HAYES, R. H., WHEELRIGHT, S. C. & CLARK, K. B. (1988). Dynamic Manufacturing. The Free

Press: New York.

HENDERSON, R. M. & CLARK, K. B. (1990). Architectural innovation: The reconfiguration of

existing product technologies and the failure of established firms. Administrative Science

Quarterly, 35, 9–30.

HENDERSON, R. M. & COCKBURN, I. (1994). Measuring competence? Exploring firm effects in

pharmaceutical research. Strategic Management Journal, 15, 63–84.

HOFFMAN, D. L. & HOLBROOK, M. B. (1993). The intellectual structure of consumer research:

A bibliometric study of author cocitations in the first 15 years of the Journal of Consumer

Research. Journal of Consumer Research, 19(4), 505–517.

ISI Journal Citation Report, Social Science 2008 Edition. [Available at www.isiknowledge.com]

ITAMI, H. & ROEHL, T. (1987). Mobilizing Invisible Assets. Harvard University Press:

Cambridge, MA.

JACOBIDES, M. G. & WINTER, S. G. (2005). The co-evolution of capabilities and transaction

costs: Explaining the institutional structure of production. Strategic Management Journal,

26, 395–413.

JACOBIDES, M. G. (2006). The architecture and design of organizational capabilities. Industrial

and Corporate Change, 15(1), 151–171.

JÄRVINEN, J. M. J. & SILLANPÄÄ, A. K. (2007). The evolution of evolution: A bibliometric study

on evolutionary research, Helsinki University of Technology, Espoo.

KOGUT, B. & ZANDER, U. (1992). Knowledge of the firm, combinative capabilities, and the

replication of technology. Organization Science, 3(3), 383–397.

53

LAWRENCE, P. R. & LORSCH, J. W. (1967). Organization and Environment: Managing

Differentiation and Integration. Irwin: Homewood, IL.

LEONARD-BARTON, D. (1992). Core capabilities and core rigidities: A paradox in managing

new product development. Strategic Management Journal, 13, 111–125.

LEVINTHAL, D. A. & MARCH, J. G. (1993). The myopia of learning. Strategic Management

Journal, 14, 95–112.

LEVITT, B. & MARCH, J. G. (1988). Organizational learning. Annual Review of Sociology, 14,

319–340.

LIPPMAN, S. A. & RUMELT, R. P. (1982). Uncertain imitability: An analysis of interfirm

differences in efficiency under competition. The Bell Journal of Economics, 13(2), 418–438.

MARCH, J. G. & SIMON, H. A. (1958). Organizations. Wiley: New York.

MARCH, J. G. (1991). Exploration and exploitation in organizational learning. Organization

Science, 2(1), 71–87.

MARTINSONS, M. G., EVERETT, J. E. & CHAN, K. (2001). Mapping the scholarly development

of strategic management. Journal of Information Science, 27(2), 101–110.

MOWERY, D. C., OXLEY, J. E. & SILVERMAN, B. S. (1996). Strategic alliances and interfirm

knowledge transfer. Strategic Management Journal, 17(Winter Special Issue), 77–91.

NELSON, R. R. & WINTER, S. G. (1982). An Evolutionary Theory of Economic Change. Harvard

University Press: Cambridge, MA.

NERUR, S. P., RASHEED, A. A. & NATARAJAN, V. (2008). The intellectual structure of the

strategic management field: An author co-citation analysis. Strategic Management Journal,

29, 319–336.

NEWBERT, S. L. (2007). Empirical research on the resource-based view of the firm: An

assessment and suggestions for future research. Strategic Management Journal, 28, 121–156.

NONAKA I. & TAKEUCHI, H. (1995). The Knowledge-Creating Company. Oxford University

Press: New York.

NONAKA, I. (1994). A dynamic theory of organizational knowledge creation. Organization

Science, 5(1), 14–37.

PAULIK, M. C., WEBER, C. V., CURTIS, B. & CHRISSIS, M. B. (1995). The Capability Maturity

Model: Guidelines for Improving the Software Process. Addison-Wesley: Boston.

PENG, D. X., SCHROEDER, R. G. & SHAH, R. (2008). Linking routines to operations

capabilities: A new perspective. Journal of Operations Management, 26, 730–748.

PENROSE, E. T. (1959). The Theory of the Growth of the Firm. Basil Blackwell: Oxford.

54

PETERAF, M. A. (1993). The cornerstones of competitive advantage: A resource-based view.

Strategic Management Journal, 14, 179–191.

PILKINGTON, A. & LISTON-HEYES, C. (1999). Is production and operations management a

discipline? A citation/co-citation study. International Journal of Operations & Production

Management, 19(1), 7–20.

PILKINGTON, A. & MEREDITH, J. (2009). The evolution of the intellectual structure of

operations management—1980–2006: A citation/co-citation analysis. Journal of Operations

Management, 27, 185–202.

PONZI, L. J. (2002). The intellectual structure and interdisciplinary breadth of Knowledge

Management: A bibliometric study of its early stage of development. Scientometrics, 55(2),

259–272.

PORTER, M. E. (1980). Competitive Strategy. Free Press: New York.

PORTER, M. E. (1985). Competitive Advantage. Free Press: New York.

PRAHALAD, C. K. & HAMEL, G. (1990). The core competence of the corporation. Harvard

Business Review, May–June, 79–91.

PRIEM, R. L. & BUTLER, J. E. (2001a). Is the resource-based “view” a useful perspective for

strategic management research? Academy of Management Review, 26(1), 22–40.

PRIEM, R. L. & BUTLER, J. E. (2001b). Tautology in the resource-based view and the

implications of externally determined resource value: Further comments. Academy of

Management Review, 26(1), 57–66.

PRITCHARD, A. (1969). Statistical bibliography or bibliometrics? Journal of Documentation, 25,

348–349.

RAMOS-RODRÍGUEZ, A.-R. & RUÍZ-NAVARRO, J. (2004). Changes in the intellectual

structure of strategic management research: A bibliometric study of the Strategic

Management Journal, 1980–2000. Strategic Management Journal, 25, 981–1004.

SCHILDT, H. A. & MATTSSON, J. T. (2006). A dense network sub-grouping algorithm for co-

citation analysis and its implementation in the software tool Sitkis. Scientometrics, 67(1),

143–163.

SCHILDT, H. A. (2004). SITKIS: Software for Bibliometric Data Management and Analysis v1.0.

MANUAL. Helsinki University of Technology: Institute of Strategy and International

Business. [Available at: www.hut.fi/~hschildt/sitkis]

SCHILDT, H. A. (2005). SITKIS: Software for Bibliometric Data Management and Analysis v1.5.

Helsinki University of Technology: Institute of Strategy and International Business.

[Available at: www.hut.fi/~hschildt/sitkis]

55

SCHILDT, H. A., ZAHRA, S. A. & SILLANPÄÄ, A. (2006). Scholarly communities in

entrepreneurship research: A co-citation analysis. Entrepreneurship Theory and Practice,

30(3), 399–415.

SCOTT, J. (2000). Social Network Analysis: A Handbook. Sage Publications: Thousand Oaks,

CA.

SMALL, H. (1973). Co-citation in the scientific literature: A new measure of the relationship

between two documents. Journal of the American Society for Information Science, 24(4),

265–269.

SPIRING, F., LEUNG, B., CHENG, S. & YEUNG, A. (2003). A bibliography of process capability

papers. Quality and Reliability Engineering International, 19, 445–460.

STALK, G., EVANS, P. & SHULMAN, L. E. (1992). Competing on capabilities: The new rules of

corporate strategy. Harvard Business Review, March–April, 57–69.

SZULANSKI, G. (1996). Exploring internal stickness: Impediments to the transfer of best

practice within the firm. Strategic Management Journal, 17(Winter Special Issue), 27–43.

TEECE, D. J. (1986). Profiting from technological innovation: Implications for integration,

collaboration, licensing and public policy. Research Policy, 15, 285–305.

TEECE, D. J., Pisano, G. & Shuen, A. (1997). Dynamic capabilities and strategic management.

Strategic Management Journal, 18(7), 509–533.

THOMPSON, J. D. (1967). Organizations in Action. McGraw-Hill: New York.

TUSHMAN, M. L. & ANDERSON, P. (1986). Technological discontinuities and organizational

environments. Administrative Science Quarterly, 31, 439–465.

WANG, C. L. & AHMED, P. K. (2007). Dynamic capabilities: A review and research agenda.

International Journal of Management Reviews, 9(1), 31–51.

WERNERFELT, B. (1984). A resource-based view of the firm. Strategic Management Journal, 5,

171–180.

WHEELWRIGHT, S. C. & CLARK, K. B. (1992). Revolutionizing Product Development. Free

Press: New York.

WHITE, H. D. & MCCAIN, K. W. (1989). Bibliometrics. In M. E. Williams (Ed.), Annual Review

of Information Science and Technology, Volume 24. Elsevier: Amsterdam.

WILLIAMSON, O. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. Free

Press: New York.

WILLIAMSON, O. E. (1985). The Economic Institutions of Capitalism: Firms, Markets,

Relational Contracting. Free Press: New York.

56

WILLIAMSON, O. E. (1991). Strategizing, economizing, and economic organization. Strategic

Management Journal, 12(Winter Special Issue), 75–94.

WILLIAMSON, O. E. (1999). Strategy research: Governance and competence perspectives.

Strategic Management Journal, 20, 1087–1108.

WINTER, S. G. (1987). Knowledge and competence as strategic assets. In D. J. Teece (Ed.), The

Competitive Challenge: Strategies for Industrial Innovation and Renewal. Harper & Row: New

York.

WINTER, S. G. (2003). Understanding dynamic capabilities. Strategic Management Journal, 24,

991–995.

ZAHRA, S. A. & GEORGE, G. (2002). Absorptive capacity: A review, reconceptualization, and

extension. Academy of Management Review, 27(2), 185–203.

ZANDER, U. & KOGUT, B. (1995). Knowledge and the speed of transfer and imitation of

organizational capabilities: An empirical test. Organization Science, 6(1), 76–92.

ZOLLO, M. & WINTER, S. G. (2002). Deliberate learning and the evolution of dynamic

capabilities. Organization Science, 13(3), 339–351.

57

APPENDICES

Appendix 1: Journals Included in the Study

Journal

Impact

factor

# of

records

Academy of Management Journal 6.079 47

Academy of Management Learning & Education 2.889 2

Academy of Management Review 6.125 48

Administrative Science Quarterly 2.853 12

European Journal of Operational Research 1.627 113

Harvard Business Review 1.793 91

IEEE Transactions on Engineering Management 1.156 53

Industrial and Corporate Change 1.165 64

Industrial Marketing Management 1.403 76

Information & Management 2.358 54

International Journal of Operations Production Management 1.725 95

International Journal of Technology Management 0.526 208

Journal of Business Research 0.943 72

Journal of Consumer Psychology 2.841 1

Journal of International Business Studies 2.992 74

Journal of Management 3.080 34

Journal of Management Information Systems 2.358 35

Journal of Management Studies 2.558 88

Journal of Marketing 3.598 44

Journal of Marketing Research 2.574 8

Journal of Operations Management 2.420 53

Journal of Organizational Behavior 2.441 7

Journal of Product Innovation Management 2.650 70

Journal of Retailing 4.095 5

Long Range Planning 1.617 53

Management Science 2.354 79

Marketing Science 3.309 17

MIS Quarterly 5.183 40

Organizational Behavior and Human Decision Processes 2.740 7

Organizational Research Methods 3.019 2

Organization Science 2.575 98

Organization Studies 1.857 46

R & D Management 2.043 59

Research in Organizational Behavior 2.444 2

Research Policy 2.655 134

Strategic Management Journal 3.344 221

Technology Analysis & Strategic Management 0.735 67

Technological Forecasting and Social Change 1.761 54

Technovation 1.907 122

58

Appendix 2: Citation Statistics – Raw and relative frequencies of the most

cited documents

Rank Citations Rel. freq. Rank Citations Rel. freq. Rank Citations Rel. freq. Rank Citations Rel. freq. Rank Citations Rel. freq.

Teece, Pisano & Shuen (1997) 1 609 22,1 % 0 0 0,0 % 0 0 0,0 % 3 177 20,4 % 1 432 30,1 %

Nelson & Winter (1982) 2 586 21,3 % 1 2 5,4 % 2 62 15,0 % 1 213 24,5 % 4 309 21,5 %

Barney (1991) 3 579 21,0 % 0 0 0,0 % 5 34 8,3 % 2 197 22,7 % 2 348 24,2 %

Cohen & Levinthal (1990) 4 508 18,5 % 0 0 0,0 % 8 27 6,6 % 4 170 19,6 % 3 311 21,7 %

Kogut & Zander (1992) 5 412 15,0 % 0 0 0,0 % 13 20 4,9 % 6 146 16,8 % 6 246 17,1 %

Wernerfelt (1984) 6 386 14,0 % 2 1 2,7 % 6 29 7,0 % 5 155 17,9 % 8 201 14,0 %

Prahalad & Hamel (1990) 7 350 12,7 % 0 0 0,0 % 3 56 13,6 % 7 140 16,1 % 12 154 10,7 %

Penrose (1959) 8 346 12,6 % 0 0 0,0 % 7 29 7,0 % 9 108 12,4 % 7 209 14,6 %

Dierickx & Cool (1989) 9 320 11,6 % 0 0 0,0 % 10 25 6,1 % 8 119 13,7 % 9 176 12,3 %

Leonard-Barton (1992) 10 302 11,0 % 0 0 0,0 % 14 20 4,9 % 10 106 12,2 % 10 176 12,3 %

Eisenhardt & Martin (2000) 11 300 10,9 % 0 0 0,0 % 0 0 0,0 % 30 44 5,1 % 5 256 17,8 %

Porter (1985) 12 300 10,9 % 1 2 5,4 % 1 69 16,7 % 12 97 11,2 % 16 132 9,2 %

Porter (1980) 13 261 9,5 % 1 2 5,4 % 4 41 10,0 % 11 101 11,6 % 18 117 8,1 %

March (1991) 14 257 9,3 % 0 0 0,0 % 22 14 3,4 % 19 73 8,4 % 11 170 11,8 %

Peteraf (1993) 15 248 9,0 % 0 0 0,0 % 21 16 3,9 % 14 88 10,1 % 14 144 10,0 %

Nonaka & Takeuchi (1995) 16 238 8,6 % 0 0 0,0 % 25 5 1,2 % 13 92 10,6 % 15 141 9,8 %

Amit & Schoemaker (1993) 17 219 8,0 % 0 0 0,0 % 17 18 4,4 % 15 87 10,0 % 20 114 7,9 %

Henderson & Clark (1990) 18 217 7,9 % 0 0 0,0 % 11 23 5,6 % 16 80 9,2 % 21 114 7,9 %

Williamson (1985) 19 210 7,6 % 0 0 0,0 % 12 22 5,3 % 17 75 8,6 % 22 113 7,9 %

Grant (1996b) 20 207 7,5 % 0 0 0,0 % 0 0 0,0 % 26 56 6,5 % 13 151 10,5 %

Teece (1986) 21 202 7,3 % 0 0 0,0 % 18 17 4,1 % 21 64 7,4 % 17 121 8,4 %

Nonaka (1994) 22 191 6,9 % 0 0 0,0 % 24 6 1,5 % 20 69 7,9 % 19 116 8,1 %

Williamson (1975) 23 183 6,6 % 0 0 0,0 % 9 27 6,6 % 27 56 6,5 % 25 100 7,0 %

Levitt & March (1988) 24 178 6,5 % 0 0 0,0 % 19 17 4,1 % 24 61 7,0 % 26 100 7,0 %

Barney (1986) 25 177 6,4 % 0 0 0,0 % 20 17 4,1 % 18 75 8,6 % 29 85 5,9 %

Grant (1996a) 26 173 6,3 % 0 0 0,0 % 23 7 1,7 % 28 56 6,5 % 23 110 7,7 %

Tushman & Anderson (1986) 27 169 6,1 % 0 0 0,0 % 15 20 4,9 % 29 49 5,6 % 27 100 7,0 %

Szulanski (1996) 28 166 6,0 % 0 0 0,0 % 0 0 0,0 % 22 62 7,1 % 24 104 7,2 %

Henderson & Cockburn (1994) 29 165 6,0 % 0 0 0,0 % 26 5 1,2 % 23 62 7,1 % 28 98 6,8 %

Grant (1991) 30 159 5,8 % 0 0 0,0 % 16 19 4,6 % 25 58 6,7 % 30 82 5,7 %

2004–2009

1436 citing articles

1992–1997

412 citing articles

1998–2003

868 citing articles

1986–2009

2753 citing articles

1986–1991

37 citing articles

59

Appendix 3: Clusters in 1986–1991

Author(s) Year Journal / Book Citations Cluster

Bonczek et al. 1981 Foundations of Decision Support Systems 2 0

Cyert & March 1963 A Behavioral Theory of the Firm 2 0

Newman & Scodro 1988 Journal of Purchasing and Materials Management 2 0

Porter 1985 Competitive Advantage 2 0

Bell 1985 The Journal of the Operational Research Society 3 A

Parker & Bell 1988 Visual Interactive Modeling on Microcomputers 2 A

Sprague & Carlson 1982 Building effective decision support systems 2 A

Chandler 1990 Scale and Scope: Dynamics of Industrial Capitalism 2 B

Donaldson & Lorsch 1983 Decision Making at the Top 2 B

Moskowitz 1980 Everybody's Business 2 B

Dess & Davis 1984 Academy of Management Journal 2 C

Porter 1980 Competitive Strategy 2 C

Lippman & Rumelt 1982 The Bell Journal of Economics 2 D

Nelson & Winter 1982 An Evolutionary Theory of Economic Change 2 D

Teece 1987 The Competitive challenge 2 D

Hall 1971 Psychology Today 2 E

Mintzberg et al. 1976 Administrative Science Quarterly 2 E

Quinn 1980 Strategies for change 2 E

60



Cohen & Levinthal 1990 Administrative Science Quarterly 23 0

Cyert & March 1963 A Behavioral Theory of the Firm 18 0

Lawrence & Lorsch 1967 Organization and Environment 24 0

Porter 1985 Competitive Advantage 69 0

Stalk et al. 1992 Harvard Business Review 18 0

Teece 1986 Research Policy 17 0

Amit & Schoemaker 1993 Strategic Management Journal 18 F

Barney 1986 Management Science 34 F

Barney 1991 Journal of Management 17 F

Dierickx & Cool 1989 Management Science 25 F

Grant 1991 California Management Review 19 F

Itami & Roehl 1987 Mobilizing Invisible Assets 18 F

Leonard-Barton 1992 Strategic Management Journal 20 F

Penrose 1959 The Theory of the Growth of the Firm 29 F

Porter 1980 Competitive Strategy 41 F

Prahalad & Hamel 1990 Harvard Business Review 56 F

Wernerfelt 1984 Strategic Management Journal 28 F

Argyris & Schön 1978 Organizational learning 22 G

Levitt & March 1988 Annual Review of Sociology 17 G

March & Simon 1958 Organizations 19 G

Thompson 1967 Organizations in Action 23 G

Henderson & Clark 1990 Administrative Science Quarterly 23 H

Nelson & Winter 1982 An Evolutionary Theory of Economic Change 62 H

Tushman & Anderson 1986 Administrative Science Quarterly 18 H

Kogut & Zander 1992 Organization Science 20 I

Lippman & Rumelt 1982 The Bell Journal of Economics 19 I

Winter 1987 The Competitive challenge 17 I

Clark & Fujimoto 1991 Product development performance 22 J

Wheelwright 1992 Revolutionizing Product Development 18 J

Williamson 1975 Markets and Hierarchies 27 K

Williamson 1985 The Economic Institutions of Capitalism 22 K

Hayes et al. 1988 Dynamic manufacturing 23 L

Hayes & Wheelwright 1984 Restoring Our Competitive Edge 18 L

61

62



Clark & Fujimoto 1991 Product development performance 67 0

Grant 1991 California Management Review 58 0

Henderson & Cockburn 1994 Strategic Management Journal 62 0

Amit & Schoemaker 1993 Strategic Management Journal 87 M

Barney 1986 Management Science 75 M

Barney 1991 Journal of Management 197 M

Dierickx & Cool 1989 Management Science 119 M

Kogut & Zander 1992 Organization Science 146 M

Lippman & Rumelt 1982 The Bell Journal of Economics 58 M

Nelson & Winter 1982 An Evolutionary Theory of Economic Change 213 M

Penrose 1959 The Theory of the Growth of the Firm 108 M

Peteraf 1993 Strategic Management Journal 88 M

Porter 1985 Competitive Advantage 97 M

Porter 1980 Competitive Strategy 101 M

Prahalad & Hamel 1990 Harvard Business Review 140 M

Teece et al. 1997 Strategic Management Journal 177 M

Wernerfelt 1984 Strategic Management Journal 155 M

Cohen & Levinthal 1990 Administrative Science Quarterly 170 N

Nonaka 1994 Organization Science 69 N

Nonaka & Takeuchi 1995 The Knowledge-Creating Company 92 N

Szulanski 1996 Strategic Management Journal 62 N

Winter 1987 The Competitive challenge 68 N

Zander & Kogut 1995 Organization Science 64 N

Levitt & March 1988 Annual Review of Sociology 61 O

March 1991 Organization Science 73 O

March & Simon 1958 Organizations 59 O

Henderson & Clark 1990 Administrative Science Quarterly 80 P

Leonard-Barton 1992 Strategic Management Journal 106 P

Teece 1986 Research Policy 64 Q

Williamson 1985 The Economic Institutions of Capitalism 75 Q

63

64



Day 1994 Journal of Marketing 94 0

Dyer & Singh 1998 Academy of Management Review 112 0

Eisenhardt 1989 Academy of Management Review 95 0

Barney 1991 Journal of Management 348 R

Cohen & Levinthal 1990 Administrative Science Quarterly 311 R

Dierickx & Cool 1989 Management Science 176 R

Eisenhardt & Martin 2000 Strategic Management Journal 256 R

Grant 1996b Strategic Management Journal 151 R

Kogut & Zander 1992 Organization Science 246 R

Leonard-Barton 1992 Strategic Management Journal 176 R

Nelson & Winter 1982 An Evolutionary Theory of Economic 309 R

Penrose 1959 The Theory of the Growth of the Firm 209 R

Prahalad & Hamel 1990 Harvard Business Review 154 R

Teece et al. 1997 Strategic Management Journal 432 R

Wernerfelt 1984 Strategic Management Journal 201 R

Zollo & Winter 2002 Organization Science 130 R

Amit & Schoemaker 1993 Strategic Management Journal 114 S

Henderson & Cockburn 1994 Strategic Management Journal 98 S

Peteraf 1993 Strategic Management Journal 144 S

Levinthal & March 1993 Strategic Management Journal 100 T

Levitt & March 1988 Annual Review of Sociology 100 T

March 1991 Organization Science 170 T

Henderson & Clark 1990 Administrative Science Quarterly 114 U

Tushman & Anderson 1986 Administrative Science Quarterly 100 U

Teece 1986 Research Policy 121 V

Williamson 1985 The Economic Institutions of Capitalism 113 V

Williamson 1975 Markets and Hierarchies 100 V

Grant 1996a Organization Science 110 W

Nonaka 1994 Organization Science 116 W

Nonaka & Takeuchi 1995 The Knowledge-Creating Company 141 W

Szulanski 1996 Strategic Management Journal 104 W

Porter 1985 Competitive Advantage 132 X

Porter 1980 Competitive Strategy 117 X

65

66

Appendix 7: Changes in Clusters

Cluster in Cluster in Cluster in

Author(s) Year 1992–1997 1998–2003 2004–2009

Amit & Schoemaker 1993 F M S

Argyris & Schön 1978 G – –

Barney 1986 F M –

Barney 1991 F M R

Clark & Fujimoto 1991 J 0 –

Cohen & Levinthal 1990 0 N R

Cyert & March 1963 0 – –

Day 1994 – – 0

Dierickx & Cool 1989 F M R

Dyer & Singh 1998 – – 0

Eisenhardt 1989 – – 0

Eisenhardt & Martin 2000 – – R

Grant 1991 F 0 –

Grant 1996a – – W

Grant 1996b – – R

Hayes & Wheelwright 1984 L – –

Hayes et al. 1988 L – –

Henderson & Clark 1990 H P U

Henderson & Cockburn 1994 – 0 S

Itami & Roehl 1987 F – –

Kogut & Zander 1992 I M R

Lawrence & Lorsch 1967 0 – –

Leonard-Barton 1992 F P R

Levinthal & March 1993 – – T

Levitt & March 1988 G O T

Lippman & Rumelt 1982 I M –

March 1991 – O T

March & Simon 1958 G O –

Nelson & Winter 1982 H M R

Nonaka 1994 – N W

Nonaka & Takeuchi 1995 – N W

Penrose 1959 F M R

Peteraf 1993 – M S

Porter 1985 0 M X

Porter 1980 F M X

Prahalad & Hamel 1990 F M R

Stalk et al. 1992 0 – –

Szulanski 1996 – N W

Teece 1986 0 Q V

Teece et al. 1997 – M R

Thompson 1967 G – –

Tushman & Anderson 1986 H – U

Wernerfelt 1984 F M R

Wheelwright 1992 J – –

Williamson 1975 K – V

Williamson 1985 K Q V

Winter 1987 I N –

Zander & Kogut 1995 – N –

Zollo & Winter 2002 – – R

67

CHAPTER 4

OPERATIONALIZATION OF CAPABILITIES: A REVIEW

MIRVA PELTONIEMI


OLA LAAKSONEN


ULRIIKKA TIKKANEN

Aalto University School of Science and Technology, Turku School of Economics [email protected]

1 INTRODUCTION

Despite the popularity of capabilities in explaining competitive advantage (Henderson and

Cockburn, 1994, Teece et al., 1997, Eisenhardt and Martin, 2000, Zollo and Winter, 2002),

empirical work relating to the concept has not to date developed common methodological

solutions and established operationalizations for different types of capabilities. Capability

remains a problematic construct both conceptually and empirically. Thus there is a need to

look at how the problem of operationalizing capabilities has been solved. The present paper

aims at examining empirical capabilities research to shed light on the variables and

operationalizations used.

68

Recent reviews (Newbert, 2007, Wang and Ahmed, 2007, Armstrong and Shimizu, 2007) on the

resource-based view of the firm and dynamic capabilities have touched the issue of empirical

grounding, but still their emphasis has been on conceptual rather than empirical issues. We

follow Armstrong and Shimizu’s (2007) approach in reviewing the empirical work through

classifications, but take the analysis to the level of variables: How do capabilities manifest in

empirical material? When does a capability exist? When is there a lot of a capability and when

only a little? Some reviews on empirical work assess the level of empirical support for a theory

through meta-analysis (e.g. Newbert, 2007, Wang and Ahmed, 2007, Ketchen et al., 1997,

Dalton et al., 2003, Campbell-Hunt, 2000) whereas others assess the choices made in

operationalization (e.g. Karren and Barringer, 2002, Zwijze-Koning and de Jong, 2007, Lohrke,

2009, Añón Higón et al., 2009). The present paper falls under the latter category.

2 METHOD

2.1 Identification of empirical capabilities articles

The first decision was to choose the journals to be included in our study. We chose to include

Strategic Management Journal, Administrative Science Quarterly, Academy of Management

Journal, Research Policy, Journal of Management, Journal of Management Studies and

Organization Science due to their high impact factors, high ratio of empirical articles and

general high regard in the academic community. The search was conducted in ISI Web of

Science and includes all articles published in 1986-2009.

The method applied in identifying empirical capabilities articles followed that used by David

and Han (2004) and Newbert (2007) in reviews on empirical work on the transaction cost

economics and the resource-based view of the firm, respectively. We proceeded as follows.

1. Ensure relevance by requiring that selected articles contain the word capabilit* in title,

abstract or keywords.

2. Ensure empirical content by requiring that selected articles also contain at least one of

the following method-related keywords in title, abstract or keywords: data, empirical,

test*, finding*, result, results, qualitative, quantitative, or evidence.

3. Ensure empirical content by reading through the remaining abstracts and discard those

that do not contain empirical work.

69

4. Ensure thematic relevance by reading through the remaining papers and discard those

that do not have capabilities as an empirical construct.

Table 1 presents the numbers of articles left from each journal after each round of elimination.

In the end we were left with 174 articles.

Table 1. The elimination process.

Total SMJ ASQ AMJ JM JMS OrgSci ResPol

capabilit* 596 218 10 40 25 72 98 133

Method-related

keywords 410 163 7 28 15 41 64 92

Empirical content 354 131 7 24 11 37 58 86

Capability as

empirical

construct

174 74 1 9 7 15 27 41

Figure 1 shows the distribution of the articles found in this manner across years. It is

noteworthy that empirical capabilities research appears to have found its way to top journals as

late as mid-1990s.

Figure 1 – Empirical capabilities articles across years in SMJ, ASQ, AMJ, ResPol, JM, JMS and OrgSci.

0

5

10

15

20

25

30

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

70

2.2 Coding scheme

After a careful reading of the articles a coding scheme was established. It comprises three

factors, namely capability type, data type, and variable type. Firstly, coding was performed on

the basis of the type of capability under study. These were identified as technological,

organizational, and market capabilities. Capabilities relating to technology, R&D, and

manufacturing were coded as technological capabilities. Capabilities relating to management,

change, and partnering were coded as organizational capabilities. Capabilities relating to

internationalization, pricing, sales, and marketing were coded as market capabilities.

Figure 2 – Coding based on capability type.

Secondly, coding was performed on the basis of data type. The following list of 10 data types

emerged from the material during the study. Figure 3 shows the distribution of studies based

on data type.

1. Survey, subjective

2. Survey, objective

3. Patents and scientific articles

4. Experience in years

5. Innovative actions

6. R&D expenditure

7. Education level

8. Interviews

9. R&D collaboration

10. Financial data

11. Other

0

20

40

60

80

100

Technological Organizational Market

71

Figure 3 – Coding based on data type.

Many studies used survey data and the surveys can be divided into those that ask about the

managers’ views on the capabilities of the firm compared to their competitors and those that

ask about factual data. An example of the former is Bhattacharya and Gibson’s (2005) study on

the flexibility of employee skills, which were measured through a survey with questions on the

managers’ perceptions on the ability of the employees to make changes. An example of the

latter is the study by Schroeder et al. (2002) on manufacturing capabilities measured through a

survey with questions relating to equipment and employee skills. Patents, scientific articles,

and their citations have been a popular data type used to determine the amount of a firm’s

capability in a particular technological area (e.g. Penner-Hahn and Shaver, 2005, Quintana-

Garcia and Benavides-Velasco, 2008) and the similarity of the capability stocks of firm pairs

(e.g. Mowery et al., 1998, Rothaermel and Boeker, 2008). Experience in years relates to

measuring acquired capability stock through the number of years that a firm has been active in

a particular technology (e.g. Nerkar and Roberts, 2004), product class (e.g. Kaplan, 2008), or

market area (e.g. Song, 2002).

Innovative actions comprise measures relating to innovations (e.g. Romijn and Albaladejo,

2002), new product launches (e.g. Zahra and Nielsen, 2002), and sales of new products (e.g.

Faber and Hesen, 2004). R&D expenditure has been used to reflect the capabilities of a single

firm (e.g. Yeoh and Roth, 1999) or a nation (e.g. Daniels, 1997). Similarly, education level has

been used to determine firm-level (e.g. Boeker and Wiltbank, 2005) and country-level (e.g.

0

10

20

30

40

50

60

72

Archibugi and Coco, 2005) capabilities. Interview data has been used in studies that are

exploratory and qualitative (e.g. Montealegre, 2002, Haas and Hansen, 2005). R&D

collaboration relationships have been used by Faber and Hesen (2004), for example, to signal

the extent of a firm’s technological capabilities. Financial data, such as wages (Del Canto and

Gonzalez, 1999), sales (Bierly et al., 2009), exports (Mahmood and Singh, 2003), and

productivity (Brush and Artz, 1999), have been used to find differences between firms operating

in the same industry. Finally, there are miscellaneous rather rarely used data types, such as

employee turnover (Traore and Rose, 2003), the existence of a design office (Bougrain and

Haudeville, 2002), and advertising intensity (Tan and Mahoney, 2006).

Finally, the articles were classified on the basis of the type of the used capability variables as

follows.

1. Existence of a capability. The value of this type of capability variable is either 1 or 0. It

determines whether a capability exists or not.

2. The amount of a capability. The value of this type of capability variable is numeral

usually ranging from zero to infinite or from zero to some upper limit which denotes

the maximum amount of capability the firm may have.

3. Similar or different capabilities. The value of this type of capability variable is either 1 or

0. It determines whether the capabilities possessed by two firms or the same firm at two

different points in time are similar or different.

4. The degree of similarity of capabilities. The value of this type of capability variable is

numeral and describes the extent to which the capabilities of two organizations are

overlapping or related.

Figure 4 – Coding based on variable type.

0

20

40

60

80

100

120

Existence of a capability

Amount of a capability

Similar/ different capability

Degree of similarity in capabilities

73

Many of the reviewed articles used several types of operationalizations and were thus coded

under two or more classes relating to capability type, data type, and variable type. Hence the

coding operates over operationalization instances rather than articles.

3 RESULTS

3.1 Capability type – data type

Table 2 presents cross-tabulations for capability types and data types with the number of

articles in parentheses. Technological capabilities tend to be measured through patents and

scientific articles. Of the 174 articles in the sample 35 studied technological capabilities through

patents, scientific articles, or their citations. As is well known, patents capture a limited share

of a firm’s innovative capability (see e.g. Patel and Pavitt, 2000), while patents may also be filed

to give false signals to competitors. Some industries are more prone to patenting than others.

Furthermore, the studies using patent data seldom explicate whether it is the ability to produce

patents, the ability to produce superior products based on the patents, or a combination of the

two, that they are measuring. In addition, patents capture a very limited scope of capabilities.

On the bright side, patent data is easily available. Reitzig and Puranam (2009) have been

innovative in broadening the scope of capabilities measured through patent data by measuring

organizational capabilities through the ability to obtain rapid patent protection.

Table 2 – Cross-tabulation of capability type and data type.

Survey,

subject

ive

Survey,

objecti

ve

Patent

s

Experi

ence in

years

Innova

tive

actions

R&D

expen

diture

Educati

on

level

Inter

view

s

R&D

collabo

ration

Financial

data

Othe

r Sum

Technolo

gical

12%

(17)

8%

(12)

25%

(35)

3%

(4)

8%

(11)

11%

(15)

4%

(6)

11%

(15)

1%

(2)

4%

(6)

13%

(19)

100%

(142)

Organizat

ional

28%

(37)

8%

(11)

5%

(7)

5%

(7)

5%

(6)

5%

(6)

2%

(3)

18%

(24)

1%

(1)

12%

(16)

11%

(14)

100%

(132)

Market 22%

(9)

7%

(3)

10%

(4)

2%

(1)

17%

(7)

10%

(4)

0%

(0)

12%

(5)

2%

(1)

10%

(4)

7%

(3)

100%

(41)

74

Organizational capabilities are often measured through subjective surveys, experience in years,

and interviews. Subjective surveys often appear as the last resort in obtaining data on a

difficult to measure capability. One of their weaknesses is the limitedness of the respondents’

knowledge on the subject. For example, asking managers to rate their firm against the

competitors in how fast and broadly data on consumer satisfaction is circulated in the business

unit may be too challenging a task. In addition to their own operations the respondents should

be familiar with the situation industry-wide. However, surveys asking for subjective views

allow data collection on many types of capabilities and even though the data is not readily

available, with adequate resources the data collection task is doable. Similar strengths and

weaknesses apply to using interview data. Measuring capabilities through the number of years a

firm has been active in a given business is a straightforward solution and makes comparison

easy. Even though experience may signal the extent of the opportunity to develop capabilities,

it does not necessarily capture the realized capability accumulation.

Market capabilities are often measured through innovative actions, such as the number of new

product launches or the sales of newly launched products. Also subjective surveys have been

used for this purpose. However, a high number of new product launches may also be

interpreted as failures in prior product launches which induce a need for additional launches.

Thus it would be appropriate to exclude new products with low sales from such market

capability measures.

3.2 Variable type - data type

Table 3 presents the cross-tabulations for variable types and data types. Interviews have often

been used to determine the existence of a capability. Such studies tend to be qualitative and

aim at exploring the capabilities a firm possesses. However, the reliability of determining the

existence of a capability in this manner is questionable, because none of the reviewed studies

have come to the conclusion that a given capability does not exist on the basis of the interview

data.

75

Table 3 – Cross-tabulations based on variable type and data type.

Existence Amount Similar/different

Degree of similarity

Sum

Survey, subjective

19% (10)

62% (33)

13% (7)

9% (5)

100% (53)

Survey, objective 32% (9)

57% (16)

7% (2)

4% (1)

100% (28)

Patents & scientific articles

15% (7)

57% (27)

13% (6)

15% (7)

100% (47)

Experience in years

14% (1)

71% (5)

14% (1)

0% (0)

100% (7)

Innovative actions

24% (5)

62% (13)

14% (3)

0% (0)

100% (21)

R&D expenditure 13% (3)

87% (20)

0% (0)

0% (0)

100% (23)

Education level 0% (0)

100% (8)

0% (0)

0% (0)

100% (8)

Interviews 36% (16)

47% (21)

16% (7)

2% (1)

100% (45)

R&D collaboration

25% (1)

25% (1)

25% (1)

25% (1)

100% (4)

Financial data 35% (9)

38% (10)

15% (4)

12% (3)

100% (26)

Other 21% (7)

50% (17)

15% (5)

15% (5)

100% (34)

The amount of a capability a firm or a nation possesses has often been determined through

R&D expenditure and education level. Both measures are suitable for different aggregates, such

as clusters and industries. R&D expenditure fails to capture the outputs and the effectiveness of

the R&D processes (cf. Dutta et al., 2005). Similarly, education level may reflect the national

policies or those of firm recruitment rather than the actual capability stock. However, such

figures concerning different countries, and perhaps even concerning different firms, are easily

available and enable comparisons.

Whether capabilities are similar or different between two firms or the same firm at two

different points in time has been measured in many different ways and none have become

decisively popular. The degree of similarity of the capabilities of two firms has mostly been

determined through patents and their cross-citations. In this way it is possible to determine the

capability overlap or relatedness of two firms.

76

3.3 Variable type – capability type

Table 4 presents the cross-tabulations for variable types and capability types. Existence,

amount, and similarity of capabilities have been used more often than the degree of similarity.

It is not surprising that technological capabilities have often been studied through the degree

of similarity as patent classifications enable such measurements.

Table 4 – Cross-tabulations based on variable type and capability type.

Existence Amount Similar/different

Degree of similarity

Sum

Technological 18% (19)

54% (57)

15% (16)

13% (14)

100% (106)

Organizational 26% (27)

54% (57)

16% (17)

4% (4)

100% (105)

Market 18% (6)

56% (19)

18% (6)

9% (3)

100% (34)

4 CONCLUSIONS

The most popular operationalizations to date in capabilities research have been to measure

technological capabilities through patents, organizational capabilities through surveys,

interviews, and experience in years, and market capabilities through innovative actions.

Measuring the amount of a capability has been the dominant approach, even though similarity

and difference of capabilities might offer more interesting findings.

Based on these findings it appears that there is still a lot of room for imagination and

refinement in operationalizing capabilities. Moreover, the less frequently used

operationalizations may offer ideas for more innovative approaches. Multiple coders could be

used to improve the reliability of the findings.

77

REFERENCES

AÑÓN HIGÓN, D., BOZKURT, Ö., CLEGG, J., GRUGULIS, I., SALIS, S., VASILAKOS, N. &

WILLIAMS, A. M. 2009. The determinants of retail productivity: A critical review of the

evidence. International Journal of Management Reviews, 11.

ARCHIBUGI, D. & COCO, A. 2005. Measuring technological capabilities at the country level: A

survey and a menu for choice. Research Policy, 34, 175-194.

ARMSTRONG, C. E. & SHIMIZU, K. 2007. A review of approaches to empirical research on the

resource-based view of the firm. Journal of Management, 33, 959-986.

BHATTACHARYA, M. & GIBSON, D. E. 2005. The effects of flexibility in employee skills,

employee behaviors, and human resource practices on firm performance. Journal of

Management, 31, 622-640.

BIERLY, P. E., DAMANPOUR, F. & SANTORO, M. D. 2009. The Application of External

Knowledge: Organizational Conditions for Exploration and Exploitation. Journal of

Management Studies, 46, 481-509.

BOEKER, W. & WILTBANK, R. 2005. New venture evolution and managerial capabilities.

Organization Science, 16, 123-133.

BOUGRAIN, F. & HAUDEVILLE, B. 2002. Innovation, collaboration and SMEs internal research

capacities. Research Policy, 31, 735-747.

BRUSH, T. H. & ARTZ, K. W. 1999. Toward a contingent resource-based theory: The impact of

information asymmetry on the value of capabilities in veterinary medicine. Strategic


CAMPBELL-HUNT, C. 2000. What have we learned about generic competitive strategy? A

meta-analysis. Strategic Management Journal, 21, 127-154.

DALTON, D. R., DAILY, C. M., CERTO, S. T. & ROENGPITYA, R. 2003. Meta-analyses of

financial performance and equity: Fusion or confusion? Academy of Management

Journal, 46, 13-26.

DANIELS, P. L. 1997. National technology gaps and trade - An empirical study of the influence

of globalisation. Research Policy, 25, 1189-1207.

DAVID, R. J. & HAN, S. K. 2004. A systematic assessment of the empirical support for

transaction cost economics. Strategic Management Journal, 25, 39-58.

DEL CANTO, J. G. & GONZALEZ, I. S. 1999. A resource-based analysis of the factors

determining a firm's R&D activities. Research Policy, 28, 891-905.

DUTTA, S., NARASIMHAN, O. & RAJIV, S. 2005. Conceptualizing and measuring capabilities:

Methodology and empirical application. Strategic Management Journal, 26, 277-285.

78

EISENHARDT, K. M. & MARTIN, J. A. Year. Dynamic capabilities: What are they? In, 2000.

1105-1121.

FABER, J. & HESEN, A. B. 2004. Innovation capabilities of European nations - Cross-national

analyses of patents and sales of product innovations. Research Policy, 33, 193-207.

HAAS, M. R. & HANSEN, M. T. 2005. When using knowledge can hurt performance: The value

of organizational capabilities in a management consulting company. Strategic


HENDERSON, R. & COCKBURN, I. 1994. Measuring competence - Exploring firm effects in

pharmaceutical research. Strategic Management Journal, 15, 63-84.

KAPLAN, S. 2008. Cognition, capabilities, and incentives: Assessing firm response to the fiber-

optic revolution. Academy of Management Journal, 51, 672-695.

KARREN, R. J. & BARRINGER, M. W. 2002. A review and analysis of the policy-capturing

methodology in organizational research: Guidelines for research and practice.

Organizational Research Methods, 5, 337-361.

KETCHEN, D. J., COMBS, J. G., RUSSELL, C. J., SHOOK, C., DEAN, M. A., RUNGE, J., LOHRKE,

F. T., NAUMANN, S. E., HAPTONSTAHL, D. E., BAKER, R., BECKSTEIN, B. A.,

HANDLER, C., HONIG, H. & LAMOUREUX, S. 1997. Organizational configurations and

performance: A meta-analysis. Academy of Management Journal, 40, 223-240.

LOHRKE, F. T., HOLLOWAY, B. B., WOOLLEY, T. W. 2009. Conjoint Analysis in

Entrepreneurship Research A Review and Research Agenda. Organizational Research

Methods, Forthcoming.

MAHMOOD, I. P. & SINGH, J. 2003. Technological dynamism in Asia. Research Policy, 32, 1031-

1054.

MONTEALEGRE, R. 2002. A process model of capability development: Lessons from the

electronic commerce strategy at Bolsa de Valores de Guayaquil. Organization Science,

13, 514-531.

MOWERY, D. C., OXLEY, J. E. & SILVERMAN, B. S. 1998. Technological overlap and interfirm

cooperation: implications for the resource-based view of the firm. Research Policy, 27,

507-523.

NERKAR, A. & ROBERTS, P. W. 2004. Technological and product-market experience and the

success of new product introductions in the pharmaceutical industry. Strategic


NEWBERT, S. L. 2007. Empirical research on the resource-based view of the firm: An

assessment and suggestions for future research. Strategic Management Journal, 28, 121-

146.

79

PATEL, P. & PAVITT, K. 2000. How technological competencies help define the core (not the

boundaries) of the firm. In: DOSI, G., NELSON, R. R. & WINTER, S. G. (eds.) The nature

and dynamics of organizational capabilities Oxford: Oxford University Press.

PENNER-HAHN, J. & SHAVER, J. M. 2005. Does international research and development

increase patent output? An analysis of Japanese pharmaceutical firms. Strategic


QUINTANA-GARCIA, C. & BENAVIDES-VELASCO, C. A. 2008. Innovative competence,

exploration and exploitation: The influence of technological diversification. Research

Policy, 37, 492-507.

REITZIG, M. & PURANAM, P. 2009. Value appropriation as a organizational capability: The

case of IP protection through patents. Strategic Management Journal, 30, 765-789.

ROMIJN, H. & ALBALADEJO, M. 2002. Determinants of innovation capability in small

electronics and software firms in southeast England. Research Policy, 31, 1053-1067.

ROTHAERMEL, F. T. & BOEKER, W. 2008. Old technology meets new technology:

Complementarities, similarities, and alliance formation. Strategic Management Journal,

29, 47-77.

SCHROEDER, R. G., BATES, K. A. & JUNTTILA, M. A. 2002. A resource-based view of

manufacturing strategy and the relationship to manufacturing performance. Strategic


SONG, J. 2002. Firm capabilities and technology ladders: Sequential foreign direct investments

of Japanese electronics firms in East Asia. Strategic Management Journal, 23, 191-210.

TAN, D. & MAHONEY, J. T. 2006. Why a multinational firm chooses expatriates: Integrating

resource-based, agency and transaction costs perspectives. Journal of Management

Studies, 43, 457-484.



TRAORE, N. & ROSE, A. 2003. Determinants of biotechnology utilization by the Canadian

industry. Research Policy, 32, 1719-1735.

WANG, C. L. & AHMED, P. K. 2007. Dynamic capabilities: A review and research agenda.

International Journal of Management Reviews, 9, 31-51.

YEOH, P. L. & ROTH, K. 1999. An empirical analysis of sustained advantage in the US

pharmaceutical industry: Impact of firm resources and capabilities. Strategic


ZAHRA, S. A. & NIELSEN, A. P. 2002. Sources of capabilities, integration and technology

commercialization. Strategic Management Journal, 23, 377-398.

80

ZOLLO, M. & WINTER, S. G. 2002. Deliberate learning and the evolution of dynamic

capabilities. Organization Science, 13, 339-351.

ZWIJZE-KONING, K. & DE JONG, M. 2007. Evaluating the communication satisfaction

questionnaire as a communication audit tool. Management Communication Quarterly,

20, 261-282.

81

CHAPTER 5

CAPABILITY EVOLUTION IN THE FINNISH FOREST CLUSTER: A

QUALITATIVE APPROACH

ULRIIKKA TIKKANEN

Aalto University School of Science and Technology, Turku School of Economics [email protected]

1 INTRODUCTION

Discretionary differences exist within an industry and do matter as opposed to the view that

the differences between the firms are not discretionary but rather reflect the differences

between the contexts in which firms operate. It is often claimed that the competitive advantage

is determined by a combination of supply-side and demand-side factors. On the demand-side,

the productive activities of the firm must correspond to a market need. On the supply side, the

firm must have the capabilities not only to serve that market need, but to serve it more

efficiently than its competitors. Not only specialized knowledge can provide a basis for

sustainable advantage, because it resides in individuals; and because the rents generated by

specialized knowledge are more likely to be appropriated by individuals than by the firm. The

critical source of competitive advantage is knowledge integration (Nelson 1991, 61; Grant 1996,

380).

Strategy, structure and capabilities can be seen as defining a relatively stable firm character.

Strategy connotes a broad set of commitments made by a firm that define its objectives and the

ways it intends to pursue them: it can be written down or lay in the management culture of the

firm. Structure is defined as the way the activities in the firm are governed and organized: it

defines what the firm actually does, given the broad strategy. Both strategy and structure call

82

forth and mold core capabilities of the firm: the firm can be understood as a hierarchy of

organizational routines, which in turn are building blocks of the capabilities of the firm (Nelson

1991, 61, 67-68). Choices around how to capture value all help determine the architecture or

design of a business. Having a differentiated yet effective and efficient “strategic architecture”

to an enterprise´s business model is important (Teece 2007, 1330) Under competition, superior

profitability is likely to be more associated with resource and capability-based advantages than

with positioning advantages resulting from market and segment selection.

Distinctive organizational capabilities may bear their importance insofar as they can be shown

to shape persistently the destiny of individual firms – profitability, growth and probability of

survival. The relative competitive position of the firm is determined by how its resources and

capabilities are acquired, developed and deployed. A successful organization derives

competitive strength from its excellence in a small number of capability clusters where it can

sustain a leadership position over time (Dosi et al, 2003; Maritan, 2001; Grant 1996).

Also Ethiraj, Kale, Krishnan and Singh (2005, 28) note that organizational capabilities reflect

the evolutionary process of deliberate firm-specific investments along with the largely tacit

“learning-by-doing” in which firms engage. This, in turn, results in heterogeneity of firms and

the consequent differences in their performance. The process reflects the competence of

individuals and of the organizing principles by which relationships among individuals, groups,

and members in an industrial network are structured and coordinated. These principles of

coordination of individual and functional competence generate the capabilities of a firm. These

capabilities concern the ability to manufacture major industrial innovations (Zander & Kogut

1995).

1.1 The objective and the structure of the study

The purpose of the current study is to develop an understanding of the factors that have affected

the capability evolution in the Finnish forest cluster.

The second chapter introduces the concept of organizational capabilities. To understand the

main themes of this study, it is important to understand the evolution of the term “capabilities”

and the different aspects of the theme.

The third chapter focuses on data and methods used in this particular study. First, the

qualitative methods of research are discussed in general. Second, the chosen research method,

83

interviews, is analyzed. In the fourth sub-chapter, the methods of collecting data in this

particular study are described. Then, the interview outline in this particular study is introduced

and the different aspects of the interview outline are discussed. The last sub-chapter reports

the analysis-phase of the empirical process.

The fourth chapter concentrates on the empirical findings of this study. The data are divided

into capability-evolution related themes which have emerged from the data. Finally,

conclusions are made.

2 ORGANIZATIONAL CAPABILITIES

2.1.1 Capability definitions and evolution

There is still no exhaustive definition of organizational capabilities. Researchers emphasize

differing themes and capabilities could even be seen floating “in the literature like an iceberg in

a foggy Arctic sea, one iceberg among many, not easily recognized as different from several

icebergs near by” (Dosi et al, 2003). According to Winter´s commonly cited definition, “an

organizational capability is a higher-level routine (or a collection of routines) that, together

with its implementing input flows, confers upon an organization´s management a set of

decision options for producing significant outputs of a particular type” (Winter 2003, 991).

There is a broad consensus that capabilities derive from organizational routines: they involve

organized activity and they are typically exercised repetitiously. Routinization of organizational

activities embeds capabilities into organizational memory, engendering a unique configuration

of firm resources. However, it is argued that routines are not the only building blocks of

capabilities (Knight & Cavusgil 2004, 127; Dosi et. al. 2008, 1167). As Dosi et. al (2008, 1168) note,

capabilities involve organized activity, and the exercise of a capability is typically repetitious in

many cases and the individual skills are the building blocks of those routines. They derive the

idea that a successful large corporation derives competitive strength from its excellence in a

small number of capability clusters where it can sustain a leadership position over time.

Several authors propose different classifications of capabilities. For example, Loasby (1998)

defines capabilities as “the least definable kinds of productive resources” which make possible

different future activities. He distinguishes capabilities between direct and indirect capabilities.

The former are required for productive activities and the latter for transactions. Indirect

capabilities can be divided into two groups: gaining control of other capabilities or by obtaining

84

access to them. Both direct and indirect capabilities develop through specialization. Indirect

capabilities reduce the costs of particular transactions by embodying knowledge of particular

circumstances and particular relationships (Loasby 1998, 152). According to Day (1994),

capabilities are complex bundles of skills and collective learning, exercised through

organizational processes that ensure superior coordination of functional activities. He divides

capabilities into three categories: “Outside-In Processes” (e.g. market sensing and technology

monitoring), “Inside-Out Processes” (e.g. HRM and manufacturing) and “Spanning processes”

(e.g. pricing, purchasing and new product development).

During the last decade, the main stress of capability discussion has been transferred towards

dynamic capabilities-approach. For example, Helfat and Peteraf (2003, 999) classify capabilities

as either operational or dynamic. An operational capability generally involves performing an

activity, such as manufacturing a particular product, using a collection of routines to execute

and coordinate the variety of tasks required to perform the activity. Dynamic capabilities do

not involve production of a good or provision of a marketable service. Instead, dynamic

capabilities build, integrate, or reconfigure operational capabilities. Operational as well as

dynamic capabilities include two sets of routines: those which perform individual tasks and

those which coordinate these tasks. The need to coordinate tasks implies that a capability

involves coordinated effort by individuals. In turn, Winter (2003) divides capabilities into a

hierarchy. In his analogy, zero-level capabilities show “how we earn a living now”. Without

them, the firm could not collect the revenue from its customers that allow it to buy more

inputs and and do the whole process over again. By contrast, capabilities that would change the

product, the production process, the scale, or the markets served are not at the zero level. New

product development is a prototypical example of a first-order “dynamic capability”.

Teece (2007, 1319) argues that sustainable advantage requires more than the ownership of

difficult-to-replicate (knowledge) assets. It also requires unique and difficult-to-replicate

dynamic capabilities. These capabilities can be harnessed to continuously create, extend,

upgrade, protect, and keep relevant the enterprise´s unique asset base. Dynamic capabilities

can be disaggregated into the capacity 1) to sense and shape opportunities and threats, 2) to

seize opportunities, and 3) to maintain competitiveness through enhancing, combining,

protecting, and, when necessary, reconfiguring the business enterprise´s intangible and

tangible assets. Dynamic capabilities include capabilities required to adapt to changing

customer and technological opportunities. They also embrace the enterprise´s capacity to

shape the ecosystem it occupies, develop new products and processes, and design and

implement viable business models.

85

Nevertheless, scholars have differing views about the contents and formation of the concept.

Teece, Pisano and Shuen (1997), in their early framework on the agenda, define dynamic

capabilities as the outcome of the firm´s processes, positions and paths. The first include

questions of coordination/integration, learning and reconfiguration and transformation. The

second include technological, financial, reputational, structural, institutional and market assets

and organizational boundaries. The third are shaped by path dependencies and technological

opportunities. Augier and Teece (2006, 406-407) claim that dynamic capabilities have multiple

origins, some rooted in routinized behavior, some in asset selection/investment choices, and

others rooted in creative and differentiated entrepreneurial acts, which involve unusual skills

that are not particularly imitable. Teece´s framework indicates that the extent to which an

enterprise develops and employs superior (non-imitable) dynamic capabilities will determine

the nature and amount of intangible assets that it will create or assemble and the level of

economic profits it can earn in the future (Teece 2007). Zollo and Winter (2002, 340), and

Wang and Ahmed (2007, 35) in turn, concentrate on the behavioral orientation of the concept.

According to them, dynamic capabilities are learned and stable patterns of collective activity

through which the organization systematically generates and modifies its operating routines in

pursuit of improved effectiveness. The purpose of dynamic capabilities is to integrate,

reconfigure, renew and recreate its resources and, most importantly, upgrade and reconstruct

its core capabilities in response to the changing environment to attain and sustain competitive

advantage.

In fact, change appears to be one of the most central concepts in the most recent dynamic

capabilities research. Teece (2007, 1320) claims that dynamic capabilities include the capacity to

shape the ecosystem the enterprise occupies, to develop new products and processes, and to

design and implement viable business models. On the other hand, it has been argued that

while some capabilities may deal specifically with adaptation, learning, and change process, all

of them have the potential to accommodate change. However, learning, change and adaptation

do not necessarily require the intervention of dynamic capabilities as intermediaries. Change

could occur by force majeure from the environment. Such change behaviors do not depend on

dynamic capabilities: Winter proposes “ad hoc problem solving” for a term for such behavior. It

is not routine; thus, ad hoc problem solving and dynamic capabilities are two different ways to

change. Dynamic capabilities typically involve long-term commitments to specialized resources

(see Winter 2003; Helfat and Peteraf 2003, 998).

Maintaining superior performance ultimately requires the continual renewal of competitive

advantages through innovation and the development of new capabilities. Grant (1996) claims

86

that there are two dimensions to such renewal: extending existing capabilities to encompass

additional types of knowledge and reconfiguring existing knowledge into new types of

capabilities.

Continuous innovation in competitive environments tends to be the result of the deployment

and extension of a continuing core of capabilities rather than the constant creation of new

capabilities. Achieving flexible integration, either through continually integrating new tacit

knowledge or through constantly reconfiguring existing knowledge, is likely to impose

substantial costs in terms of reducing the efficiency of knowledge integration. Capabilities also

change over time, as the company builds on its core capabilities and absorptive capacity to

identify the new market opportunities, and to develop the new project- and functional

capabilities required to satisfy new commercial objectives, and to introduce the organizational

changes to meet increasing demand (Grant 1996; Davies & Brady 2000).

2.1.2 Capabilities in mature industries

As forest sector companies operate in a relatively mature industry, it makes sense to investigate

previous research relating to the capabilities on other industries which have reached their

maturity stage. As Mitchell (1994, 575) notes, organizational capabilities are retained within a

product market as firms age and grow. Capabilities change over time, as the company builds on

its core capabilities and absorptive capacity to identify the new market opportunities, develop

the new project and functional capabilities required to satisfy new commercial objectives, and

introduce the organizational changes to meet increasing demand (Davies & Brady 2000, 948-

949). Relative to new enterprises, established firms often possess a wider array of resources and

capabilities that they can leverage in additional markets. By virtue of size and longevity,

established firms may hold larger and more developed stocks of individual resources and

capabilities (Helfat & Lieberman 2002, 734).

The science and technology policy discourse has been leaning toward a more innovation-

centered approach with an increasing emphasis on the development of technologies and the

application of knowledge at the expense of the creation of knowledge. According to Knight and

Cavusgil (2004), the aspects of capabilities relate to the shifting character of the business

environment which leads to appropriately adapting, integrating and re-configuring knowledge

based capabilities toward the changing environment.

87

Kogut and Zander (1992) introduce the concept of combinative capability, which refers to

generating new applications from existing knowledge (e.g. innovations). By combinative

capabilities, they mean the intersection of the capability of the firm to exploit its knowledge

and the unexplored potential of the technology. They claim that switching to new capabilities

is difficult, as neither the knowledge embedded in the current relationships and principles is

well understood, nor the social fabric required to support the new learning known. It is the

stability of these relationships that generates the characteristics of inertia in the capabilities of

an organization. Van den Bosch, Volbreda and de Boer (1999, 556-557) extend this view by

distinguishing three types of combinative capabilities a firm has at its disposal: systems

capabilities, coordination capabilities and socialization capabilities. Systems capabilities are

used to integrate explicit knowledge: they provide a memory for handling routine situations.

Coordination capabilities enhance knowledge absorption through relations between members

of a group. Socialization capabilities may influence absorptive capacity by specifying broad,

tacitly understood rules for appropriate action under unspecified contingencies. Socialization

capabilities are found in firms with a strong identity. Here, one can find a coherent set of

beliefs, a high degree of shared values, a common language, and a strongly agreed-upon kind of

appropriate behavior.

New product and process development projects are obvious, visible arenas for conflict between

the need for innovation and retention of important capabilities (Leonard-Barton 1992). Klepper

(2002) claims that the individual capabilities and the experience of the founder in related

industries is closely tied to the performance of the firm in the automobile industry. His

research of diversifying and de novo entrants to the automobile industry showed that prior

experience imparted competitive advantage. Even though diversifying entrants from related

industries (e.g. bicycles) were compelled to learn new techniques of manufacturing and

structure the organization more precisely, technological change often led to new innovations.

As Teece (2007, 1328) notes, the existence of established assets and routines exacerbates

problems of excessive risk aversion. Specifically, both the isolation effect and the certainty

effect can be intensified by the existence of established assets, causing incumbent enterprises

to become comparably more risk averse than new entrants. In terms of innovative activity, this

excessive risk aversion leads to biased decision making and limits the probability that

incumbent enterprises will explore risky radical innovations. He also argues (2007, 1335) that a

key to sustained profitable growth is the ability to recombine and to reconfigure assets and

organizational structures as the enterprise grows, and as markets and technologies change.

Incumbent enterprises possessing fixed assets may further tend to limit their new investments

88

to innovations that are “close-in” to the existing asset base. They tend to narrowly focus search

activities to exploit established technological and organizational assets. This effect makes it

difficult for these enterprises to see potential radical innovations. Incumbent enterprises tend

to frame new problems in a manner consistent with the enterprise´s current knowledge base,

assets, and/or established problem-solving heuristics and established business model. This

means that managers may not successfully address opportunities or potential innovations even

when they do recognize them.

2.1.3 Forest industry

Traditionally, the technological competencies and know-how have been the leading triggers of

development in the forest industry. As Tremblay (1999, 798) notes, technological capabilities

embody the resources required to manage and actualize the generation of technical change.

These resources are accumulated and embodied in people (skills, knowledge and experience)

and organizational systems. However, there has been much evidence that the forest sector has

been realizing that competitive advantage rests on more than just technological excellence.

In empirical studies of the forest industry, both tangible and intangible firm resources have

been found to have an important role in creating value-added, enhancing competitiveness, and

achieving success in a modern business environment. To become and stay competitive,

industries must develop and update their technological knowledge bases much quicker than

ever before. Survival in the increasingly tightening market competition requires strategic

decision-making and the constant development of business and manufacturing processes as

well a more innovation-centered approach.

However, break-through innovations are not usually the most important driving factors of

competitive position and performance of the firms in the pulp and paper industry. It seems that

the firms that succeed in implementing even small gains in productivity on year-to-year basis

via investments in upgrading and modernization, as well as making changes to the supply-

chain, would in the medium-to-longer run gain competitive advantage relative to those firms

who were not successful in implementing such strategies (Ghosal & Nair-Reichert 2008).

Process innovations in the forest product industry seem to occur in improvements in raw

material utilization, computer-aided manufacturing and customized machinery, for example.

Also these categories listed under innovations tend to fall on incremental improvements to

enhance efficiency. For example, R&D efforts in the pulp and paper industry are not focused on

89

radical innovations. They are instead oriented towards improving the quality and features of

existing products and improving manufacturing processes and acquisition of embodied

technology available in international markets instead of on the development of in-house

technology (see Diaz-Balteiro et. al. 2006; Hovgaard & Hansen 2004; Tremblay 1999).

As Lähtinen and Toppinen (2008) note in their investigation on Finnish sawmills, companies

are not alike even in a resource-dependent, mature industry operating within the same

business environment. The internal characteristics of firms differ from each other and cause

variation in the company-wise financial performance. The core capabilities of the firm

influence the outcome of wood product innovations (Bull & Ferguson 2006). The presence of

appropriate technology, governance structure and firm wide learning culture will increase the

likelihood that the product will achieve success.

3 DATA & METHODS

3.1 Qualitative research methods

Qualitative research genres have become increasingly important modes of inquiry for the social

sciences and applied fields. Qualitative research is pragmatic, interpretive and grounded in the

lived experiences of people. These interests take qualitative researchers into natural settings,

and foster pragmatism in using multiple methods for exploring a topic. Hence, qualitative

research is a broad approach to the study of social phenomena (Marshall & Rossman 2006, 1-2).

The processes and phenomena of the world are described before theorized, understood before

explained, and seen as concrete qualities before abstract quantities. The qualitative stance

involves focusing on the cultural, and situated aspects of human thinking, learning, knowing,

acting, and it is opposed to “technified” approaches to the study of human lives (Kvale &

Brinkmann 2009, 12).

According to Malhotra and Birks (2007, 152) qualitative research can be defined as an

unstructured, primarily exploratory design based on small samples, intended to provide insight

and understanding. Qualitative research is a mixture of the rational, explorative and intuitive,

where the skills and experiences of the researcher play an important role in the analysis of data.

Qualitative research is often focused on social process and not on social structures, which is

often the case in quantitative research (Ghauri et al. 1995, 84). Typical examples requiring

qualitative research are research problems focusing on uncovering a person’s experience or

90

behavior, or trying to uncover and understand a phenomenon about which very little is known

(Ghauri et al. 1995, 85), like is the case in this research.

The aim of qualitative research is to illustrate real life, dealing with quality and meanings rather

than absolute quantity. The idea is that reality is incoherent, and therefore reality cannot be

dealt in arbitrary ways, but as clearly as possible (Hirsjärvi et al. 2003). A foundation for

qualitative research is to describe real life by understanding the research subject

comprehensively (Hirsjärvi, Remes & Sajavaara 2000, 152). Also, not having hypotheses is

characteristic to qualitative research (Eskola – Suoranta 2003, 19–20). In order to get a

comprehensive understanding on the study subject, a qualitative approach to this research

would be justified (see Zalan – Lewis 2004, 510). Moreover, the author’s aim is to find new

insights and to learn about this subject, thus supporting the choice of qualitative research

approach (see Eskola – Suoranta 2003, 19–20).

Basing one’s decisions only on a qualitative approach is not sufficient for carrying out a viable

research. One should in addition decide on the research strategy. Five major research strategies

are experiments, surveys, archival analysis, histories and case studies. This poses a challenge to

the author, as it is important to choose a strategy that would be convenient to carry out, as well

as that would be generating accurate research information (Yin 1985). Every type of empirical

research has an implicit, if not explicit, research design. The design is the logical sequence that

connects the empirical data to a study´s initial research questions and, ultimately, to its

conclusions (Yin 2003, 20). In qualitative research, conformability is seen as the criteria for

neutral approach. This can be reached when the researcher has conformed with various

techniques that the research is trustworthy and applicable to other scenarios (see Lincoln &

Guba 1985). A well and carefully done research design makes it possible for a reader to follow

the flow of the research and estimate its trustworthiness.

When design requirements have been specified, decisions must be made on how requirements

should be met and information collected. The type of required primary data depends upon the

research problem and research design. The normal choices of primary data collection in

qualitative research are observations, surveys and interviews in addition to other written or

visual material (Ghauri & Gronhaug 2002; Marshall & Rossman 2006). Practical reasons often

justify the choice of an interview. An interview can be more informative, less time consuming

and leave out other methods. Koskinen, Alasuutari and Peltonen (2005) also note that it is easy

to be misguided when observing a certain phenomenon if one does not confirm his or her

91

findings by interviewing. The selection of how to conduct the interview and who should be

interviewed, have a great importance in selecting the case for the research.

3.2 Interview as a data collection method

According to Kvale & Brinkmann (2009), the qualitative research interview attempts to

understand the world from the subjects´ point of view, to unfold the meaning of their

experiences, and to uncover their lived world prior to scientific explanations. The research

interview is based on conversations of daily life and is a professional conversation: it has a

purpose and it involves a specific approach and technique. An interview is a conversation that

has a structure and a purpose, and it is also an active process where interviewer and

interviewee produce knowledge through their relationship. The production of data goes

beyond a mechanical following of rules and rests upon the interviewers´ skills and situated

personal judgment in the posing of questions. Knowledge of the topic of the interview is in

particular required for the art of posing second questions when following up interviewee´s

answers (Kvale & Brinkmann 2009, 82). According to Stone (1978, 67) the major difference

between data collection via the questionnaire and the interview is that in the case of the former

technique the respondent reads the questions and records his responses to the questions, while

in the case of the latter method the interviewer both presents the questions to the subject and

records the elicited responses and thus they can be considered as alternative data collection

tools.

Interview reports have tended to evoke rather standardized objections about their quality from

the mainstream of modern social science (Kvale & Brinkmann 2009). Some claim that

interviews are not objective, but subjective. Still, the objectivity of the interview has to be

discussed specifically for each of the multiple meanings of objectivity, as relevant to the

interview inquiry in question. It is also often argued that interviews are too person dependent –

flexible context sensitive and dependent on the personal interrelationships. Furthermore,

interviews are claimed to be not trustworthy – unacknowledged bias may entirely invalidate the

results of an interview inquiry. A recognized bias or subjective perspective, may, however,

come to highlight specific aspects of phenomena investigated and bring new dimensions

forward. It is also claimed the interview findings are not generalizable because of too few

subjects. Still, in postmodern conceptions of social sciences the goal of global generalization is

replaced by a transferability of knowledge from one situation to another, taking into account

the contextuality and heterogeneity of social knowledge (Kvale & Brinkmann 2009, 169-171).

92

According to an epistemology that takes as its starting point the elimination of human

subjectivity in research, the qualitative interview based on interpersonal interaction is

unscientific. Although no single authoritative definition of science exists, according to which

the interview can be unequivocally be categorized as scientific or unscientific. Interview data

consist of meaningful statements based on interpretations; they are thus not strictly separated.

Interview statements can be ambigious and contradictory and the findings may not be

intersubjectively reproducible. The open structure of research interviewing is an asset as well as

a problem field in interview investigations: no standard rules or procedures exist for

conducting an interview (Kvale & Brinkmann 2009).

The key questions when planning an interview investigation are:

why: clarifying the purpose of the study

what: obtaining pre-knowledge of the subject matter to be investigated

how: becoming familiar with different techniques of interviewing and analyzing, and

deciding which to apply in order to obtain the intended knowledge.

In general, there are three different types of interviews classified by their prescriptive.

Structured interview refers normally to survey research interviews, where the researcher sets

the questions and the running order of them and offers even answer possibilities. A semi-

structured interview allows the interviewee more freedom. Answering in own words, deflection

from the question outline and even proposal of own questions are made possible for the

interviewee. An unstructured interview strives to minimize the influence of the researcher on

the interview. At clearest form, the interviewer has only a general interest that he/she wants to

talk about with the interviewee (Koskinen et al. 2005). In the course of an interview, the

subjects can change their descriptions of, and attitudes toward, a theme. They may discover

new aspects of the themes they are describing, and suddenly see relations that they had not

been aware of earlier (Kvale & Brinkmann 2009, 31).

In an interview, the aim is to collect information in a methodically determined way. Interviews

are conducted under circumstances set by the interviewer (Hirsjärvi & Hurme 1995). While

interviewing, the interviewer can make observations and control the interview situation and

environment. The interviewer can note possible disturbing effects during the interview or

perceive unwillingness of the interviewee that affect on the validity of the research (Malhotra &

Birks 2006). As Kvale and Brinkmann (2009, 164) note, the quality of an interview is decisive for

93

the quality of the subsequent analysis, verification, and reporting of the interview findings. It

can be monitored through six quality criteria:

the extent of spontaneous, rich, specific and relevant answers from the interviewee

the extent of short interviewer questions and longer interviewee answers

the degree to which the interviewer follows up and clarifies the meanings of the

relevant aspects of the answers

to a large extent, the interview being interpreted throughout the interview

the interviewer attempting to verify his or her interpretations of the subject´s answers

over the course of the interview

the interview being “self-reported”, a self-reliant story that hardly requires additional

explanations.

Due to the exploratory nature of this research, an interview is considered relevant: getting

information from a specialist who has deep knowledge on the industry puts in advance the

most relevant issues to be further developed. An exploratory interview is usually open, with

little preplanned structure. The interviewer introduces an issue, an area to be charted, or a

problem to be uncovered, then follows up on the subject´s answers and seeks new information

about and new angles on the topic (Kvale & Brinkmann 2009, 105-106). Due to the availability

of the interviewer, interviewing allows open-ended questions and the interviewer can ensure

that the complex instructions or sequences can be adhered to (Brewerton & Millward 2001, 74)

unlike when colleting data via questionnaires. This was the case in the course of this research

as well.

3.3 Data

3.3.1 Data collection

The data were collected using semi-structured interviews arranged along themes to provide a

wide scope for the data collection. Thematizing an interview study involves clarifying the

theme of the study – developing a conceptual and theoretical understanding of the phenomena

to be investigated in order to establish the base to which new knowledge will be added and

94

integrated (Kvale & Brinkmann 2009, 106). The interviewee can thus answer in his own words

even though the questions have been pre-formulated (see Eskola & Suoranta 2003, 86).

Presenting the questions is not made too exact not to distort the interview and the data; the

ultimate ends of the questions are tried to keep unknown.

The ability to ask open-ended questions is one of very important features in conducting semi-

structured interviews. An open-ended question encourages a full meaningful answer by use of

the subject's own knowledge and understanding. Unlike closed-ended questions whereby the

questions encourage short and or single-word answer. Open-ended questions also tend to be

more objective and less leading than closed-ended questions. Open ended question are

typically begin with words such as "Why" and "How", or phrases such as "Tell me about”. Open–

ended questions are basically statements which implicitly demand for responses (Reis & Judd

2000, 237) in a bid to get the interviewees tell more about their understanding and knowledge

in the field. According to Yin’s categorization, the open-ended nature of the interview

questions is realized by asking for the facts of a matter as well as asking what his/hers opinions

and emphases on that matter are. Thus, actually the interviewee is rather an informant than a

respondent. If he/she also proposes corroboratory evidence, they will be of critical value to this

study (see Yin 1985).

After deciding on the data collection method, the decision from whom the data should be

collected was considered. To cover the different aspects of the industry as comprehensively as

possible, a matrix based on the significant aspects of the value-chain was created. Six

product/service areas were identified: wood supply; chemicals; machinery; pulp, paper and

traditional wood products; new products (biofuel, wood plastic composite etc.) and consulting

and research. During the interview process, one additional area was considered significant

enough to be added to the interview matrix, namely environmental issues. The possible

informants were also differentiated by six functional areas they engage in, mainly operations,

logistics and current technologies; R&D and future technologies; strategic management and

corporate-level capability development; HR, knowledge management and personnel

development; marketing and sales; and finance.

95

Operations,

logistics and

current

technologies

R&D and

future

technologies

Strategic

management

and corporate-

level capability

development

HR, knowledge

management,

and personnel

development

Marketing and

sales

Finance

Wood Supply

Firm examples:

Metsäliitto

Chemicals

Firm examples:

Kemira

Machinery

Firm examples:

Metso, Andritz,

Tamfelt

Pulp, paper, and

traditional wood

products

Firm examples:

UPM, M-Real,

Metsä-Botnia,

Myllykoski

New products

Firm examples:

Stora Enso, UPM,

VTT

Consulting and

research

Firm examples:

Pyöry, VTT

Figure 1 – Interview matrix

Interviewees were selected according to the recommendations by the ForestCluster Ltd., and

contacted via e-mail. Interviewees represented a wide organizational membership: the

96

positions of informants ranged from senior vice-president level to the managers of product

marketing, directors of sales and business controllers as detailed in table 1.

Table 1 – Interviewee positions

Position

Business Controller

1

Chief Financial Officer

1

Chief Technology Officer

2

Director, Product Development

1

Development Manager, Wood Supply

1

Executive Vice President, Business Development

1

Manager, Pulp Chemistry, R&D and Technology

1

Product Marketing Manager

1

Professor

1

Sales Director

2

Senior Vice President

6

Environmental Affairs and Corporate Responsibility (1)

Human Resources and Total Quality Management (1)

Printing & Writing Paper (1)

Production (1)

Technology (1)

Sales and Marketing (1)

Vice President

12

R&D (2)

Environment (1)

Business Development (1)

Human Resources (1)

Operations, Plywood Business (1)

Pulp Mill Systems, Technology (1)

Sales (2)

Technology (3)

Total

30

97

During the time period from September 2009 to April 2010, a total of 30 interviews were

conducted. Most of the interviews took place in the premises of the organization the

interviewee represented, excepting one interview which was conducted through telephone. In

most of the interviews, two interviewers were present. The duration of the interviews ranged

between 40 minutes and one and a half an hour, the average being approximately one hour.

The interviews were conducted in Finnish because both the interviewers and the interviewees

spoke Finnish fluently. Hence, the quotations represented in following sections are translated

from Finnish to English.

In the beginning of the interview, the interviewees were explained the purpose of the research.

They were also asked if the interviews could be audio-recorded and notes could be taken.

However, the issue of confidentiality was emphasized. All except one of the interviews were

audio-recorded. Detailed notes of the non-recorded interview were taken during the interview

and then transcribed as soon as possible after the interview. The interviewees were also

explained some key concepts related to the interview questions in order to make sure that they

understood these concepts same way than the interviewer did. The actual interview questions

asked related to the background of the interviewee and their experience in the industry. After

these introductory questions, the actual focus of this research was introduced.

After the interview, the transcription was sent to the informants via e-mail for further remarks

and corrections. Besides, a draft of preliminary results was sent to the interviewees in April 2010

for prospective comments.

3.3.2 Interview outline

A central instrument that guides the interview is a clear outline. The two principal functions

are to make sure the interviewee presents the required questions and to let the interview flow

as naturally as possible. Researchers argue that the outline works as a memory tool that helps

to realize what has already been covered and what still needs to be emphasized (Koskinen et al.

2005).

An interview outline was formed based on the review of the literature and several discussions

between the interviewing team. The interview outline was divided in three themes:

Exploitation. It refers to the capabilities which are related to the incremental

improvements. Exploitation actions are considered important because they provide the

basis for more radical innovations.

98

Exploration. It refers to the capabilities which require actions of exploring the

environment for new ways of doing things.

Transformation. It refers to the capabilities necessary to redesign the processes and

structures in the organization.

Under these three themes, eleven possible capability-areas were differentiated as introduced in

the following figure.

Figure 2 – Interview outline

The interviewer suggested the informant either a) to select the capability-areas which the

informant considered being significant strengths or weaknesses according to his/her

experience, or b) to go through the eleven capability-areas point by point. The informants were

asked to speak as the representatives of the collective (ForestCluster Ltd.) but it was kept in

mind that naturally the backgrounds and different organizations affected the opinions of the

interviewees. Hence, the informants were urged to take the role of the spokesperson for their

organization as well. Interviewees were encouraged to describe the matters of concern by their

own words, and the follow-up questions were made to clarify the meanings of the relevant

aspects of answers. Interviewers gave the respondent also an opportunity to emphasize themes

which were not included in the interview outline. The interview outline was reviewed along the

interviewing process, even though no changes of large scale were considered important. Some

of the industry experts interviewed were provided some additional information and

instructions, in order to ensure that they understood the questions as the interviewer meant

them.

99

3.3.3 Data analysis

The interpretation of qualitative data may be seen as the most difficult task in a case study. A

challenge is that data analysis must be connected to data collection during the whole research.

Consequently theory develops as the data is gathered (Ghauri 2004, 117). The process of data

analysis includes clarifying and understanding text data, preparing the data for interpretation

(Creswell 2003, 190). Generic steps may be followed to analyze the data (Creswell 2003, 191–192):

(1) Organizing and preparing the data for analysis, (2) Obtaining a general sense of the data by

reading it through, (3) Analyzing the data in a more detailed manner by coding it, (4) Using the

coding process to construct a description of the phenomenon and themes for analysis, (5)

Advancing how the description and themes will be represented in the research text, and (6)

Making an interpretation on the data. Analysis of data does not take place in a single stage after

collecting the data. It is a continuous, systematic process which runs simultaneously with data

collection. In analyzing qualitative data, the aim is to produce new information by clarifying

the data with the purpose of condensing the data without losing information. In other words, it

aims at increasing the informative value of data by clarifying and simplifying scattered data.

Data analysis is the process of bringing order, structure and meaning to the mass of

unstructured data. While undertaking fieldwork one should be searching for common themes

start coding or develop some early concepts (Daymon & Holloway 2002; Ghauri & Gronhaug

2002).

According to Eskola and Suoranta (2003), there is three ways of analyzing interviews. They can

be resolved and directly analyzed by trusting the intuition of the researcher. It is also possible

to resolve the data and use coding. The third way is to combine the above with resolving and

coding that leads the researcher to data analysis. Interviewer may learn throughout an

investigation. The conversations with the subjects can extend and alter his or her

understanding of the phenomena. The interviewees bring forth new and unexpected aspects of

the phenomena studied; and during analysis of the transcribed interviews new distinctions may

be discovered (Kvale & Brinkmann 2009, 112).

The researchers first transcribed the interview word for word for getting a tool for analysis,

except for one interviewee who wished her interview not to be transcribed. It is easier to have a

glance over the entity of the interview, and thus all the material will be handled, not forgetting

anything of importance from the analysis. Second, the transcriptions were read through for

obtaining a general sense of what it involves.

100

Then, the interviews were reread thoroughly several times, simultaneously marking the

similarities and differences among the words, phrases, terms, and labels the informants (see

Corley, Gioia & Nag 2007; Corley & Gioia 2004). After reading through the transcriptions, a

computer-based analysis program, Atlas.ti, was utilized to support and systematize the data

coding. It enables the recording and cross-referencing the codes emerging from the data. With

the support of the program, there was an effort to discern the codes that were similar in their

essence in order to discern first-order categories. According to Corley and Gioia (2007), first-

order categories employ language expressing similar ideas among the informants. Then, the

coding of interviews was continued in this manner until no more distinct conceptual patterns

shared by the informants could be ascertained.

Following the analysis strategy introduced by Corley, Gioia and Nag (2007); and Corley and

Gioia (2004), the development of the first-order themes was followed by discerning

relationships among these categories. These links enabled clustering them into second-order

themes, which refer to researcher-induced concepts in a more abstract level. The second-order

themes were then collated into overarching dimensions in order to finish a framework linking

the issues emerging from the data.

In the following chapter, the results of the data analysis are represented by introducing one

dimension at a time. The figures in the beginning of the chapters describe the dimensions,

second-order themes, and first-order categories that are discussed in the sub-chapters.

3.3.4 Trustworthiness of the study

The issue of trustworthiness is for the researcher to persuade his or her audiences that the

findings of the study are worth taking into account of and paying attention to. To enable this

persuasion, researchers have conventionally found it practical to consider the following four

issues: truth value, applicability, consistency and neutrality of study (Lincoln & Guba 1985).

According to Ghauri et al. (1995, 95) data which cannot be statistically analyzed and are

difficult to measure in numbers are often called qualitative; such as strong, weak, difficult and

easy. One main problem of analyzing qualitative data is that, on one hand the number of

observations is so low and, on the other hand, the information on the case(s) is so in-depth that

it is very easy for the researcher to be drawn into the sheer volume of cases. With qualitative

methods the analysis is also difficult because data collection and analysis are often done

simultaneously, and sometimes the research problem is even formulated or reformulated at the

same time (Ghauri et al. 1995, 95- 96).

101

Issues of reliability and validity go beyond technical or conceptual concerns and raise

epistemological questions about the objectivity of knowledge produced through interview

research. In principle, a well-crafted interview can be an objective research method in the sense

of being unbiased. Contrary to common opinion, knowledge produced in interviews need not

be subjective, but qualitative interviews may, in principle, be an objective mode of inquiry with

respect to several key meanings of objectivity. Reliability pertains to the consistency and

trustworthiness of research findings; it is often treated in relation to the issue of whether a

finding is reproducible at other times and by other researchers. This concerns whether the

subjects will change their answers during an interview and whether they will give different

replies to different interviewers (Kvale & Brinkmann 2009).

Validation comes to depend on the quality of craftsmanship during an investigation, on

continually checking, questioning and theoretically interpreting the findings. The validity of an

investigation rests upon the soundness of the theoretical presuppositions of a study and upon

the logic of the derivations from theory to the research questions of the study. The validity of

knowledge produced involves the adequacy of the design and the methods used. The validity of

interviewing pertains to the trustworthiness and the quality of interviewing – careful

questioning to the meaning what is said and a continual checking of the information obtained

(Kvale & Brinkmann 2009).

The objective of testing reliability is to minimize the errors and biases in a study. The general

way of approaching the reliability problem is to make as many steps as operational as possible

(Yin 2003, 37-38). The responsibility of the researcher is to provide a data base that makes

transferability judgments possible for potential appliers. The purpose of transferability is to

provide proper description of the methodological choices and case selection in order to enable

possible transfer (Lincoln & Guba 1985). Lincoln and Guba (1985) emphasize that researchers

cannot achieve complete transferability in qualitative studies, but the data should be described

in such detail that the potential appliers are able to judge the transferability themselves. Thus,

for a research to be transferable, the data must be collected in a transferable way, and the

selection of research focus must be explicit. Kirk and Miller (1985, 22) refer to transparency as

“apparent validity” and it implies, that the reader of a qualitative research report is able to see

the basic processes of data collection (Rubin & Rubin 1995, 85). A transparent report allows the

reader to assess the intellectual strengths and weaknesses, the biases, and the

conscientiousness of the interviewer (Rubin & Rubin 1995, 85). In the course of this research,

the transparency has been achieved through thorough reporting the data collection process in

detail (see previous chapters), so that readers are able to evaluate how the research was

102

conducted. In order to able easy access to the data collected, it has been saved in the form of

recordings, and transcripts as well.

Several steps were taken to ensure the trustworthiness of the data. The reliability of this

research was increased mainly via the reliability of interview schedules. Before the interviews,

the respondents were explained the purpose of this research as well as the interview procedure.

They were notified about the audio-recorders and about the notes taken. However, it was

emphasized that the interviewees were not to be identified in the final results. The anonymity

to some extent even increases the reliability, as the interviewees may feel more confident

commenting on issues anonymously.

During the interviewing process, regular meetings were arranged between the industry experts

and the research group, in which the industry experts were shown drafts and preliminary

results of the interview study to comment on. The preliminary results of the study were also

represented in two seminars arranged to enable correspondence between research groups and

the representatives of ForestCluster Ltd., and to serve as a sounding board for propositions, and

to solicit questions about the data collection and analysis procedures.

103

4 EMPIRICAL RESULTS

4.1 Capability selection

The first overarching dimension of the capability evolution process assembled according to the

comments made by the interviewees was capability selection. The aim of the following chapter

is to introduce the factors that the informants have considered important for the capability

selection process in the Finnish forest cluster.

Figure 3 – Findings on capability selection

4.1.1 History

According to the industry experts interviewed, there seems to be a broad consensus that

history has affected the capability selection in the Finnish forest industry. Still, the effects on

the capability selection are both positive and negative.

Industry experts considered the evolution of the industry one of the most significant factors

affecting the selection of certain capabilities. The industry development was first led by buying

licenses from the North American companies where the top knowledge laid after the Second

World War. The informants noted that the knowledge acquired that way was efficiently

utilized, and generated a basis for future technology development. The strong growth of

consumption of printed media from the 1960s created a growing market of different paper

104

products. Consumption of paper and cardboard products was incrementally increasing every

year. The enormous technological development progress from the 1970s to the 2000s was

gradually based on the profitability and technological system development. In the former

decades, it was typical to develop new products and innovations from the viewpoint that they

could lead to improvements in functional aspects. Hence, the informants suggested that it was

logical to concentrate on the capabilities which are related to development of technology and

processes.

The co-operation between Finnish forest industry firms seems to have been more intense in the

former decades. The joint research and development facilities enabled co-operative ventures

and innovations which have been introduced successfully. The establishment of joint sales and

marketing was mentioned as one of the distinctive features of the past decades. Interviewees

suggested that the joint marketing co-operatives were of high importance for the profitability,

and provided forest companies a strong position abroad. For example, it was noted that “Finns

have been largely the leaders of the technological area, and that has been based to the fact that

these clusters have been active inside the companies operating in multiple industries, and the

cooperation has been easier to understand.”

As ”the main idea was that not every firm should build their own sales channels – that we do

certain areas together and join forces, and that was a very smart move, and it kind of forced us

to co-operate”, these co-operative actions also deepened the networking capabilities between

domestic firms. The networks encouraged the firms to develop capabilities relating to reactive

adaptation, described by one informant as “rapid and flexible reactions” to market situations as

well.

However, also differing views were represented. Sales and marketing being shared, companies

had to face severe competition in other areas. “Marketing was co-operative. What was left,

then; it did not make sense to try to do anything else than compete in the areas of building

even more factories and negotiating the finances needed. It was quite primitive.” Some

informants regarded them leading into alienation between companies and their customers

which, in turn, resulted in the situation in which the capabilities related to customer

relationships were hindered. As one interviewee indicated, ”sales companies and competitors

thought that discourse with customers was unnecessary – it has left behind a culture in which

the communication with customers was not seen valuable.”

The environmental conditions in Finland, especially from the viewpoint of the forest industry

might have been among the most challenging in the world but also a trigger for development.

105

For example, logistics have been a very challenging task, and adapting to the Finnish

environmental conditions around the year have been the trigger into shaping capabilities

related to e.g. logistics. As an interviewee suggested, “When we have had to run this business

all year round, in every circumstance – we have improved here, or at least have been able to

adjust into these circumstantial factors.”

4.1.2 Industry logic

The industry experts interviewed considered forest industry still very conservative. The product

life-cycles have been long and the new product innovations fairly infrequent compared to other

industries. As one interviewee from pulp and paper industry noted, ”if you compare paper and

cardboard industry to tele communications, the product life-cycles tend to be about 52 weeks

[in tele communications industry]. And here [in paper industry], the product life-cycles are

approximately 100 years. New product innovations, they are relatively unusual.” Hence, the

commercialization capabilities have not been selected into development, as there has not been

a constant flow of new products. The conservativeness of the industry is also visible in the

attitudes towards change. As two informants from the paper industry indicated, ”we prefer to

do things the way it has always been done, the same products”; and “we are a paper producer,

we make newsprint, and it will be similar until the end.”

The commercialization of new products has not been that straight-forward without verification

of the credibility and proven savings. Hence, the conservative nature of the industry has

hindered capabilities of new product development and commercialization. The capabilities

selected into development have been mostly technological and related to manufacturing and

improving efficiency. As stated, “The way of thinking has been production-inspired, that is we

have developed technologies to improve efficiency.”; and “the operations models and processes

which have been pitched to the utmost.”

The bulk products seem to be the main source of core business but the profits of producing

them in Finland are ever declining because of the high costs of raw material and labor. Hence,

there is a strong need for “new products which might be sources of the cash flow in the future”.

Tightening profit margins have developed strong cost pressures to the companies in the

industry and hence, the improvements have been mostly incremental. As one informant

explained, ”paper is a commodity product and its real price will be declining until it reaches

zero. We have advanced productivity but the actual products.. Our forest companies, they base

their actions on the wood and do not fiddle with the value chain.”

106

It has been indicated that the forest industry products have been traditionally seen as

commodity and hence it is difficult for one production unit to differentiate itself from the

competing ones because the products sold are “of uniform quality and do not create changes”

in the processes of the customer. Product differentiation, if not impossible to conduct, was

considered unprofitable in most cases. “In these bulk markets, product differentiation creates

only new costs: development costs, logistics-related costs.. When you develop differentiated

product variants, all they create is costs – storage costs, for example.”

Some informants were also quite negative towards the capability of the industry to renew itself.

As a chief technological officer stated, “the innovators might come outside the forest industry.

Bio refineries and that kind of things.. The forest industry has been quite incompetent in

redefining itself”. Still, some informants saw possibilities in the “businesses created nearby”. It

was stated that “the forest companies could take their share” by offering their current

infrastructure to the new companies generated near by.

The internationalization of the industry has had its own influence on the capabilities in which

companies are willing to invest in. According to interviewees, the paper industry was the first

to internationalize in Finland. That image has traditionally facilitated the recruitments into the

forest industry. Several informants suggested that the most skilled and hard-working people

tended to seek their way into the forest sector. As an SVP from chemicals industry claimed, “it

was like US Navy: join the navy, see the world. And those who went abroad, they worked 24

hours a day.” In the past decade, the amount of competitors has increased when companies

outside the Western countries have entered the field. Even though Western companies are still

suspicious of the quality of the outputs of the more unfamiliar producers the “border line has

been shifting towards the approval of even more crucial aspects relating to production from a

wider array of firms.” As an informant from machine building industry indicated, “five years

ago no Western factory would even consider the possibility of importing equipment from

China, no way. But then, let´s say about two years ago, they began to approve some parts which

were not in contact with processes. They were like, okay – why not they do it, it is not the end

of the world.”

4.1.3 National culture

Among other factors introduced, it seems that national culture direct the capability selection

process. In certain situations, honesty – “obeying the laws and willingness to do the right

things”, gives a positive added value to the business but in some other situations it does not.

107

Positive added value could be created when the customer values the reliability of the Finnish

way of dealing with situations. The Finns were said to quite commonly underestimate their

products: “this is quite good, but then there are certain aspects that could be improved”. Some

customers may appreciate the modest attitude but some may claim that it creates uncertainty.

Typical feature mentioned of Finnish national culture, modesty, is noted as one of the factors

that hinder marketing capabilities of the forest cluster companies. Some informants claimed

that modesty could lead as far as international competitors “stealing” the ideas of the Finnish

counterparts. Interviewees assumed that also the “introvert nature of the Finns” might hinder

the capability be able to commercialize their ideas and sell the products or services.

Some informants claimed that modesty and negativity have led to an atmosphere in which the

striving for consensus has been a factor hindering innovative capabilities. Some interviewees

commented that it has been a challenge to present differing views in the organizations.

However, the changing work environment and new generation of employees have shaken the

traditional way of thinking. One interviewee noted that ”working in the community is

nowadays quite different. When new generations of employees are recruited, the ways of co-

operation tend to be distinct from the previous ones.” It was also pointed out that the

willingness to discuss controversial topics “differs from the traditional Finnish paper mill

community” when new generations of employees change the way of thinking.

4.1.4 Firm culture

The static nature of the industry has affected, for example, the managerial capabilities. “The

besetting sin” has been that employees with strong specialized knowledge bases are promoted

to managerial positions, and consequently “he or she may not be able to adapt to changing

role.” They might still “want to concentrate on their specialized area of knowledge, and ignore

the importance of the leadership of people.” Also the vacancy mindset has had its consequences

on the rigidity of the industry. The organization structures have been inflexible because of the

many organization levels created around central employees. As a VP from machine building

industry remarked, “when we move up in the organization, everyone has a certain, limited box

– kind of an own organization, which builds up several mini-organizations.” It has also been

noted that ”when organizations grow and develop more organization levels”, the possibilities

for proactive adaptation decrease.

108

The mindset mentioned above has also had its effects on the capabilities to absorb and exploit

the information from outside the company borders. As one informant argued, the traditional

way of thinking has been based “on the thought that we know our stuff and can manage on our

own.. We haven´t admitted that we could learn from the knowledge outside the organization.”

Even if the organizations would be willing to remove some vacancies to strengthen the

adaptability, they would have to negotiate a unanimous decision with the trade unions which,

according to one interviewee, “have made the decision making process extremely stiff.”

According to the interviewees, the openness towards new ideas and “radical ways of thinking”

has been traditionally underestimated in the Finnish forest cluster firms. Interviewees

indicated that “when someone comes out with an idea, threatening factors are usually

highlighted” instead of “trying to develop the idea further and discover the possible positive

aspects.” Technological solutions might have taken the primary role in both implementation of

”new ideas” and redesign of business processes, yet the main challenges are seldom purely

technological. That might be the case because technological problems are often more straight

forward to solve because they can be better observed, defined and they have a “right” solution.

Also, breaking the long-standing structures and building a new operations model is considered

challenging because of the critique towards new ways of doing things and previous failures. As

one informant stated, “A considerable portion of staff argue that it will not be possible, and

base their arguments on the fact that it wasn´t possible in, for example, 1973.”

However, willingness to develop leading products and technologies has been a significant

driving force in the industry. “The engineer-like willingness to take ideas further” has created

the role of Finnish companies as the leading technology producers. When investing, there has

been an attitude of “always wanting to bring in some new, to be forerunners.” Among

informants, these attitudes were criticized as well. As noted by one interviewee, the industry

players “somehow got stuck in being technology slaves. I mean, we have always tried to develop

bigger, more beautiful and more complex.” An interviewee from the chemicals industry even

used the word “over-engineering” to describe the prevailing orientation in the research and

development in the industry.

4.1.5 Environmental change

Changes in the business environment, for example distribution of manufacturing into the

countries in which the raw material and labor costs are more affordable, have posed challenges

for the knowledge transfer. An interviewee from the chemicals industry noted that maintaining

109

competitiveness might require ”looking into the countries where the pulp making will be in the

future. We can still develop technology even if it would be used somewhere else – there is a

strong demand for know-how.” The comments reflect the idea that in changing environment,

organizations are forced to focus on knowledge sharing capabilities. When establishing new

production plants abroad, for example, the companies need to acquire the capabilities to

transfer the process knowledge needed for the successful establishment of production plants.

As a SVP from machine building industry cited, “in the global business, the core knowledge is

very limited resource. It is a big challenge how you can put this resource into supporting China,

Japan, Brazil and all those countries at the same time.”

Still, several interviewees were concerned about the extent of knowledge spilling over abroad.

The interviewees have regarded the declining path of the North American firms as an

exemplary on the issue. When the investment rate in North America was its highest, “the

North-American companies flourished”. The top knowledge seemed to be a couple of decades

ago still in North America and then “the Scandinavian forest industry began to develop and

also the technological know-how followed.” Some of the informants also pointed out that many

companies “are thinking of themselves, not the national interest”. As an interviewee from the

machine building industry noted, “if this industry is mitigated in Finland, we will have to move

with our research and development resources towards the countries which have the best

hands-on know-how.”

The informants regarded the capability to produce high-quality outputs a particular strength of

the industry. However, it was stated that the importance of quality factors has been declined

and the competition is based mainly on the price of the product. As an informant noted, “Low

cost good enough – there is no need to produce the best quality.” There has been a growing

need for complexity reduction, thus ”not that many basically similar product types are needed”.

As an informant stated, ”we don´t have to provide the best quality anymore – instead, we

should have the quality fit for purpose. Good enough is enough.”

According to the interviewees, the cost pressures have limited the opportunities to develop new

products. “There is always the threat that the hunt of cost savings drives us to a situation where

development is not worthwhile any longer.” It has been indicated that “many competitors have

succeeded without developing anything”. Instead, different kinds of capabilities have reasserted

their importance. Capability to build relationships with customers, for example, has been noted

to be of high significance. As an interviewee from paper industry noted, “We have clearly aimed

at building partnerships with the customer so that we can further improve and develop the

110

products we sell well. It could be bulk sales but we can also add value by ensuring that the

performance of the product in the customer´s processes is of high value.”

Traditionally, the forest industry has received some special treatment from the government.

The policy makers have considered that the “relatively big and responsible domestic buyers are

a good thing.” It has also been indicated that former solution and “gimmick” towards the

profitability challenges was the devaluation of the Finnish currency. Joining the Euro area has

abolished that possibility and “the problems won´t be solved so easily anymore.” Tightening

competition regulations have had an effect on the selection of the capabilities firms want to

acquire and develop, as well. According to interviewees, the openness and transparency in the

forest industry has been declining radically in the last decades. One of the interviewees even

stated that the “operations of the Competition Office have made discourse impossible.” Several

interviewees argued that the competition regulations have had a considerable effect on the

possibilities to develop collaboration capabilities and build networks in which information

could be exchanged. “Nowadays we can co-operate through different research arenas but the

actual development which might lead to innovation – there, the boundaries are rapidly faced.”

In some cases, restriction of collaboration is considered going to the extremes. For example,

one informant told us a story of him and his colleague from competing organization settling a

casual appointment of going to see a ice-hockey game together. “We both had to report the

event in detail to our superiors, and write a certain memo about the reason for the meeting and

guarantee that we will not be discussing work-related topics, and sign on the dotted line.”

111

Table 2 – Representative quotations on capability selection

Theme

Selection

History ”The printing paper segment grew beyond measure – we were kind of stuck.”

“We have invested a lot in the 1970s and 1980s into these factories and

clearly emphasized larger capacity – more production volume. It can be seen

even today when comparing the factory sizes globally.”

”Formerly, it was very helpful that you could call any one in any firm –

everybody knew each other.” ”Let´s say, from the 1970s to the 2000s, there

was a vast technological development path. It was mainly based on

developing efficiency and production systems. That, in turn, enabled the

development of machinery. In that moment, there was a need for that kind of

system – the consumption of paper and cardboard were increasing every

year.”

Industry logic “It´s something like buying gasoline for your car.. The differences between

the gasoline if you buy from Neste or ST1, or if the oil is imported from

Russia or if it is imported from Saudi-Arabia or from Norway.. The end

product is practically the same.”

”Pulp is even more a raw material [than paper]. For example, gold has its

price, silver has its price, and also the pulp has a certain price that everyone

can read in Financial Times. So, it is pure commodity.”

National culture ”It could be that in Finnish forest industry, there is too much of a “family-

centered” ethos - reluctance to exchange views with others.”

“the Finns have unnecessary modesty and shyness which originates from the

Finnish national character.. That we belittle our accomplishments”.

Firm culture ”In our value chain, there are a lot of areas where things are done in the same

way as always, and producing the same products is promoted because it

seems to be enough to succeed.”

”There was this engineer-like willingness to take ideas further. When

investing, we always wanted to bring in some new, to be forerunners.”

112

Environmental

change

“The challenge is how to form comprehensive outlook of the innovations

and findings made in the different sectors”

“There is always the threat that the hunt for cost savings drives us to a

situation where further developments are not worthwhile any longer.”

4.2 Capability gaps

The second overarching dimension of the capability evolution process assembled according to

the comments made by the interviewees was capability gaps. The aim of the following chapter

is to introduce the factors that the informants have considered significant especially in the

future but, according to the informants, in which the forest sector has not paid sufficient

attention to.

Figure 4 – Findings on capability gaps

4.2.1 Risk taking

Several interviewees have claimed that the capability to sell the new and innovative solutions to

customers is probably a weakness in the industry. The ideal situation would be to “find a first

preferred customer, that we collaborate with them, develop a product for which our customer

has privilege in that particular market.” However, “tested” technologies are preferred. As an

interviewee from pulp and paper industry and the other from chemicals industry claimed, “if

they build a factory, they are willing to take risks to a certain point..”; ”We have a ready product

113

but we haven´t found the customer who would be willing to buy it.” A R&D manager also

indicated, ” If one has developed a new piece of machinery, customers are not ready to bring it

on line before they see that it functions well in two or three other factories.” According to

respondents, customers are actually interested in new products and services “but there should

be vast production volumes” if they are willing to take the risk and invest on completely new

solutions or to modify existing structures. As stated by a R&D manager, “Changing the big

picture, it requires kind of a snap in the brain so that you go along.” However, the interviewees

pointed out that risk sharing possibilities should be created, for instance, through ”developing

new businesses” which should begin with “promoting entrepreneurship.”

However, carefulness and even risk aversion are considered logical as risk sizes can be

measured in billions. When ”it is the case of a couple of millions, the firms are willing to take

the risks and try out new. But considering the change of the whole production unit, the fear of

malfunctions is often overwhelming. It is about several dozens, even hundreds of millions. No

one is willing to risk the functionality of the organization as a whole.” The networking

capabilities with customers should be also advanced because it was seen decreasing risk

aversion.

The negative experiences from the past tend to affect the decisions concerning risk taking. An

informant named an example of a large pulp mill which was built in Germany in the 1990s and

was based on completely new technology. Several mistakes were made, and quite rapidly they

had to close the whole production unit. “These kinds of examples of the radical thought have

induced that the firms are not so enthusiastic to search for large changes.” Hence, the

interviewees call for the attitudes which would be more accepting towards possible failures. ”In

EU, and especially in Finland.. There should be this, encouraging.. That not everything should

be perfect at once – that one could experiment, and succeed, and fail.”

4.2.2 Recruitment

Several informants pointed out that the development of the present human resources has been

a particular strength in the forest industry companies. Hence, the companies have invested in

maintaining and broadening the capability base of the employees. However, the need of

substance knowledge of the application area has led to the recruitment of the employees with

homogenous backgrounds. As a SVP of research and development stated, “Our problem is that

we have too many paper engineers when we should have physicists, chemists, mathematicians,

and then a couple of paper engineers.” The recruitment of people with differing backgrounds

114

was seen as important factor in renewal process. The current needs of broader employee base

were clearly emphasized in several interviews, but the “profiles are too often created based on

the existing structure.. We should clearly reform in a larger scale” As an informant from

machine building industry stated, the recruitment should also support the changing strategies:

“to become a global service provider, we need a wide range of employees. We should hire them

all around the world and they should mainly be local talents. From knowledge-perspective, it

becomes a challenge that we should have a network through which we could recruit skilled

people around the world.”

Traditional educational structures in Finland were considered well-established but they might

“feed these competences built in the past. And then, the young people study both qualitatively

and quantitatively the things that were important in the past.” Also, the attractiveness of the

industry was seen as poor. As stated, ”the educational base has traditionally been very good but

the problem is that young people are not funneled into the industry anymore.” The image of

the industry, created mostly be public media reporting “plant closures, lay-offs and such” has

led to a situation in which “the attractiveness of the whole industry has been such that lately no

distinct talents have found their way to the relevant education. When looking into the

admittance statistics into the universities, this industry is quite poor rated.”

4.2.3 Executive message

Creating a positive image for the industry is also seen important as a means of attracting skilled

individuals, “when we have a positive drive, it creates an environment in which people want to

come to.” As a chief technology put it, “We are in big trouble right now because we should

report our issues truthfully - We have plenty of things with which we can be able to compete in

the future.” The overall understanding was that the leaders should not delude either general

public or current employees with misinformation or palliate the truth. Instead, they should

create an image of the industry in which “something new will arise from these ruins”. As a

means of creating the image, one interviewee proposed that the managers ”should not always

complain in the media how miserable this is.. Instead, they should be giving an example that

this is not a dying business.”

As interviewees indicated, the charisma of the messenger has an impact on the comprehension

of the message. As one informant noted “on account of everything, it would be worthwhile to

have media-capable leaders” in the industry. Other interviewee noted that “most of the CEOs

are, kind of, gray.” He also pointed out that ”they are part of the group of engineers who act

115

based on the facts” and even though he did not deny that they are working hard, he stressed

that “they have not enough charisma to be front men of this industry”. Another informant

noted that the front men “are not painting a big picture about the direction where we are

headed, or give a good example.”

According to the interviewees, delivering the executive message should be given more

importance. The core communication “should be verified from the executive group, and it

should be veracious and transparent” and the companies should “have clear core messages, and

repeat them” because there is always someone present who hears that message for the first

time. However, informants recognized that those messages are not so often delivered to general

public, and the emphasis of the ”communications relate to the stock exchange values,

information about new sales and such.”

Even the good news may be difficult to pass to the general public. Many interviewees noted

that delivering positive news may result to unreasonable demands. ”If you say anything more or

less positive, there is always someone demanding something”. The “unreasonable demands”

seemed to be related to the media fuelling an erroneous impression that change and

innovations could be created “overnight”. As one interviewee told us, “the thought of renewal

through research and development taking place during few months is ridiculous.”

4.2.4 Ideas

All companies need a rich portfolio of ideas where to choose the ones which will be discussed

and developed further. According to the informants, there are several areas which have a strong

profit potential but which are not sufficiently exploited. For example, several respondents

considered environmental knowledge and corporate social responsibility issues as being fields

of know-how which could have potential, if sufficiently commercialized, to increase their

importance in the future.

Also offering complete solutions is increasing its significance in the forest industry. Several

informants have emphasized the importance of finding the synergies between the whole

supply-chain and delivering larger concepts. As a VP from machine building sector noted, “we

don´t get the inquiry to deliver certain parts of the factory that often any more. It is more and

more like, deliver the whole concept.” For example, the combination of products and services is

regarded as a potential area of growth. It seems that utilizing the product and service

combinations has occurred earlier as well but the companies have not been able to

116

commercialize the service-side of the combination. As one informant noted, the service aspects

related to products have been ”a tool for broadening our market share. We have not been able

to commercialize it, though.” After-sales support, for instance, was one particular issue

mentioned several times during the interviews from which the forest companies could take

advantage of.

4.2.5 Customer needs

The informants have stressed the growing importance of the knowledge of the end customer´s

needs. As stated by one informant, ”it is not necessarily crucial what your primary customer

thinks about the product. Often the more crucial factor is that what their customer

appreciates.” The networking capabilities through the whole supply-chain are regarded

insufficient among the interviewees. As an interviewee pointed out, the companies should

better ”understand that the customer has its own customers, and that their needs could wiggle

the whole supply chain.”

Complexity in the supply chain has been stated to be one of the key barriers hindering the

innovative capabilities. Complicated products require certain level of expertise from all the

members of the supply chain, and therefore represent a barrier to involving intermediaries. As

one informant from paper industry claimed, “We aren´t just getting over that there are so many

people that block the message away. For example, in the fine paper segment, you have to sell

the product for us, and we have many difficulties in understanding that, and then we have to

sell the product to our wholesalers. And then they have to sell the product to the 200 000

printing companies which have to sell that to their millions of customers.” The more

complicated the products are, the more training should be provided for partners. There has

been a broad consensus that strengthening the networking capabilities with customers could

lead to significant profits. Stability and continuity were considered important in inter-firm

relationships, as well. As a Senior Vice president of sales and marketing told us, ”we should

have more co-operational projects with our customers in which the product, kind of sells itself,

and probably the whole process and market launch would be faster, then.”

Several interviewees stated that the “engineer-like mindset” mentioned in the previous chapter

could lead to the ”development of the products for the sake of the development, and not for the

sake of the customer needs.” Industry experts claimed that the knowledge of customer needs is

delivered insufficiently between the units with market-connection and the departments of

research and development: “they do not necessarily realize that the information they possess

117

could be very valuable in the next stage in the organization.” A general impression was that too

often the different business units could not articulate the needs of the customers clearly

enough to the production units. As one interviewee complained, “We could get more agility out

of our production units if we could reason them that this is the actual need – why we are

actually making these products.”

The technology push-factor was also considered having too much significance among the sales

and marketing processes. When developing new products, the customer demands were often

considered in the early development process but tended to lose their importance in the

progress. As a VP of research and development told us regarding one example of the product

development process, ”Maybe the basic idea came from the upcoming customer need but then

it quite rapidly turned into.. How we could actually do this, which technology to use, and how

to promote this selected technology. It carried too much weight, the push-element.” As another

informant put it: ”Our product development forums, they lack customer connection” and posed

a question: “Are they too far from the real world?”

There was also a concern that the customers being not able to articulate their needs might be

forcing the companies to pay attention to the products which are not actually profitable. The

basic problem seems to be that the customers “do not necessarily know, or be able to bring it

out in a way that it could be easily understood”. Sometimes, the articulated customer needs

seemed to be even too straight-forward. A Vice President of R&D operations explained, “if we

ask our customers what they really want – they cannot put that in words. They might say that

they want as good quality as cheap it it is possible – but those things we already know.” That

seems to be a challenge in the R&D departments in which “the visions might be based on

current actions rather than predicting and estimating the upcoming need”.

118

Table 3 – Representative quotations on capability gaps

Theme

Capability gaps

Risk taking ”The structures are heavy, the factories and integrations are of great size, in

pulp and paper… It is not so easy to change those, they are really tough

decisions to make.”

”When you have a 50-million-project, then there is a real opportunity but

there are also numerous risks.”

“When the sums are significant, they have naturally wanted to minimize the

risks so that the processes would start up as soon as possible and they would

do what they have promised to do.”

Recruitment ”On the one hand, we have a need for in-depth knowledge.. Doing big things

in a narrow area. On the other hand, we need generalists who can see the

connections, and are able to piece together and combine these issues.”

”Probably, on average, the situation is that we have too many people with

similar knowledge and such – people with similar abilities. But of course, we

shouldn´t go too far.. It could be quite difficult to control, then.”

Executive message ”Naturally there are several points of views if you ask different managers:

one claims that if we go straight to the subject, it will be fine in the end. And

the other relies on logical argumentation and tries his best in developing a

mutual understanding.”

”The image which they build is very depressing for their own employees and

for everyone. If you will succeed, you should stop whining and get into

work.”

”If you say anything more or less positive, there is always someone

demanding something.”

Ideas ”The production unit would do some [corporate social responsibility]

enhancements, aim at something.. And then they could send the authorities a

message of “we can do this, isn´t that a good thing to do?””

119

4.3 Capability development

The third overarching dimension of the capability evolution process assembled according to

the comments made by the interviewees was capability selection. The purpose of the following

chapter is to describe the ways the forest sector has been developing their capabilities.

Figure 5 – Findings on capability development

4.3.1 Organizational change

When describing the efforts the organizations in the forest sector have taken to develop their

capabilities in particular areas, the organizational change aspects became evident. According to

the interviewees, the streamlining of organizations has been an important mode of operation to

change the established structures. As one informant described the process in his organization,

“until the current decade, the mills were quite independent units. Then we combined our

forces to operate as company”. Another SVP reflected the change in his organization during the

last decade as follows: “In Finland there was a great transformation.. Not every plant should

have its own HR manager, and he/she, in turn, did not have to have three assistants.”

According to informants, redesigning the tasks and the processes has resulted in corporate

cultures in which solving the emerging problems becomes a part of work routines.

Customer needs ”The customer needs solutions it didn´t know it needed before – Not that we

force our customers to buy something they do not really need.. We should

really help them to succeed and then take a part of the profits”

”It is a problem that we do not sell the product the right way, or we cannot

utilize the know-how we actually have"

”The customer knowledge should be spread unbiased through the

organization. So that the people in the customer interface do not make

interpretations on their own”

120

Competence focus change was also a theme emerging from several interviews. Focusing on one

segment and investing in the development of that particular sector, “expertise in a particular

area”, was seen as means of enhancing capabilities. As an interviewee from the pulp industry

explained, ”Our company has had to transform from a large producing organization into a

market pulp supplier.. So we have to break our habits and thoughts.”

The organizations have attempted to change through mergers and acquisitions as well. For

example, an informant from machine building industry regarded ”acquisitions of smaller

technology firms” as a natural part of development of technologies. As he indicated, ”when

there has been skilled innovators and entrepreneurs, they have developed and commercialized

technology” in a smaller scale, and ”it has been profitable for us to buy the whole firm to

acquire that technology”. As a result, the entrepreneurial capabilities may have been

transferred into the organization.

4.3.2 HR

Maintaining and broadening competence bases of the employees have been largely based on

the fact that the knowledge requirements seem to have been increasing gradually. As an

interviewee explained, for example “the middle management assignments have become wider

of context.” Deeper understanding of the issues related to the work assignments seem to be

needed more often. Training in an industry is the formal procedure which a company uses to

facilitate employees’ learning so that their resultant behavior contributes to the attainment of

the company’s goals and objectives. When the knowledge bases of the employees grow, they

are offered more challenging tasks. As an informant from machine building sector noted, it is

crucial to offer interesting and multi-faceted positions to retain the most promising employees:

”younger people do not want simple, narrow tasks anymore”. “Those days are gone when you

worked for the same firm from cradle to grave”, and retention of key personnel and clarifying

their commitment to organization facilitates the development of organizational capabilities.

Some informants pointed out the importance of the tools for rewarding employees. The

cultures of using differing rewarding tools seem to be quite young in the Finnish forest

industry, and therefore unknown by both superiors as well as the subordinates.

As an important tool for broadening the knowledge bases of the individuals employed in the

industry, the interviewees have mentioned job rotation opportunities. Nowadays, the

companies are of sufficient size to offer opportunities for job rotation inside the firms. As a SVP

of human resources indicated, “The crucial issue for development could be the job rotation

121

between the firms – that people move between organizations and also from the supplier

organization to customer organization. That is how we could get different angles of views”

The job rotation aspects are also related to the international opportunities. There is a broad

consensus that the forest industry offers possibilities for the individuals willing to work abroad.

Globally, “the positions offered are various”. That aspect has been an advantage when

developing the HR capabilities, and also when collaboration capabilities are considered. Even

though “the ideal situation would be that we could recruit locally”, the knowledge level abroad

is not usually sufficient and firms must send their employees from Finland. As an informant

from a paper company noted, “We should export the know-how, and there is only two ways to

do it. Either you move abroad or you engage in export business. It is clear that more and more

people have to earn their living abroad in a way or another.” However, several interviewees

stressed that it is not very simple to find employees who are willing to move abroad and “who

are able to succeed in an international arena”. Also activating people “to take the opportunity

in the earlier phase of the career” was also seen as a significant factor. Another issue emerging

from the interviews was taking advantage of all the profits possible from the internationally

experienced employees has been a concern. As one interviewee from chemicals industry told

us, “the companies may be not able to repatriate the employees.. Are they able to exploit the

experience which they have invested in?” When an expatriate located in South America, for

instance, comes back to home country ”they shouldn´t put him to manage issues of Northern

Finland.”

122

Table 4 – Representative quotations on capability development

Theme

Development

Organizational

change

”Everyone has an own role, and it releases resources in reporting, for

example.”

”That you explore and develop one segment more profoundly. It will create

better outcome than that you try to do all kinds of things.”

”We handle side by side these breakthrough-affairs and incremental

improvement. They are not exclusive to each other. – One of the basic

principles of incremental improvement is that we should improve but we

have to see the overall effects on the organization of these improvements.”

Human resources ”We have wide-ranging assignments, so more intimate knowledge is also

needed.. – So we are searching for completely new working methods and

boundaries”

“They should people with wider perspectives.. Who could be the key perons

in the cluster when these new technologies are developed.”

”There are different kinds of tasks in production, there are teams, change

positions in the organization.”

”It is an important part of human resource development, that they travel the

world and understand what is happening abroad, understand the markets as a

whole.”

4.4 Capability outcomes

The fourth overarching dimension of the capability evolution process assembled according to

the comments made by the interviewees was capability outcomes. The following chapter

describes the factors in which one could see if the capability development process has been

successful.

123

Figure 6 – Findings on capability outcomes

4.4.1 Executive commitment

It seemed that the positive as well as negative outcomes of the capability development process

are reflected in executive commitment. The changing markets and organizational restructuring

have posed challenges to the top management. For example, an interviewee from a paper

industry company told us that “the matrix organization requires more efforts from the

management. Communications, definitions of policies, to name a few examples.. So that there

wouldn´t be any contradictory interpretations.” The uniformity of actions, “ensuring the

understanding of the messages sent” and “sensitiveness of carrying out the changes” were seen

as ways of showing the commitment to employees as well as to other stakeholders of the

company. The interviewees regarded important that the management motivates the employees

“to do things in different ways” and promotes internal entrepreneurship, and an “atmosphere in

which people want to be innovative and find the ways of thinking differently.” However,

promoting new ideas requires a lot of effort from the idea generator, meaning well prepared

presentations to superiors, and willingness to invest a considerable amount of time and effort

into the promoting process.

The commitment and respect of the management have appears to be visible, for example, in

the tolerance of taking risks which might lead to new innovations. The informants considered

significant that someone with sufficient authority should “stand behind the decisions” in order

to “gain support from the whole organization.” The commitment of the top management was

seen as a factor that “inspires people”. As an interviewee indicated, “they show by their

individual behavior that there are abilities and willingness to take risks and stick their neck

out.”

Encouragement was another topic which was discussed in the interviews. If an idea is accepted

but the implementation process takes a long time, it might destroy the innovative spirit. As

stated, the negative executive commitment is visible in the decision making processes if

124

management “do the definition of policy needed.. It remains in the front line and then they

give advise afterwards.. It does not take things further.”

4.4.2 Indicators

How are the outcomes of the capability creation and development processes measured, then?

On one hand, the industry experts mentioned different efficacy indicators which might

describe the outcomes of the capability evolution process. On the other hand, innovations

seemed to be regarded as an indicator of successful capability development.

Concentrating on the technological capabilities and improving efficiency has enabled forest

industry to continuously increase its production volumes. As one interviewee indicated, they

are mostly “simple values, like mean velocity of paper machines”. There has been a broad

consensus among the informants that the forest industry has been successful in creating

capabilities related to improvements in efficiency of the processes. As a SVP stated, there has

been “amazing development in unit sizes”. Investments in machinery and improving

production technology stemming from the “cost reduction mentality” have “made possible to

produce more tons with less costs”. As one interviewee from paper industry noted, “Paper per

employee-indicators, they have been increasing year by year.”

According to the interviewees, the outcomes of the restructuring have enabled more flexible

processes and have facilitated constant flow of information. As a business controller put it,

“Our trump card is clearly that we are able to direct our efforts towards right things. We are

able to conduct better analyses and inform the management level more quickly”. Another

evident issue about the restructuring processes has been that restructuring enables the cost

savings when “less crew is needed.” Improving the supply chain management systems has also

been a factor in which the benefits could be seen. As an interviewee explained, “We are able to

manufacture and distribute the products effectively and they are of high quality.”

According to the interviewees, the product and process innovations indicate the success of

capability development. Even though some informants suggested that “innovativeness may not

traditionally have been that much emphasized in the industry”, especially the number or the

lack of “customer-related innovations, and new products approved by the customer” was

mentioned as an indicator reflecting the quality of the capabilities developed. However, new

ideas of doing things can be also related to cost-efficiency. For example, a Vice President of

sales and marketing noted, “decreasing the quantity of the different product types” has been a

125

recent transformation activity through which his company has been able to strengthen its

competitive position as well.

Table 5 – Representative quotations on capability outcomes

Theme

Outcomes

Executive

commitment

“Communication, self-control, ensuring the understanding of the messages,

repeating them and uniformity of the actions of the management.”

“Developing new products and innovations should be logical in this industry. And it

comes down from the highest level.. That we decide to do certain things and it has

continuity.”

”Big investments are made but the decisions fall flat.. There isn´t ability to follow

through with the reforms.”

Indicators ”If a mill was built in the 1970s, the capacity was about 50 to 200 thousand tons, in

a single factory. Nowadays, we are talking about million tons.”

”The application of process management model has improved the cost efficiency

and we also operate with smaller human resources now.”

”The technological innovations leading to larger production volumes are strongly

shown by the fact that when comparing the average sizes globally, we are clearly

successful.”

”Paper production pro employee-indicators have increased year by year.”

”The number of customer-related innovations, and then the new products which are

accepted by the customer and create added value.”

126

4.5 Conclusions and discussion

Figure 7 - Findings

Sustainable advantage requires more than the ownership of difficult-to-replicate assets: it also

requires organizational capabilities. These capabilities can be harnessed to continuously create,

extend, upgrade, protect, and keep relevant the enterprise´s unique asset base (Teece 2007).

The organization selects the capabilities which it desires to acquire and develop further

according to varying criteria. In the Finnish forest industry, history, industry logic, national and

organizational cultures, as well as environmental change seem to be the factors which have

affected most significantly the selection decisions.

Because of the industry logic, even small gains seem to matter as many traditional forest

products such as sawn lumber, wood chips and pulp are traded internationally as commodities

and break-through innovations are not usually the most driving factors in the traditional forest

industry companies (see Sathre & Gustavsson 2009; Ghosal & Nair-Reichert 2008). It tends to

127

lead to selection of the capabilities which could enhance the efficiency of the processes. That

seems to be in line with the previous research themes on the forest industry which suggest that

innovations tend to be incremental. Among other issues, the static nature of the organizational

cultures and the conservativeness of the industry have affected the capability selection process.

The interviewee insights on the subject tend to follow the logic introduced by Tremblay (1999)

and Hovgaard and Hansen (2004), as the improvements of the features of the existing products,

manufacturing capabilities and concentration on process innovations have been among the

main development efforts in the Finnish forest sector.

Dynamic capabilities include capabilities required to adapt to changing customer and

technological opportunities. They also embrace the enterprise´s capacity to shape the

ecosystem it occupies, develop new products and processes, and design and implement viable

business models (Teece 2007). The results of the present study indicate that the Finnish forest

cluster has traditionally been able to select the capabilities needed for the excellence in the

technological area. However, the selected capabilities have often been related to incremental

improvements, such as upgrading the machinery to match the current standards and be of a

better quality than the base of the competitors. As indicated by Augier and Teece (2006), the

asset investment choices may actually be one source of the capabilities. Accordingly, some

capabilities are rooted in routinized processes. In this case, these might include the reactive

adaptation to market situations or efficient division of labor in which the Finnish forest cluster

seems to be relatively successful. Additionally, some capabilities are rooted in creative and

differentiated entrepreneurial actions involving unusual skills (Augier & Teece 2006). Among

the industry informants, there was a broad consensus that these kinds of capabilities are

commonly underdeveloped in the Finnish forest sector.

The strong urge to stay ahead in the technological development by investing in the product

development actions seemed to be regarded as a particular strength of the industry among the

interviewees. Nevertheless, there seemed to be a consensus that the firms might not be able to

stay competitive in the future just by maintaining the current capability bases. When existing

firms enter a new market in which they do not currently participate in, they should either

develop new capabilities or alter the existing ones. A particular set of capabilities serves as a

platform into other markets and related product areas (Helfat & Lieberman 2002; Zander &

Kogut 1995). It seemed that the capabilities needed in changing markets are rather different

from those required in the past decades. There was a mutual understanding that

internationalization of the industry and various knowledge transfer challenges have affected

the capability selection decisions in the Finnish forest sector. For instance, the excellence in

128

producing outputs of extremely high quality seemed to have lost its importance in the past

decades because the customers seemed to appreciate the products which were more affordable

and of “sufficient quality”.

Capability gaps, meaning the reasons why certain capabilities the industry experts regarded as

important but which were not currently been selected and developed extensively further were

discussed extensively in the interviews. It seemed that the executive message generated by the

forest industry leaders was not comprehensively communicated to the general public, and not

even to the employees of the organizations. That, in turn, resulted in challenges in recruitment

of the skilled individuals who might have the potential of generating innovations and

distinctive capabilities to recombine their knowledge to improve the innovation (see Zander &

Kogut 1995). It was argued that the homogeneity of the educational backgrounds also hindered

capabilities which enhance innovation by breaking down the thought worlds that arise because

people not only know different things, but know those things differently (see Eisenhardt &

Martin 2000).

In the Finnish forest industry, risk aversion tends to affect the possibility that incumbent firms

will explore radical innovations. As stated in the interviews, the firms seem to limit their

investments to innovations that are close-in to the existing asset base (Teece 2007) and to focus

search activities to exploit established technological and process-related assets. This effect

makes it difficult for these enterprises to see potential radical innovations (Teece 2007). It was

also evident that networking capabilities between suppliers and customers should be enhanced

to promote the possible innovations.

Deep knowledge of the customers´ needs tend to enhance the innovative capabilities and the

creation of new ideas. They typically involve participation of people who bring together

different sources of expertise. These sources of expertise are essential for superior products

because they address a unique aspect of product quality (Eisenhardt & Martin 2000). In the

Finnish forest industry, the communication and information flow between departments seem

to be insufficient. Most organizational capabilities require integrating the specialist knowledge

bases of a number of individuals. Inventors shape the way of which these capabilities evolve by

using networks to overcome problems such as uncertainty and incomplete information (Nerkar

& Paruchuri 2005; Grant 1996). For instance, the capability of mobilizing a new product is

determined by the communication of tacit and explicit knowledge both internally and

externally (Bull & Ferguson 2006). The interviewees indicated that these knowledge bases

might not be sufficient enough to support the development and commercialization of new

129

innovations. For instance, the knowledge associated with product development, the technology

in question and an understanding of sales and marketing practices were not seen as

confronting sufficiently. The divisions of marketing and research and development seem to

lack the communication channel through which they could exchange views and transfer the

information.

The development of the capabilities seems to occur through organizational change and several

human resource development actions. Firms tend to develop their adaptive capability to

identify and capitalize on emerging market opportunities through evolution of organizational

forms (Wang & Ahmed 2007). That seems to be true in the Finnish forest industry as well.

Achieving internal cooperation requires good incentive alignment and shared goals throughout

the organization. Particularly in large traditional hierarchies, achieving functional integration is

a major organizational accomplishment (Augier & Teece 2006). Accordingly, the development

of selected capabilities was often carried out by reorganizing the structures of the companies.

Removing the unnecessary barriers and organizational levels were considered important to

enhance the capabilities needed in changing business environments.

Creation, absorption, integration and reconfiguration of knowledge within the organization,

and between the firm and organizations external to it, is important (Teece 2007; Verona &

Ravasi 2003) in capability development. Interviewees stressed the importance of the human

resource development activities as a tool for capability development. Attracting the skilled

employees and broadening their knowledge bases by, for instance, job rotation opportunities or

possibilities to work abroad were mentioned as means of developing organizational

capabilities.

The capabilities of learning from partners, integrating external information and transforming it

into firm-specific knowledge (see Verona & Ravasi 2003; Wang & Ahmed 2007) seem to be

demonstrating in the actions the firms take to create different networks between the different

players in the industry, as well as when mergers and acquisitions were discussed. Several

interviewees noted that firms appear to take account not only the resources they have, but also

of gaps between their pre-entry resources and those required for entry. Thus, established firms

with critical resource gaps are more likely to enter markets using modes of entry such as

acquisition, joint venture, or parent spin-offs (Helfat & Lieberman 2002, 734). For instance, the

acquisitions of smaller firms possessing the skills relevant to acquiring organization was seen as

an important way of developing technological capabilities.

130

Galunic and Eisenhardt (2001) emphasize the role of the managers: managers manipulating the

resources into new productive assets in the context of changing markets, shape dynamic

capabilities. In fact, the executive commitment was mentioned by several informants as

manifesting the efforts taken in the capability development. The negative outcomes of

executive commitment, in turn, were regarded as avoiding the responsibilities and not being

able to insure the employees and other stakeholders of the actions taken.

The firm could be investing in the capability to achieve lower production costs from the

economies of scale that result from the increased capacity (Maritan 2001) as discussed in the

interviews. It seems that the outcomes of “zero-order capabilities” (see Winter 2003) which

indicate how firms “earn their living now” tend to be the ones the Finnish forest sector has

been successful in developing. Accordingly, the interviewees regarded different efficiency

indicators as the most important manifestations of successful capability development.

131

REFERENCES

AUGIER, M. & TEECE, D. J. 2006. Understanding complex organization: the role of know-how,

internal structure, and human behavior in the evolution of capabilities, Industrial and

Corporate Change, 15, 395-416.

AUGIER, M. & TEECE, D. J. 2009. Dynamic Capabilities and the Role of Managers in Business

Strategy and Economic Performance, Organization Science, 20, 410-421.

BLYLER, M. & COFF, R.W. 2003. Dynamic capabilities, social capital, and rent appropriation:

Ties that split pies. Strategic Management Journal, 24, 677-686.

BONARDI, J. P., HOLBURN, G. L. F. & BERGH, R. G. V. 2006. Nonmarket strategy

performance: Evidence from US electric utilities. Academy of Management Journal, 49,

1209-1228.

BREWERTON, P. & MILLWARD, L. 2001. Organisational Research methods. Sage Publications:

United Kingdom.

BULL, L. & FERGUSON, I. 2006. Factors influencing the success of wood product innovation in

Australia and New Zealand. Forest Policy and Economics, 8, 742-750.

CAPRON, L. & Mitchell, W. 2009. Selection Capability: How Capability Gaps and Internal

Social Frictions Affect Internal and External Strategic Renewal. Organization Science,

20, 294-312.

CARPANO, C., RAHMAN, M. 2006. International competition in mature, localized industries:

Evidence from the US furniture industry. Journal of Business Research 59, 630-637.

COLLIS, D. J. 1994. Research note: how valuable are organizational capabilities? Strategic

management journal, Winter special issue 15, 143-152.

CORLEY, K. & GIOIA, D. A. 2004. Identity ambiguity and change in the wake of a corporate

spin-off. Administrative Science Quarterly, 49, 173-208.

DAVIES, A. & BRADY, T. 2000. Organisational capabilities and learning in complex product

systems: towards repeatable solutions. Research Policy, 29, 931-953.

DAY, G. S. 1994. The capabilities of market-driven organizations, Journal of marketing, 58, 37-

52.

DIAZ-BALTEIRO, L., HERRUZO, A. C., MARTINEZ, M. & GONZALEZ-PACHON, J. 2006. An

analysis of productive efficiency and innovation activity using DEA: An application to

Spain´s wood-based industry. Forest Policy and Economics, 8, 762-773.

DOSI, G., FAILLO, M. & MARENGO, L. 2008. Organizational capabilities, patterns of

knowledge accumulation and governance structures in business firms: An introduction.

Organization Studies, 29, 1165-1185.

ESKOLA, J. & SUORANTA, J. 2003. Johdatus laadulliseen tutkimukseen. Vastapaino: Finland.

132

ETHIRAJ, S. K., KALE, P., KRISHNAN, M. S. & SINGH, J. V. 2005. Where do capabilities come

from and how do they matter? A study in the software services industry. Strategic


GALUNIC, D. C. & EISENHARDT, K. M. 2001. Architectural innovation and modular corporate

forms. Academy of Management Journal, 44, 1229-1249.

GAVETTI, G. 2005. Cognition and hierarchy: Rethinking the microfoundations of capabilities'

development. Organization Science, 16, 599-617.

GHAURI, P., GRONHAUG, K. & KRISTIANSLUND, I. 1995. Research Methods in Business

Studies – A practical guide. Prentice Hall: United Kingdom.

GHOSAL, V. & NAIR-REICHERT, U. 2009. Investments in modernization, innovation and gains


536-547.

GRANT, R. M. 1996. Prospering in dynamically-competitive environments: Organizational

capability as knowledge integration. Organization Science, 7, 375-387.

HARDSTONE, G. A. P. 2004. Capabilities, structures and strategies re-examined: Incumbent

firms and the emergence of complex product systems (CoPS) in mature industries.

Technology Analysis & Strategic Management, 16, 173-196.

HELFAT, C. E. & Lieberman, M.B 2002. The birth of capabilities: market entry and the

importance of pre-history. Industrial and Corporate Change, 11, 725-760.

HELFAT, C. E. & PETERAF, M. A. 2003. The dynamic resource-based view: Capability

lifecycles, Strategic Management Journal, 24, 997-1010.

HIRSJÄRVI, S. & HURME, H. 1995. Teemahaastattelu. Yliopistopaino: Finland.

HIRSJÄRVI, S., REMES, P., & SAJAVAARA, P. 2000. Tutki ja kirjoita. Tammi: Finland.

HOOPES, D. G. & MADSEN, T. L. 2008. A capability-based view of competitive heterogeneity.

Industrial and Corporate Change, 17, 393-426.

HOVGAARD, A. & HANSEN, E. 2004. Innovativeness in the forest products industry. Forest

Products Journal, 54, 26-33.

KIM, L. 1998. Crisis construction and organizational learning: Capability building in catching-

up at Hyundai Motor.

KIRK, J. & MILLER, M. L. 1985. Reliability and validity in qualitative research. Sage Publications:

United States of America.

KLEPPER, S. 2002. The capabilities of new firms and the evolution of the US automobile

industry.

KNIGHT, G. A & CAVUSGIL, S. T. 2004. Innovation, Organizational Capabilities, and the Born-

Global Firm, Journal of International Business Studies, 35, 124-141.

133

KOGUT, B. & ZANDER, U. 1992. Knowledge of the firm, combinative capabilities, and the

replication of technology. Organization science, 3, 383-397.

KUSUNOKI, K., NONAKA, I. & NAGATA, A. 1998. Organizational Capabilities in Product

Development of Japanese Firms: A Conceptual Framework and Empirical Findings.

Organization Science, 9, 699-718.

KVALE, S. & BRINKMANN, S. 2009. InterViews. Learning the Craft of Qualitative Research

Interviewing. Second edition. SAGE Publications, Los Angeles.

LAZONIK, W. & PRENCIPE, A. 2005. Dynamic capabilities and sustained innovation: strategic

control and financial commitment at Rolls-Royce plc. Industrial and Corporate Change,

14, 501-542.

LEONARD-BARTON, D. 1992. Core capabilities and core rigidities: a paradox in managing new

product development. Strategic Management Journal, Summer Special Issue 13, 111-125.

LINCOLN, Y.S. & GUBA, E.G. 1985. Naturalistic Inquiry. Beverly Hill, CA: Sage.USA

LOASBY, B. J. 1998. The organisation of capabilities, Journal of economic behavior &

organization, 35, 139-160.

LÄHTINEN, K. & TOPPINEN, A. 2008. Financial performance in Finnish large- and medium-

sized sawmills: The effects of value-added creation and cost-efficiency seeking. Journal

of Forest Economics, 14, 289-305.

MALHOTRA, N. K. & BIRKS, D. F. 2007. Marketing Research, An applied Approach. Pearson

Education Limited: Spain.

MARITAN, C. A. 2001. Capital investment as investing in organizational capabilities: An

empirically grounded process model. Academy of Management Journal, 44, 513-531.

MARSHALL, C. & ROSSMAN, G. B. 2006. Designing Qualitative Research. Fourth edition. Sage

Publications, Thousand Oaks.j

MITCHELL, W. 1994. The Dynamics of Evolving Markets: The Effects of Business Sales and Age

on Dissolutions and Divestitures. Administrative Science Quarterly, 39, 575-602.

NAG, R., CORLEY, K. G., GIOIA, D. A. 2007. The intersection of organizational identity,

knowledge, and practice: Attempting strategic change via knowledge grafting. Academy

of Management Journal, 50, 821-847.

NELSON, R. R. 1991. Why do firms differ, and how does it matter. Strategic Management

Journal, 12, 61-74.

NERKAR, A. & PARUCHURI, S. 2005. Evolution of R&D capabilities: The role of knowledge

networks within a firm. Management Science, 51, 771-785.

REIS, H. T. & JUDD, C. M. 2000. Handbook of research methods in social and personality

psychology; Cambridge University Press,

134

REPENNING, N. P. & STERMAN, J. D. 2002. Capability traps and self-confirming attribution

errors in the dynamics of process improvement, Administrative Science Quarterly, 47,

265-295.

ROTHAERMEL, F. T. & HESS, A. M. 2007. Building dynamic capabilities: Innovation driven by

individual-, firm-, and network-level effects. Organization Science, 18, 898-921.

RUBIN, H. J. & RUBIN, I. S. 1995. Qualitative Interviewing – The Art of Hearing Data. Sage

Publications: United States of America.

SAKO, M. 2004. Supplier development at Honda, Nissan and Toyota: comparative case studies

of organizational capability enhancement. Industrial and Corporate Change, 13, 281-308.

STONE, E. 1978. Research methods in organizational behaviour. Scott, Foresman and

Company: United States of America.

TEECE, D. J. 2007. Explicating dynamic capabilities: The nature and microfoundations of

(sustainable) enterprise performance. Strategic Management Journal, 28, 1319-1350.



TREMBLAY, P. J. 1999. Organize to generate innovation. International Journal of Technology

Management, 17, 797-813.

VAN DEN BOSCH, F. A. J., VOLBERDA, H. W. & DE BOER, M. 1999. Coevolution of Firm

Absorptive Capacity and Knowledge Environment: Organizational Forms and

Combinative Capabilities. Organization Science, 10, 551-568.

VERONA, G. & RAVASI, D. 2003. Unbundling dynamic capabilities: an exploratory study of

continuous product innovation. Industrial and corporate change, 12, 577-606.

WANG, C. L. & AHMED, P. K. 2007. Dynamic capabilities: A review and research agenda.

International Journal of Management Reviews, 9, 31-51.

WINTER, S. G. 2003. Understanding dynamic capabilities. Strategic management journal, 24,

991-995.

YIN, R. K. 1985 & 2003. Case Study Research. Design and Methods. Third edition. Sage

Publications, Thousand Oaks.

ZANDER, U. & KOGUT, B. 1995. Knowledge and the speed of the transfer and imitation of

organizational capabilities – an empirical test. Organization Science, 6, 76-92.

ZOTT, C. 2003. Dynamic capabilities and the emergence of intraindustry differential firm

performance: Insights from a simulation study. Strategic Management Journal, 24, 97-

125.

135

CHAPTER 6

EVOLUTION OF CAPABILITIES IN THE FINNISH FOREST INDUSTRY: A QUANTITATIVE APPROACH

JAAKKO LINNAKANGAS


1 INTRODUCTION

1.1 Background and motivation

An essential question for organizations is how to create and sustain competitive advantage in

the long term. Over time researchers have tried to answer this question with a number of

theories. In the last decades the dominant ones have proposed that organization’s resources

and capabilities are behind firms’ superior performance. Competition forces companies to

continuously create and reconfigure their resources and capabilities to fit the changing

competitive environment (e.g. Augier and Teece 2009).

The same applies for clusters. In the globalized world companies’ activities have spread over

country borders and things such as comparative advantage have become less important in

competition. Cluster’s performance has become very dependent on how it can develop its

capabilities (Porter 1998) and on the knowledge the cluster possesses (Tallman et al. 2004). Its

innovativeness depends on its ability to recombine technological and organizational

capabilities (Heidenreich 2005). Knowledge and capabilities of a cluster are getting more and

more attention, even in traditional mature industries which also can be knowledge intensive

(Porter 1998).

136

The forest industry has traditionally been one of the main drivers of Finnish economy. Today,

Finnish forest industry is undergoing an era of, perhaps its history’s most prominent, change.

As a result, companies have experienced a decline in their performance. Changes, for example,

in technology, competition, and demand are reshaping the setting in which companies are

operating. Global competition, developing countries’ growth, changes in energy price, and so

forth, are all challenges for Finnish forest companies. Finnish forest cluster must capitalize its

current capabilities in new ways or develop new ones.

In a recent article concerning dynamic capabilities authors call for more research in several

areas (Easterby-Smith et al. 2009). Among other things, they call for more longitudinal studies

in the research of capability creation and evolution. They suggest research on capabilities’ link

to functional capabilities such as IT, R&D and marketing. They wish to see research including

more traditional industries. And they note that attention should be given to links between

capabilities and micro issues such as managerial cognition.

1.2 Objectives and scope of the study

The main objective of this study is to identify capabilities that have been developed in the

Finnish forest cluster firms in the last ten years. Another objective is to find out whether

companies have developed similar capability portfolios with each other and to find out which

capabilities have been developed simultaneously.

The objectives of the study can be presented in the following research question and its sub-

questions:

How have firms in the Finnish forest cluster developed their capability portfolios in the

last ten years?

What capabilities have they developed?

Have they developed capabilities consistently?

How are developed capabilities linked to innovation?

Regarding the first sub-question, firms have conceptually unlimited number of activities that

they do. Therefore it is necessary to define what counts as capability in this study. Because

sample firms differ in terms of their technological fields, capabilities must not be assessed

through outputs (for example through patents). Second sub-question is aimed at answering to

whether firms have developed some capabilities not only occasionally but consistently. This is

137

because organizations maintain and develop their capabilities by exercising them (Nelson and

Winter 1982). The third sub-question is aimed at linking developed capabilities to cluster’s

innovativeness because the study is made as a part of innovation programme.

To achieve these objectives, it is necessary to define criteria upon which the assessment of

capability development is based on – to define what is counted as capability development and

how it is measured. In this study the focus is on firms’ capability portfolios. Therefore how

individual capabilities are developed is not in the scope of this study but which capabilities are

developed. Assessment of capabilities’ or capability portfolios’ appropriateness to the

competitive environment is also not in the scope of this study. Therefore capabilities’ link to

performance is not covered, and it is not discussed whether companies have developed ‘good’

or ‘bad’ portfolios of capabilities.

1.3 Structure of the study

The remainder of the study is organized as follows. Next two chapters (chapters 2 and 3)

discuss prior literature of the subject and chapter 4 summarizes them together and the theory

framework is presented. After the literature part the empirical part starts with the used

methodology in chapter 5. It is followed by the research process in chapter 6, which leads to

results and discussion in chapters 7-8.

2 CAPABILITIES

2.1 Introduction

The literature on strategic management has given more and more attention to ‘capabilities’ in

the last decades as an extension to the resource-based view (Dosi et al. 2000:13; Easterby-Smith

et al. 2009:S1). A myriad of definitions for capabilities has emerged, and many different forms

of capabilities have been suggested to exist. For example Teece et al. (1997) discuss dynamic

capabilities, Winter (2000) discusses operational capabilities, whereas Mayer and Salomon

(2006) note that capabilities come in many forms, such as technological, managerial,

operational, marketing-based, etc. Collis (1994) states that there are almost as many definitions

for organizational capabilities as there are authors on the subject. This chapter is supposed to

shed light on the subject of capabilities. First, a practically inseparable resource-based theory of

a firm is presented.

138

2.2 The resource-based view

2.2.1 Definition of the RBV

The resourced-based view (RBV) is one of the many theories trying to explain why some firms

succeed and others do not. The theory is often put head-to-head with the traditional industrial

organization model (for example in Wernerfelt 1984; Barney 1991; Grant 1996a; Teece et al. 1997;

Barney 2001) which proposes that firm’s long-term performance results mainly from how a firm

can respond and adapt to the pressure from ‘competitive forces’. The industrial organization

model emphasizes factors such as how easy it is for new companies to enter the industry, how

intense is the competition, and how easily a product can be substituted with another. It is

assumed that to be in competitive advantage a company has to have in some way unique

product or service compared to competitors – for example in terms of product performance. In

brief, the model emphasizes the link between external environment and firm’s actions when

trying to explain differences in firms’ performances.

The resource-based model takes a rather divergent view. RBV assumes that the unique

products are eventually imitable by competitors, and therefore the competitive advantage

gained with them can only be temporary (e.g. Barney 1991). Penrose (1959) has been

acknowledged to be the first one who rose the idea that firm’s profitability and growth should

be thought primarily as a result from organization’s unique characteristics. These distinctive

characteristics are usually simply called resources. At the simplest, resources are defined as

“anything which could be thought of as a strength or weakness of a given firm” (Wernerfelt

1984:192). To be more precise, resources are defined as “all assets, capabilities, organizational

processes, firm attributes, information, knowledge etc. controlled by a firm” (Barney 1991:101).

However, the resource-based view and industrial organization model are not mutually

exclusive. RBV can be seen as a complement for the industrial organization model in a way that

resources make the competitive actions against industry forces possible (Amit and Schoemaker

1993). There have been studies made about the relative effects of industry and firm attributes

on firm’s long-term performance, and it seems that firm-specific attributes play a greater role

than the industry (Cool and Schendel 1988; Rumelt 1991; Barney 2001).

139

2.2.2 Value of resources

There are different types of resources. As an example, Wernerfelt (1984) mentions brand

names, in-house knowledge, machinery, and capital. Resources differ in how much they can

contribute to firm performance (Barney 1991). Barney introduces four well-known requirements

that can be given for resources which can lead to a sustainable competitive advantage (so-

called VRIN resources). First, they must be valuable in a way that they are meaningful for a

firm’s strategy. Second, they must be rare, since otherwise they would not be an advantage

compared to competitors who could also possess the same resources. Third, they must not be

substitutable, meaning that any other alternative resource available cannot be used by

competitors to implement a similar strategy. Last, they must be imperfectly imitable for the

same reasons than the last two requirements. In line with Barney, Amit and Schoemaker (1993)

propose that resources are strategically valuable when they are difficult to buy, sell, imitate or

substitute. According to them, also firm-specificity, durability, and scarcity add resources’

value. Moreover they argue that the value of firm’s assets depends on their complementarity,

meaning that resources’ strategic value is dependent also on other resources that a firm has in

its use. Wernerfelt (1984) notes this by emphasizing a firm’s whole resource portfolio rather

than just individual resources. Those resources and capabilities that are in this sense valuable

are strategic (Amit and Schoemaker 1993). Amit and Schoemaker propose that when a firm

enjoys a set of strategic assets it can generate organizational rents.

The greatest distinction is usually made between tangible and intangible resources (Galbreath

2005). Tangible resources are physical assets which can be touched, such as production

machines and buildings. Intangible resources are untouchable, such as knowledge, capabilities,

culture, and brands. Intangible assets are typically seen as more valuable than tangible assets

(Barney 2001:648), but some authors have argued that this is not always the case. Miller and

Shamsie (1996) argue that property-based resources are more valuable when the environment

is stable and predictable, and that knowledge contributes more to performance when the

environment is changing and unpredictable. Few can totally disagree with the proposition that

a firm with superior resources has a competitive advantage. However, because obtaining and

maintaining resources is costly, the firm with best resources is not necessarily the most

profitable one (Becerra 2008), and firms have to optimize their resources with their scarce

resources.

140

2.3 What are capabilities?

2.3.1 Defining capabilities

Organizational capabilities are often defined through so-called routines (e.g. Nelson and

Winter 1982; Grant 1996b; Eisenhardt and Martin 2000). At the simplest, routines mean how

things are done in an organization (Nelson and Winter 1982:400). More accurately, routines are

organization’s all the regular and predictable behavioral patterns (Nelson and Winter 1982:14).

The notion of routines is quite broad including all the patterned activities in an organization

from well-specified technical routines to business strategies (ibid.). Routines at the highest

level can even be as broad as describing the strategy and structure of a firm (Bruderer and

Singh 1996).

Winter defines organizational capabilities as a collection of routines, which combined with

other resources give options for different types of outputs (Winter 2000:992). There are other

similar definitions, such as capabilities being the efficiency of converting inputs into desired

outputs (Dutta et al. 2005), filling the gap between intention and outcome (Dosi et al. 2000),

firm's capacity to deploy resources - usually in combination - using organizational processes to

effect a desired end (Amit and Schoemaker 1993), and ability to perform a coordinated set of

tasks, utilizing organization’s resources, for achieving a particular end result (Helfat and

Peteraf 2003:999). Grant (1996a) sees capabilities being primarily an integration of a huge

network of routines in an organization. Most of these definitions share same ideas. Capabilities

are collective and continuous activities in a firm resulting in some types of outputs. For

example some companies can be seen as having a strong R&D capability resulting in a large

number of patents, and some can be thought as having a strong logistics capability resulting in

economical, fast or timely logistics. In using its capability an organization needs also its other

resources, such as production machines, for achieving its objectives.

Which capabilities are the most valuable for a firm? 'Core capabilities' are said to differentiate a

firm strategically from others (Leonard-Barton 1992; Kogut and Kulatilaka 2001) and 'distinctive

capabilities' to provide competitive advantage (Day 1994; D'Este 2002). Both of them are used

to describe organization’s unique skills and knowledge. In the resourced-based theories of firm-

specific advantage, particularly the concept of technological capabilities comes up (Mayer and

Salomon 2006).

141

2.3.2 Technological capabilities

Technological capabilities define the roots of a firm's sustainable competitive advantage (Lee et

al. 2001). There is no well-established definition for the concept of technology in prior

literature. Nelson and Winter (1982:104) discuss technology by writing “skills, organization, and

‘technology’ are intimately intertwined in a functioning routine, and it is difficult to say exactly

where one aspect ends and another begins”. Levinthal and March (1981) see technology as any

semi-stable way how an organization functions and deals with its environment. Zyglidopoulos

(1999) distinguishes, referring to several authors, two dimensions of technology. The narrow

dimension is formed only by codifiable explicit information which is held in patents, manuals

and blueprints, while the broad dimension is in tacit form. He argues that, from the resource-

based perspective, technology is embodied in physical and human assets in an organization –

that is routines and capabilities.

The concept of technological capabilities is as vague as the technology itself. The concept of

technological capabilities is often linked tightly to a firm’s R&D activities and has been

analyzed with firms’ products and patents (Stuart and Podolny 1996). More abstract approach

is also common. Mayer and Salomon (2006) handle technological capabilities as an ability to

fulfill customer’s requirements in many different aspects such as cost. Bell and Pavitt (Bell and

Pavitt 1995:78) define technological capabilities as “the resources needed to generate and

manage technological change, including skills, knowledge, experience and organizational

systems”. Figueiredo (2002:74) views them as “resources needed to generate and manage

improvements in processes and production organisation, products, equipment, and

engineering projects”. Nelson and Winter (1982) prefer discussing knowledge which an

organization holds as a whole and thus defines the boundaries for firm’s choices regarding

technology.

The discussion above suggests that a set of different capabilities, or dimensions of knowledge,

define firm’s technological capabilities. After all, firms are formed by different types of

capabilities in order to carry out their technological activities (Figueiredo 2003). This view is

supported by Dosi (1982) who notes that firms’ non-technological assets influence the direction

of its technological trajectory because it’s more likely that firms will develop those technologies

which can be supported by their existing assets. Wernerfelt (1984) proposes that by specifying a

resource profile for a firm, it is possible to find the optimal product-market activities for it. For

example a firm with strong marketing capability is probable to develop products which

142

especially benefit from marketing. Peteraf (1993:188) argues that firms tend to enter those

markets where firm’s resource capabilities match the market’s resource requirements.

2.4 Capabilities and path-dependence

As discussed, organizations change their routines in order to adapt to the changing

environment. However, firms cannot once in a while jump from one optimal state to another,

because changing the organization does not happen quickly, it’s costly, and because history

matters (Teece et al. 1997). Organization’s current state is molded by the path which it has

travelled and its history and current state restrict its possibilities in the future (ibid.).

History is not only a source of restrictions though. A company’s unique historical conditions

can be one reason why it can possess a set of resources and capabilities which are difficult to

imitate by the competitors and thus they can lead to a competitive advantage (Barney 1991).

Strong path-dependence is one characteristic of capabilities (Helfat and Peteraf 2003). The

phenomenon is called learning-by-doing (Pisano 2000), experience accumulation (Zollo and

Winter 2002), capability accumulation (Bell and Pavitt 1995; Figueiredo 2002; 2003), and

strengthening capabilities with use (Zahra et al. 2006). Because of learning by doing, exploiting

existing capabilities leads to cumulative and incremental improvement in them (Kogut and

Kulatilaka 2001). Building and maintaining strong capabilities therefore require consistent and

systematic development over time. However, systematic development of capabilities that were

successful in the past might not always be an advantage in the future. Accumulating old

expertise when change would be required can lead to competency traps (Levitt and March

1988:322) and capabilities can become rigidities (Leonard-Barton 1992).

2.5 Identifying and measuring capabilities

2.5.1 Obstacles in measurement

How can one say if a firm has a certain capability or not? Winter (2000) proposes that there is

no clear limit for having or not having one. The definitions in chapter 2.3.1, which use inputs

and outputs in defining capabilities, lead to an idea that capabilities could be identified by

observing the inputs i.e. the routines and resources in a firm, to see what a firm really does and

thus is capable of doing. Another way could be observing the outputs i.e. the tangible and

143

intangible results of firm activities, as a proof of an existing capability. The former might be

more complex to do because of the ambiguous nature of capabilities (Barney 1991).

What about assessing the strength of a capability? One characteristic of intangible resources is

that they are difficult to measure (Saaty et al. 2003). There exist no established measures for

capabilities, and that has been one of the problems in prior capability research (Robins and

Wiersema 1995; Dutta et al. 2005; Laamanen and Wallin 2009). Measuring capabilities is also

questionable in a sense that capabilities are firm-specific (Amit and Schoemaker 1993), which

makes them less comparable between firms. Dutta, Narasimhan et al. (2005) note that since

capabilities are in the middle of inputs and outputs, they should both be taken into account in

an assessment of capabilities. This carries conviction, considering that firms producing

identical outputs with different inputs, or vice versa, must have different strengths in their

capabilities. It has also been proposed that different external conditions can hinder the

comparison of capabilities between firms (Figueiredo 2003; Dutta et al. 2005). Helfat (1997)

points out that capabilities can be complementary in the sense that it is difficult to explain

some achieved output with some specific capability. In a similar manner Flynn and Flynn

(2004) discuss the interaction of capabilities: how they can be mutually reinforcing and

debilitating, and how other capabilities can lay the foundation for new capabilities.

To summarize, the outcomes of firm’s activities cannot be explained only by individual

capabilities. Capabilities function together with each other as a portfolio, and a changing

external environment requires changing configurations in capabilities. In developing

capabilities firms confront opportunity costs, meaning that they have limited assets for

developing capabilities (Dierickx and Cool 1989). Hence, developing one capability means less

development possibilities for other capabilities. Firm management has to choose an optimal

portfolio of capabilities to develop. Saaty, Vargas et al. (2003) suggest that resource allocation

and prioritization can be used as a measurement for intangible resources. Assuming that

management’s mental models drive the resource allocation, their attention reflects the

development of capabilities.

2.5.2 Attention for capabilities

Research on mental models of firm leaders has been seen as a potential addition to the

resource-based theory of a firm (Barr et al. 1992). It has been widely agreed that managers drive

the development of firm capabilities (Levinthal and Myatt 1994; Tripsas and Gavetti 2000;

Adner and Helfat 2003; Helfat and Peteraf 2003; Gavetti 2005; Zahra et al. 2006; Eggers and

144

Kaplan 2009; Laamanen and Wallin 2009; Sirmon and Hitt 2009). Manager cognition, that is

beliefs and mental models for decision making, drives manager action (Tripsas and Gavetti

2000; Laamanen and Wallin 2009).

Augier and Teece (2008) note that strategy processes are evolutionary in nature but involve

heavily intentional design from managers. Although many of the evolutionary theories stress

the role of environment and bounded rationality in capability evolution, it certainly does not

imply that management cannot affect the evolution of their organizations (Barron 2003). Some

evolutionary models accept the effect of environmental variables but also acknowledge that

strategic choices affect the evolution of a firm (Lamberg and Ojala 2006). This means that by

supporting certain activities by, for example, replicating well performing bundles of routines

(Dosi et al. 2008) management affects firm’s evolution. The evolution of a capability portfolio is

said to be a sort of a race where different capabilities compete of management’s attention

(Laamanen and Wallin 2009). When a capability portfolio is not working on a satisfactory level,

its constituting routines are modified through trial-and-error learning in an evolutionary

manner, and the resulting configuration significantly resemblances the predecessor

configuration (Lavie 2006).

The attention given by the management directs firm behavior (Ocasio 1997). The attention-

based view (ABV) thus helps to understand organization’s adaptation to changing environment

(ibid.). ABV is not only about management’s attention, but of whole organizations’ attention,

which is triggered by management (Sonpar and Golden-Biddle 2008) and communicated

through variety of communication channels such as annual reports and meetings (Ocasio 1997).

For example organization’s capabilities have been found to develop in accordance with the top

management’s attention (Eggers and Kaplan 2009).

Figure 1 Model of attention-based view (Sonpar and Golden-Biddle 2008)

2.6 Capabilities in the study

The capabilities in the study were iteratively chosen as a result of literature review and word

categorization in the empirical part of study. Table 1 shows capabilities in the literature and

those that were chosen in this study. Table 2 briefly summarizes them.

145

Table 1 Capabilities in the study and literature

In the study In the literature

Financial capability Cost control* (Day 1994) Financial management* (Day 1994) Financial management capability (Kochhar 1997)

Being responsible Corporate social responsiveness capability (Black 2006) Environmental health and safety* (Day 1994) Social and ethical response capabilities (Litz 1996) Stakeholder integration capability (Sharma and Vredenburg 1998)

Being strategic Strategy development* (Day 1994) Strategic planning capability (Yam et al. 2004)

Monitoring capability Market-linking capability (Song et al. 2007) Market sensing* (Day 1994) Competitive scanning capabilities (McEvily and Zaheer 1999)

Internal development HR capability (Park et al. 2004) Human resources management* (Day 1994)

Managerial capabilities Dynamic managerial capability (Adner and Helfat 2003) Leadership capability (Conger 2004) Managerial capability (Van den Bosch and Van Wijk 2001)

External sourcing Alliance capability (Kale et al. 2002; Draulans et al. 2003) Alliance management capability (Rothaermel and Deeds 2006) Integration capability (Zollo and Singh 2004) Relational capability (Lorenzoni and Lipparini 1999; Helfat et al. 2007)

Marketing capability Customer linking* (Day 1994) Customer order fulfillment* (Day 1994) Marketing capability (Grant 1991; Dutta et al. 1999; Yam et al. 2004; Vorhies and Morgan 2005; Song et al. 2007) Pricing* (Day 1994)

Procurement/logistics capability

Distribution capability (Grant 1991) Integrated logistics* (Day 1994) Logistics service capabilities (Lai 2004) Purchasing* (Day 1994) Supply chain management capabilities (Tracey et al. 2005)

Structuring Resource divestment capability (Moliterno and Wiersema 2007)

Operational capabilities Manufacturing capabilities (Schroeder et al. 2002; Yam et al. 2004) Manufacturing processes* (Day 1994) Quality management capabilities (McEvily and Zaheer 1999)

Service capability Customer service delivery* (Day 1994) Service capability (Grant 1991; Athreye 2005)

Change Dynamic capability (Teece et al. 1997)

Innovative capabilities Innovation capability (Cavusgil et al. 2003; Panayides 2006) New product/service development* (Day 1994) R&D capability (Yam et al. 2004) Technology development* (Day 1994) Technological innovation capabilities (Yam et al. 2004)

Internationalization capability

Internationalization capability (Chetty and Patterson 2002; Contractor 2007)

* From the figure “classifying capabilities” (Day 1994:41)

146

Table 2 Summary of capabilities in the study

Capability Summary

Financial capability Kochhar (1997) argues that possessing a resource which can provide a competitive advantage is not sufficient for obtaining economic rents if a firm has poor financial capabilities. According to him financial capability includes decisions, for example, related to firm’s capital structure. In Day’s (1994) model financial management and cost control are ‘inside-out’ capabilities that are needed for matching the market needs.

Being responsible According to Black (2006) the capability of being responsible is formed by a firm’s ability to recognize and meet its social responsibilities. Social responsibility means how an organization responds to expectations of different stakeholders (Black 2006). Litz (1996) argues that rather than seeing responsibility as necessity, developing such capability can be a source of strategic advantage. Sharma and Vredenburg (1998) found that those organizations which had proactive environmental strategies were able to develop superior capabilities for stakeholder integration, higher-order learning, and continuous innovation.

Being strategic In Day’s (1994) model strategy development is so-called spanning capability. Spanning capabilities are in the middle of inside-out and outside-in capabilities, and are needed for satisfying the anticipated needs of customers. Strategy is much emergent but also a result of deliberate planning (Mintzberg and Waters 1985). Yam et al. (2004) define strategic planning capability as a firm’s ability to identify internal strengths and connect them to external environment, to formulate and implement plans so that they are connected to firm’s vision and missions.

147

Monitoring competitive environment

Market sensing (Day 1994) is one type of outside-in capability which is needed to sense possibilities in the environment for deciding how to best serve them. Day sees this capability as the most important one for market-driven organizations along with customer-linking. A firm with strong market-linking capabilities is able to detect market changes and anticipate shifts in the market environment (Song et al. 2007). This ables a firm to sense opportunities before competitors and connect its other capabilities to external environment (ibid.). McEvily and Zaheer (1999) see competitive scanning capability as critical for a firm. Those capabilities enable a firm to monitor competitors’ strategies and tactics, and to collect information about the market.

Internal development Internal development is one way of obtaining new capabilities (Capron and Mitchell 2009). It includes, for example, internal training and hiring new staff (ibid.). HR capabilities are said to be pivotal for organizations (Park et al. 2004). They are needed for acquiring, developing, nurturing, and redeploying human resources in an organization (ibid.). Park et al. argue that HR capabilities can be a source of competitive advantage for a firm. Day (1994) sees human resource management as one of the market-driven company’s inside-out capabilities.

Managerial capabilities

Managerial capabilities of a firm are results of management’s collective knowledge and how it is applied (Van den Bosch and Van Wijk 2001). Dynamic managerial capabilities are needed for benefiting from organization’s resources and capabilities (Adner and Helfat 2003). Boeker and Wiltbank (2005) found that managerial capabilities have a great importance in how firms’ ventures succeed.

148

External sourcing External sourcing refers to a trading in a strategic capability in the form of purchase contracts, alliances, and acquisitions (Capron and Mitchell 2009). Kale et al. (2002) found that by investing in the development of alliance capability firms’ success rate in their alliances increased. The success enhanced in short-term by generating a positive stock response and in long-term by meeting the alliance objectives. Rothaermel and Deeds (2006) found that those firms which were able to manage their alliances effectively achieved higher levels of new product development, and the alliance management capability could act as a source of firm-level competitive advantage. Lorenzoni and Lipparini (1999) use the concept of relational capability to describe organization’s ability to interact with other companies, and argue that it has a positive effect on organization’s innovativeness.

Marketing capability Dutta et al. (1999) argue that marketing capability is an essential capability for an innovative firm. They propose that strong market-orientation is one of the most important sources of innovation – especially together with strong R&D capability. Similarly Vorhies and Morgan (2005) argue that marketing capability can be one source of competitive advantage.

Procurement/logistics capability

Tracey et al. (2005) argue that supply chain management capability should be regarded as important source of competitive advantage. These capabilities are formed by, among other things, efficient inbound and outbound transportation, warehousing, and inventory control. In Day’s (1994) model integrated logistics is one of the inside-out capabilities a market-driven organization needs.

Structuring capability Moliterno and Wiersema (2007) discuss organization’s capability to manage its resource portfolio, and particularly, by divesting some of its assets.

Operational capabilities

McEvily and Zaheer (1999) use the term quality management capability to describe knowledge which enables a firm to control its production processes. Schroeder et al. (2002) offer a comprehensive summary of literature about manufacturing capabilities and conclude that they have potential for providing competitive advantage. Day (1994) sees manufacturing capabilities as inside-out capabilities for market-driven organizations.

149

Service capability Customer service delivery is a spanning-capability in between the customer and a firm (Day 1994). Grant (1991) treats service capabilities as one factor behind differentiation and competitive advantage of a firm.

Change Dynamic capabilities (Teece et al. 1997) are behind organization’s ability to change. They come in many forms and can thus overlap with other capabilities in this study. Dynamic capabilities also include recognition of changes externally (Helfat et al. 2007). Therefore they are closely related to monitoring capability.

Innovative capabilities

Firm’s innovative performance is dependent on its innovation capability (Cavusgil et al. 2003). Sometimes it is thought to be an independent capability (Panayides 2006) but it is also acknowledged that it may come in many forms such as R&D capability or resource allocation (Yam et al. 2004).

Internationalization capability

Internationalization capability can be defined as a firm’s ability to reproduce an organization and staff in a foreign location (Contractor 2007). Contractor argues that this capability can be a distinctive one - learnt only in some organizations.

150

3 ORGANIZATIONAL EVOLUTION

This chapter discusses organizational evolution. Organizations are seen here as bundles of

resources and capabilities (Amit and Schoemaker 1993) and the evolution is thus approached

from the perspective of capability development. Before going into how organizations evolve,

the concepts of configuration and fitness are discussed, as they have been used before for

modeling evolutionary processes in strategic management (Siggelkow 2001).

3.1 Organizational configurations

Configurational approach is an important part of strategic management (Fiss 2007; Helfat et al.

2007). Any multidimensional constellation of conceptually distinct characteristics in a firm can

be thought of as a configuration (Meyer et al. 1993). The concept is quite broad, as these

characteristics can differ from organizational structures to activities and resources (Fiss 2007).

Configurations are not comprehensive descriptions of organizations but are used to describe

organizations’ strategies (Miller 1986; 1996).

The approach fits in to the resourced-based view (Black and Boal 1994; Delery and Doty 1996;

Miller 1996) and the capabilities perspective (Barron 2003; Lavie 2006; Helfat et al. 2007).

Dynamic capability literature discusses how organizations reconfigure their capabilities (Teece

et al. 1997) and how capability configurations are vital for creating value for customers (Sirmon

et al. 2007). One approach to organizational configurations is to look at capability portfolios

(Lavie 2006).

Why to look at capability portfolios rather than individual capabilities? First of all, as discussed

in earlier chapters, organization’s technological capabilities are formed by a set of capabilities

and resources rather than just one ‘technological capability’. It has been also argued that

resources are too often seen as individual entities which can lead to a competitive advantage

(Black and Boal 1994). Black and Boal propose that the way how a network of organization’s

resources is combined, or configured, is more vital for achieving competitive advantage than

individual resources. There are a number of examples of configurational approach in the

resource-based view. Kogut and Kulatilaka (2001) analyze firm’s choice of optimal capability

set. Galunic and Rodan (1998) discuss how resources can be reconfigured. Dierickx and Cool

(1989) handle assets as bundles. Flynn and Flynn (2004) note that capabilities can be built upon

each other. Some authors discuss assets (Teece 1986; Amit and Schoemaker 1993; Helfat 1997)

and capabilities (Moorman and Slotegraaf 1999) as complementaries. Teece, Pisano et al. (2000)

151

argue that technological innovations require different complementary capabilities. Being a

complementary is an essential part of configurations because it means that “the strategic value

of each asset's relative magnitude [may] increase with an increase in the relative magnitude of

other Strategic Assets” (Amit and Schoemaker 1993:39). Similarly, activities are complementary

“if doing (more of) any one of them increases the returns to doing (more of) the others”

(Milgrom and Roberts 1995:181).

Conceptually capability development takes place at different levels (Laamanen and Wallin

2009). Laamanen and Wallin propose that at the level of individual capabilities the interest is in

what ways those capabilities are developed, but at the level of a capability portfolio the interest

is in which capabilities company develops and which it does not. Levinthal (2000) argues that a

changing environment pressures an organization to re-evaluate and possibly repackage its

capabilities because a new environment can make some capabilities less and some more

valuable than they used to be. He also notes that changes in the environment may influence

the complementarity among sets of capabilities. For example new manufacturing technique

may place a premium on a different set of human-resource practices.

In brief, configurations mean different sets of choices made. As an example, if an organization

has A choices to make with B options in all of them, the total number of possible configurations

with these choices is . Delery and Doty (1996:804) summarize configurational approach well

by noting that “configurational theories are concerned with how the pattern of multiple

independent variables is related to a dependent variable rather than with how individual

independent variables are related to the dependent variable”. For example configurations can

be approached by looking at how multiple characteristics of an organization are related to

organization’s performance.

Configurations are also used in the context of organizational change (Siggelkow 2002). When

markets and technologies change, an organization must reconfigure its assets and structures

(Teece 2007). Generally it is not enough for a firm to change individual activities but it must

change a number of activities at the same time (Porter and Siggelkow 2008). By reconfiguring, a

firm can maintain its evolutionary fitness and escape unwanted path-dependencies (Teece

2007).

152

3.2 Organizational fitness

The framework of fitness was originally created in the context of evolutionary biology (Wright

1931), but is now considered relevant also to strategic management (Venkatraman and Camillus

1984) and organizational research (Ruef 1997; Beinhocker 1999; Gavetti and Levinthal 2000;

Siggelkow 2001). Though, there is no established definition for it in the field of business studies

(McCarthy and Tan 2000). The concept is usually associated with terms competitiveness,

effectiveness, profitability, return on investment, and customer satisfaction (ibid).

Organizational fitness can be distinguished into two types: internal fitness and environmental

fitness (Miller 1992). Siggelkow (2001) calls these internal fit and external fit. He approaches

internal fit as interactions of firm’s choices, how its internal logic work and external fit as the

appropriateness of the configuration to the prevailing external environment (ibid.).

Fitness landscapes are often used for illustrating the concept. Kauffman’s (1993) NK-model

discusses rugged fitness landscapes in the context of genotype mutations. The interest in

fitness landscapes was originally in organisms’ evolution - in an evolutionary process of

variation and selection. How organism’s performance changes in its environment when some of

its elements are changed? What kinds of combinations of genes are behind organism’s survival?

These types of analogies from biology are often used for modeling organizational evolution

(Nelson and Winter 1982; Bruderer and Singh 1996; Gavetti and Levinthal 2000). For example

organizational routines are treated as genes, which define organization’s possible behavior in a

similar way than genes define living organism’s behavior. As biological populations adapt to

their habitat with variations in their genes, organizations are thought to adapt to their

competitive environment by changing their routines (Nelson and Winter 1982). If capabilities

are viewed as higher level routines (Winter 2000), organizations adapt with changes in their

capabilities.

Firm’s performance can be simplified as being a function of its configuration. If a firm is

thought of as a system that has N attributes in its configuration, performance is a function of N

variables. These attributes can be activities, structures, capabilities, resources, etc. (Siggelkow

2001). A performance landscape can be created by mapping the function to N+1 dimensional

space where the extra dimension is the firm’s performance. Thus, fitness landscapes are

representations of all the possible configurations and performance of a population. For the

purpose of illustration in three dimensions, Siggelkow uses two variables in his paper. Although

visualization is difficult, the same logic can be used also with more dimensions.

153

Figure 2 Fitness landscapes

In the example landscape, consider that a company has only two capabilities to develop. It faces

a trade-off between being able to produce at low cost and to be able to differentiate. In the

landscape on the left the competitive environment is favorable for a company which can

produce at low cost (but to some extent have to also be able to differentiate) but in the

environment on the right companies benefit more from capabilities to differentiate (modified

from Siggelkow 2001). If the company tries to prosper in both of them in this particular

landscape, it does not occupy a peak but is “stuck in the middle” (Porter 1980).

The place of a firm on the map is determined by the values of the attributes (firm’s

configuration). Height of any particular point in the landscape represents firm’s performance -

that is external fit. Peaks in the landscape are configurations with an internal fit. This is

because if any component in a consistent configuration is being changed without other

components being changed, the performance can only decline as the configuration is not fully

consistent anymore. The strengths of the interactions between components define the

steepness of the peaks. This is because misalignments in systems with strong interactions are

particularly penalized. A landscape’s shape is determined by conditions in the competitive

environment, which consists of competitors, customers preferences, given technology, etc. The

shape can change as a result from changes in the competitive environment, and a firm can

change its place on the map by changing its configuration.

Siggelkow (2001) proposes that a performance landscape can be also thought of as manager’s

mental map [notions of „mental map‟ and „cognitive map‟ are often used to describe management‟s

mental models (Hambrick and Mason 1984; Barr et al. 1992)] about the performance landscape of

its environment. Gavetti and Levinthal (2000) have used similar fitness landscapes to model

managerial cognition and organizational search.

154

4 SYNTHESIS AND THEORETICAL FRAMEWORK

This section briefly summarizes the literature part of the study and a framework for evaluating

firms’ capability portfolios is presented. The framework is used in the empirical part of the

study.

4.1 Synthesis of the literature

The resource-based view (RBV) proposes that a firm’s long-term performance is largely

dependent on the resources a firm possesses (e.g. Wernerfelt 1984; Barney 1991). There exist

different types of resources a firm can have (Wernerfelt 1984). At present, capabilities and

knowledge are seen as those resources that best provide sustainable competitive advantage

(Grant 1996a). Especially changing environment gives rise to knowledge-based resources

(Miller and Shamsie 1996).

Firm’s capabilities define what types of outputs it can achieve with its inputs (Winter 2000). As

discussed, there exist a number of different types of capabilities. There has been varying views

of which of them are the most valuable for a firm. Terms of core capability (Leonard-Barton

1992) and distinctive capability (Day 1994) are used to describe organization’s unique skills and

knowledge. Dynamic capabilities (Teece et al. 1997) are those that make organizational change

possible. Technological capabilities (Figueiredo 2002; Mayer and Salomon 2006) are needed for

managing technological change and for satisfying customers’ needs. Earleir discussion proposes

that a firm’s technological capabilities are formed by its different types knowledge. That is, a set

of different capabilities.

A set of capabilities form a company’s capability portfolio (Laamanen and Wallin 2009).

Because any multidimensional constellation of conceptually distinct characteristics in a firm

can be thought of as a configuration (Meyer et al. 1993), a capability portfolio is conceptually

one type of such. Firms adapt to their changing environment by reconfiguring their capability

portfolios (Lavie 2006). That is, firms look for better fitness with their environment by

changing their configuration. This is also called searching (Levinthal 2000), and experimenting

with alternatives (Teece et al. 2000). Capabilities are very path-dependent in their nature

(Helfat and Peteraf 2003). That is why capabilities cannot be developed overnight. To build

strong capabilities in an organization they must be accumulated through experience over time

(Zollo and Winter 2002). In other words, they must be learned by doing (Pisano 2000).

155

One problem of empirical capability research has been in identifying and especially in

measuring capabilities for several reasons. In addition to finding the right measures, another

difficulty is in identifying proper data. In this framework management’s attention is seen as a

manifestation of capability development. Managers’ decisions are one important internal

selection mechanism that focuses attention to some particular routines and capabilities

(Cohendet and Llerena 2003).

The following framework is built based on the literature review. Management’s attention drives

organization’s search. It affects which capabilities firm prioritizes at different points of time

and which it develops less. Management continuously receives feedback from its operating

environment and redresses the development of capabilities accordingly. The more an

organization puts effort on some capabilities, the more it accumulates experience and thus

‘becomes better’ in them. On the other hand, organizations can also forget as capabilities are

remembered by doing (Nelson and Winter 1982). Therefore consistent development is in a

central role. Effort in one capability means less development possibilities for other capabilities

because of firm’s scarce resources. The core of the framework can be summarized in Grant’s

and Winter’s words:

“in the context of initial learning of a capability, there is generally no clear-cut or automatic

answer to the question of when an organization should be expected to cut back its learning efforts

and affirm that the desired capability has been achieved - - - learning is understood as being

marked by observable allocation of attention and resources to the task of acquiring the

capability” (Winter 2000:981)

“To the extent that capabilities are learned and perfected through repetition, capabilities develop

automatically through the pursuit of a particular strategy.” (Grant 1991:132)

156

Figure 3 Theoretical framework of the study

5 RESEARCH METHODOLOGY

This chapter begins the empirical part of the study. First, the sample companies are introduced

briefly and after that the data and how it was collected are presented. Thereafter research

methods used for analyzing the data are introduced.

5.1 Sample

The sample consists of eleven companies from the Finnish forest cluster. The included

companies are shareholders of Finnish company Forestcluster Ltd. Although universities and

other institutions are associates of Forestcluster Ltd, only companies were included since other

institutions are too different to be compared with companies. Next, the sample companies are

introduced briefly.

Andritz Group

Andritz was founded in Austria in 1852 as a foundry and machine facility. At present, company

produces customized plants, process technologies, and services for hydropower stations, for the

pulp and paper industry, the metals industry, and other industries. The company’s

headquarters are still in Austria and it has around 13700 employees over the world. Andritz has

over 150 production sites in addition to its service and sales companies. The company has five

157

business areas (hydro, pulp & paper, metals, environment & process, and feed & biofuel) of

which pulp and paper and hydro are the largest ones in terms of sales.

Ciba

Switzerland-rooted Ciba was founded in 1884 as a chemicals company. The roots of the present

Ciba reach a bit further to 1758 when a Swiss chemicals company Geigy Ltd. started with

chemical and dye products. After several mergers and new names Ciba Specialty Chemicals was

spun off from its parent company and got a name Ciba Inc. Eventually in 2009 Ciba became a

part of BASF which has about 330 production sites and 97000 employees globally. Ciba itself

has around 12500 employees and 59 production sites in 120 countries. Nowadays Ciba’s business

is divided into three areas: plastic additives, coating effects, and water & paper treatment.

Kemira

Kemira was founded in 1920 as the state sulphuric acid and superphosphate plants. Originally

Kemira supplied products for state’s powder plants and agroculture but a couple of decades

later it started producing also chemicals for Finnish industries. In the latter part of the 20th

century Kemira entered new fields through acquisitions and mergers and started to

internationalize heavily. At present, Kemira is divided into five units. Paper, Water and Oil &

Mining are focusing on water and fiber management. Tikkurila is focused on paints and

coatings business and a new ChemSolutions unit concentrates on organic acids and salts.

Kemira operates in 40 countries and has around 9400 employees.

Metso

Metso was established in 1999 when Valmet and Rauma were merged. Valmet produced paper

and board machines and Rauma concentrated on fiber technology, rock crushing and flow

control solutions. The history of Valmet roots back to 1750’s when a shipyard for building

warships was established in Suomenlinna in Helsinki. Valmet and Rauma-Raahe were born in

1940’s when several state-owned metal workshops were merged into Valmet and multiple

sawmills into Rauma-Raahe. In addition to paper machines, Valmet’s products have included

ships, aircraft, weapons, locomotives, tractors, marine engines, and elevators. In 2008 Metso

got a new organizational structure which consists of three segments: Mining & construction

technology consisting of mining and construction business lines, Energy & environmental

technology consisting of power, automation and recycling business lines, and Paper & fiber

technology which consists of paper, fiber, and tissue business lines. Metso has nearly 30000

employees over the world of which nearly one third is from Finland.

158

Metsä-Botnia

Metsä-Botnia was founded in 1973 but its oldest part was founded over 110 years ago. The

company has had several large owners including Metsäpohjanmaa, Nokia, Serlachius, Metsä-

Serla, and UPM. Today Metsä-Botnia is a subsidiary of Metsäliitto which owns, after recent

arrangements, most of the shares of the company. Its other noteworthy owner is UPM-

kymmene. Metsä-Botnia is one of the world’s top producers of chemical pulp with nearly 2000

employees.

Metsäliitto

Metsäliitto was founded in 1934 by forest owners in order to coordinate timber sales to

customers outside Finland. In the beginning of 1950’s Metsäliitto entered the mechanical forest

industry by establishing its own sawmills, and a couple of years later the company moved into

the chemical forest industry when it started to manufacture pulp, paper and board. A few

decades later Metsäliitto Group was started, formed by Metsäliitto Cooperative, Metsä-Serla,

Metsä-Botnia and Finnforest. Metsäliitto group is now one of the largest forest industry groups

in the world with more than 16000 employees in about 30 countries. It has five business areas:

wood supply, wood products, pulp, board & paper, and tissue & cooking papers. Its parent

company Metsäliitto cooperative has 130000 members and owns approximately half of the

private forests in Finland.

M-real

M-real is one of the leading producers of paperboard in Europe and a supplier of paper. Its

customers vary from printing houses to brand owners. Until 2001 the company was known by

the name Metsä-Serla Oy, which was established as a merger of Metsäliiton teollisuus Oy (one

branch of Metsäliitto cooperative) and Serlachius Oy at the end of 1986. At the same time

Metsä-Botnia became its subsidiary. Oldest roots of the firm go back to 1860’s when Serlachius

Oy’s first groundwood plant was started. M-real consists of four business areas. Consumer

packaging is responsible for supplying paperboard, specialty papers and packaging services.

Office paper business area produces uncoated fine papers. Other papers -business area is

focused on specialty papers. Market pulp & energy -area mainly sells pulp but also energy. M-

real has around 6500 employees.

Ponsse

Ponsse has the shortest history of the sample companies and is the smallest company of the

sample in terms of personnel. Its first tractor was taken into use not earlier than in the late

159

1960’s. Ponsse’s first forest tractor was made for inventor’s own use but quickly he started

supplying tractors also for others and set up Ponsse Oy in 1970. 1980’s and 1990’s were times of

growth for now internationally operating Ponsse. The company operates approximately in 40

countries and has around 1500 employees. Ponsse’s products include forest harvesters, control

systems, and services.

Stora Enso

Stora Enso was born as a merger of Swedish Stora (until 1984 Stora Kopparbergs Bergslag) and

Finnish Enso (former Enso-Gutzeit) in 1998. Out of all sample companies Stora Enso has the

longest history. Earliest traces of Stora Kopparberg’s mines in Falun can be found from the 13th

century. The first paper and sawmills were built in the latter half of the 18th century, and at the

mid-19th century company supplied its first paper machine. Paper and pulp production grew

together with Stora’s iron and steel production. In 1978 Stora left its iron business and focused

on energy and forest industries. Stora and Enso grew with several acquisitions and mergers,

and finally with large ones as they acquired Holtzmann in 1998 and Consolidated papers in

2006. In 2008 Stora Enso consisted of six business areas: newsprint & book paper, magazine

paper, fine paper, consumer board, industrial packaging, and wood products. Stora Enso is the

largest company in the sample in terms of personnel and sales.

Tamfelt

Tamfelt's history can be traced back to 1797 when a woolen and worsted mill was established in

Jokioinen. In 1882 Tamfelt took the first step from consumer products to paper industry when it

started producing felts for paper machines. Focus changed permanently when in the 1960's the

machine felt division exceeded the sales of woollen mill for the first time. In 1965 company

started producing filter fabrics and since then has focused on technical textiles. In 1981 the

company switched its name from Tampereen Verkatehdas Oy to Tamfelt Corp. Today Tamfelt

consists of Tamfelt PMC which produces paper and board machine clothing and Tamfelt

filtration whose products are filter fabrics, for example, to the forest, mining and chemical

industries. Tamfelt has around 1500 employees and is the smallest company in the sample in

terms of sales.

UPM

UPM-Kymmene was started in 1995 when Kymmene Oy and Repola Oy were merged.

Company's roots go back to 1870's when its first mechanical pulp, paper and sawmills were

started (although now its oldest mills are from the 15th century). At present, UPM consists of

160

three business groups. Energy and pulp group consists of pulp mills and hydro power assets

and is also responsible for wood sourcing, biofuels business, and some other products. Paper

group produces different types of papers such as newsprint and magazine papers. Engineered

materials group produces self-adhesive label materials and, for example, RFID technology.

UPM has nearly 25000 employees of which about half in Finland.

Figure 4 Personnel and sales of the sample companies (2008)

Sources: Corporate annual reports

5.2 Data and data collection

The data used in the study were corporate annual reports. Publicly listed companies are obliged

to publish annual reports accessible for investors. Therefore data used in the study is public

information. Law obligates firms to communicate correct information, and also that companies

must make sure that all the up-to-date information which are relevant for investors are present

in annual reports.

Annual reports have been used as a material in several business studies before. Almost a decade

ago at least 70 studies that used annual reports were identified (Stanton and Stanton 2002).

The subjects of the studies vary a lot, including the relationship between companies’ risk and

return (Bowman 1984), correlations between CEO letter to shareholders and firm performance

(Kohut and Segars 1992), relationship between communications content and stated mission

objectives (Clarke 1997), customer satisfaction (Jones 2006), and environmental disclosures

0

5000

10000

15000

20000

25000

30000

35000

Personnel

Sales (MEUR)

161

(Neu et al. 1998). Bowman (1984) proposes that content analysis of annual reports can be of real

usefulness for understanding issues of corporate strategy and can serve as primary or

supplementary source of information. Sierilä (1989) argues that annual reports are too often

neglected in strategy research.

Also studies about managers’ cognition have used annual reports as material. Kaplan, Murray et

al. (2003) used word count measures in assessing managers’ recognition and response in

technological discontinuity. Osborne, Stubbart et al. (2001) used computer assisted content

analysis in their study related to cognitive strategic groups. Recently, Eggers and Kaplan (2009)

measured top management’s attention towards capabilities using CEO letters from annual

reports. In their study discussing managerial cognition and organizational renewal Barr et al.

(1992) acknowledge that annual reports may not be the ultimate data source but few rival

sources that provide insight into changing mental models of managers exist. They note that

annual reports are too important not to be given close attention by the top management in

terms of subject framing and word level editing. One of the strengths of annual reports is also

that they are suitable for longitudinal studies because they are published in consistent

intervals.

The annual reports were downloaded from companies’ websites in PDF-format, except one

which was in webpage format only but the content was the same as in the version released on

paper. Especially in the beginning of the millennium, companies released business and

environmental reports separately. In these cases they were merged into one. Altogether the

number of used annual reports was 101 from 11 companies, meaning that nine reports were not

available from the observation period. Six companies had all the ten annual reports and the

least was seven from one firm.

162

Table 3 Annual reports in the data

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Andritz ● ● ● ● ● ● ● ● ● ●

Ciba ● ● ● ● ● ● ● ● ● ●

Kemira ● ● ● ● ● ● ● ● ● ●

Metso ● ● ● ● ● ● ● ● ● ●

Metsä-Botnia ● ● ● ● ● ● ● ● ●

Metsäliitto ● ● ● ● ● ● ● ●

M-real ● ● ● ● ● ● ● ●

Ponsse ● ● ● ● ● ● ● ● ● ●

Stora Enso ● ● ● ● ● ● ● ● ● ●

Tamfelt ● ● ● ● ● ● ●

UPM ● ● ● ● ● ● ● ● ●

5.3 Research methods

Two research methods are used in this study. Content analysis is used for measuring attention

for capabilities, and self-organizing map is used for analyzing the quantitative content analysis

data. These two methods are introduced next.

5.3.1 Content analysis as a data analysis method

The concept of content analysis has had several meanings over time. According to Krippendorff

(2004:19), it was originally defined as an “objective, systematic and quantitative description of

the manifest content of communication” and was used as a quantitative research method as

early as in the 1890’s for analyzing newspapers’ content (ibid.). Later on, content analysis has

been seen not only as quantitative, but also as qualitative method for analyzing different kinds

of, usually textual, data. What has remained as a typical property of the definitions is that

content analysis is seen as systematic and objective way of analyzing data (Kyngäs and

Vanhanen 1999). What should be pointed out is that content analysis is a technique for analyzing the

data but in the end it leaves the discussion and conclusions to the researcher (Tuomi and Sarajärvi

2009).

163

Using computers for automated content analysis has increased in popularity (Riffe et al. 1998).

At the simplest, content analysis is done by counting frequencies of words in the data but

better information can be retrieved with more advanced techniques where for example words’

contexts are examined or their relationships explored (Riffe et al. 1998). Perhaps the biggest

advantage of computer-aided analysis is the speed in which computers can go through large

amounts of data and if done properly, computers are reliable yielding always the same

information out of the data, and thus fulfilling the requirements of content analysis as being

‘systematic and objective’. This kind of data gathering is also called text data -mining, which is

one form of data mining. Data mining has been defined as extraction of implicit, previously

unknown and potentially useful information from data (Witten and Frank 2002).

As mentioned above, content analysis has had history both as qualitative and quantitative

research technique. Quantitative content analysis, either manual or automated, has been a

popular method for analyzing textual data - for instance in the field of business. Nearly all of

the studies listed in the preceding chapter using annual reports as material also used

quantitative content analysis as a research technique. Recently, these two approaches have

generated ‘a third movement’ in the form of mixed-methods research (Janasik et al. 2009).

Already fifteen years ago artificial intelligence software was used for analyzing content analysis

data in psychological studies (Gottschalk 1994). Janasik et al. propose that a technique called

‘self-organizing map’ can be a useful tool for analyzing qualitative (textual) data, by improving

the quality of inferences of the researcher and by providing a relatively objective approach. This

technique is introduced next.

5.3.2 Self-organizing map as a method for visualization

Self-organizing map, also called Kohonen map, is one of the most popular neural network

methods (Kohonen 2001). It is an algorithm that originates from artificial neural network

research, which studies algorithms that have analogies with the functioning of brain (ibid.).

What makes SOM special compared to other neural network methods is that it performs

unsupervised training (Kiang and Fisher 2008).

The self-organizing map (SOM) uses input data to produce a two-dimensional map as an

output. Input data consists of samples with n numerical attributes. The samples can therefore

be thought as vectors in n-dimensional space and attributes as the lengths of component

vectors. The output map consists of so-called nodes, also referred as neurons (see figure 6).

Node’s shape is usually hexagon but they can also be squares. Each node on the map has an

164

associated prototype vector (also called reference vector) so that the vectors resembling

each other are placed on the nodes near each other and vice versa. The nodes which are near

each other are called neighbors. The shape of the map is pre-defined so that every node on the

map has a unique static place and equal size. (Vesanto 2002)

Teaching the map happens in the following manner (Kohonen 2001; Janasik et al. 2009). First,

the grid of prototype vectors is initialized. This is done by assigning values for the reference

vectors. Random values are typically used if only little is known about the input data. Next, the

handling of input data is started. Sample vectors are drawn randomly from the input data set

and taught iteratively to SOM one by one.

Teaching starts by searching the best matching unit (BMU) for the sample. It is done by feeding

an input data vector (where i is the input data vector index) to all the units of the map and

by calculating ‘similarity measures’ between it and reference vectors (j is the map unit

index). Typically the metric used for similarity is Euclidean distance. For vectors

and the Euclidean distance can be defined as (Kohonen 2001):

The sample is associated with the most similar node, that is with closest to . The best-

matching unit (BMU) denoted by for an input vector is therefore:

where t is the training step index.

After the BMU is found, it and its neighbors’ prototype vectors are updated to be slightly closer

to the sample vector in the input space:

where is a learning-rate factor and is a neighborhood function.

Thereafter, the process of searching the BMU and updating are done iteratively for all the data

samples. Finally, the map stabilizes with the input data samples placed on the suitable nodes

(Janasik et al. 2009).

165

Self-organizing map can be visualized in many different ways (Vesanto 2002). U-matrix, PCA-

projections, visualizations of component planes, bar charts, and data histograms are usual ones.

In this study U-matrix (Ultsch and Siemon 1990) and the component planes are used. This is

because they are suitable for making inferences about similarity of the samples, samples’

properties, and relationships between variables. The component planes illustrate the values of

different variables in different parts of the map. U-matrix illustrates the distances between the

nodes, and that is why U-matrix includes extra nodes between neighboring nodes. There is only

one U-matrix for one SOM. If the data is such that it includes similar enough samples that can

be grouped into clusters, clusters can be found from the U-matrix. This is because in the areas

where values in the U-matrix are relatively low the nodes are closer to each other (resemble

each other), whereas high values differentiate areas from each other.

Figure 5 Visualizations of SOM

Figure 6 shows a few visualizations of SOM. This map consists of 24 hexagonal nodes but in

maps with larger number of samples also the number of nodes is usually larger. A map without

colors, where the closest neighborhood of d is marked, is used only to illustrate samples’

positions on the map (above the positions of samples a, b, c and d are marked to all the

visualizations). Component planes show variables’ values in different parts of the map. In this

map red represents high values and blue represents low values. For example in the right-down

166

corner where sample c is positioned, variables 1 and 2 have low values whereas variable 3 has

high values. What also can be seen from the component planes is that variables 1 and 2 are

fairly well correlated because those component planes resemble each other. The visualization

with bars is in line with component planes. Bars represent nodes’ prototype vectors. As

discussed, in U-matrix colors are used to illustrate the distances (similarity or dissimilarity)

between nodes. This is why U-matrix includes extra nodes between neighboring nodes. As can

be seen, the middle of the map has relatively large values in U-matrix. This means that the

samples in the upper part of the map are relatively much differentiated from the samples in the

bottom of the map. Red color in the U-matrix also illustrates that compared to d the sample b is

further away from a, although in other maps it would seem to be otherwise.

There have been a vast number of studies in different fields using self organizing maps as a

visualization method varying from gene research (Chen et al. 2001) to face recognition

(Lawrence et al. 1997). Self-organizing maps have been used also for studying contents of

annual reports (Back et al. 2001). As already mentioned, SOM can be used for exploratory

research by analyzing and categorization of textual data (Janasik et al. 2009).

Two-dimensional maps have been used before in business studies for different types of

purposes. Stuart and Podolny (1996) discuss the evolution of organizations’ technological

capabilities and the localness of companies’ search based on patenting data. They draw two-

dimensional maps derived from traditional multidimensional scaling techniques. Lamberg et

al. (2009) used PCA-projections for illustrating firms’ strategic consistency. This study with

self-organizing maps resembles those in the sense that idea is to show firms’ positions and

movement on the map. One of the most common uses of SOM is searching clusters of similar

samples from the data. In this study the primary focus is on changes in firms’ capability

portfolios and illustrating them with SOM. Classifying the sample firms into groups would be

welcomed addition if clusters can be found.

167

6 DATA ANALYSIS

This chapter describes the whole process of how the analysis was done in practice from content

analysis to data preparation and creation of self-organizing maps. A thorough example is used

for elucidating the process. Figure 7 illustrates the process which Vesanto (2002) proposed for

data exploration process based on self-organizing map.

Figure 6 Applying the SOM in data mining (Vesanto 2002)

6.1 Content analysis process

The downloaded annual reports were first opened in Atlas.ti-software. The word cruncher

function of Atlas.ti was used for counting the word occurrences from annual reports. Word

cruncher’s stoplist was enabled in order to avoid miscalculations because some special

characters were counted as words without the stoplist. Atlas.ti exported the data into an Excel-

file.

After the word frequencies for all the words in annual reports were listed, they were

categorized to represent organizational routines (forming capabilities). Initially the number of

different words in annual reports was 35190. The words which occurred most were left out first.

These words included the most common words such as “the, that, it, of, etc”. Also those words

which occurred hundred times or less were left out (except country names, see below), which

means that the total number of occurrences for included words was more than the number of

annual reports. These steps were done because the most common and the least common terms

are typically not good candidates for terms (Janasik et al. 2009). Next, the rest 2016 words were

gone through manually. The words were categorized into groups according to what capabilities

they can represent. Because the capability literature was already familiarized, the groups were

similar to what had been found from the literature (see table 1). Those words were left out

168

which were meaningless regarding capabilities. These words included words such as ‘during’

and ‘January’. Eventually 359 keywords and their derivatives were used for describing

capabilities. For example efficien* includes derivatives efficiency, efficient, efficiently, and so on

(asterix indicates that word includes sequence of letters in front of it and the asterix can be

replaced with any letters). The frequency for every capability was counted as a sum of the

occurrences of its keywords.

All country names were included as keywords for internationalization capability although some

of them appeared less than 100 times. Occurrences for country names were counted from the

initial list of words (with 35190 words) using excel’s countif-function. The list of world countries

for the countif-function was downloaded from Gapminder’s website

(http://www.gapminder.org/wp-content/uploads/2008/11/formal-list-of-areas-0807011.xls).

6.2 Preparing the data and building self-organizing maps

The process of using self-organizing maps for text mining followed the way which has been

proposed by Janasik et al. (2009). Authors divide the process into six steps:

1. Select a document collection.

2. Automatically or manually choose the terminology to be used in encoding the

documents.

3. Transform, based on the terminology, the documents into numerical data.

4. Initiate the SOM iterations with varying parameters.

5. Study and interpret the resulting document maps and term distributions on the

maps.

6. Formulate the inferences that can be drawn from the maps.

Results and inferences are discussed later in this study, and the first steps were already taken in

the content analysis phase. These were choosing the document collection, namely annual

reports, and choosing the terminology to be used, which was done by choosing the words from

the list Atlas.ti produced (categorization of words). In that same list the terminology was

already in a numerical form, but this step differed slightly, since it was groups of words

representing capabilities rather than individual words that were used as dimensions for

initiating SOM. Therefore this phase also included summing the words in the groups.

169

The following example shows the whole process of building self-organizing maps following the

content analysis phase, in which the words were categorized and their frequencies were known.

Instead of the full data of 101 annual reports and 15 capabilities, three samples and three

variables are used for better elucidation.

Table 4 Word frequencies for capabilities as sums of their keyword occurrences

Capability x Capability y Capability z

Keyword

Keyword

Keyword

⁞

Keyword

Keyword

Keyword

Keyword

⁞

Keyword

Keyword

Keyword

Keyword

⁞

Keyword

Frequency for

capability x

Frequency for

capability y

Frequency for

capability z

Table 5 Word occurrences in the example

Word occurrences Capability x Capability y Capability z

Annual report 1 15 30 45



Figure 7 Absolute word occurrences in the example

0

10

20

30

40

50

60

Annual

report 1

Annual

report 2

Annual

report 3

Wo

rd o

ccu

rren

ce

Capability x

Capability y

Capability z

170

The preparation of the numerical data included counting the relative occurrences of

capabilities, so that the total attention (100 %) was divided to capabilities according to the word

occurrences. Laamanen and Wallin (2009) used this approach to illustrate attention allocation,

and similar normalization was recommended by Janasik et al. in their paper concerning self-

organizing maps and text-mining. This was done in order to make sure that different lengths of

the annual reports do not distort the maps, and because the interest was in the relative

attention given for the chosen capabilities. For example in figure 8 the smaller number of

occurrences in annual report 2 could be explained with annual report’s shorter length. Table 5

shows how the attention was divided in the example annual reports. Figure 9 illustrates this

graphically and it can be seen that annual reports’ capability profiles remain unchanged but the

differences between their lengths are evened up.

Table 6 Attention for capabilities in the example

Capability x Capability y Capability z

Annual report 1 17 % 33 % 50 %



Figure 8 Illustrations of attention allocation

171

In practice the self-organizing map was built in the following way. Data was gathered into a

table-format. Vesanto (2002) used figure 10 to illustrate how the data should be arranged in

Excel and MATLAB. SOM toolbox for MATLAB (http://www.cis.hut.fi/projects/somtoolbox/)

was used for finding the best-matching units for samples, teaching the map, and graphical

representation (in addition, basic photo editor software was used for black and white maps

representing annual reports’ positions in chapters 8.2.1 and 8.2.2).

Figure 9 Table-format data used for SOM (Vesanto 2002)

Figure 11 illustrates how the capability profiles in the example would position to best-matching

units in a 42-node SOM created with random values.

Figure 10 Best-matching units (BMUs) for example profiles

172

7 RESULTS

This chapter includes the results from the content analysis phase and the self-organizing maps.

The categorized keywords are listed first, and thereafter self-organizing maps are presented.

That is followed by inferences in which the matters behind the figures are opened up. And last,

answers to the research questions are summarized.

7.1 Content analysis results

Categorization resulted in fifteen groups of keywords describing capabilities (see tables 1 and 2

for links to existing literature). The number of words that were grouped totaled 245 and their

derivatives, and 108 country names. Keywords are listed in table 6 and word frequencies for

groups are listed in the appendices. Table 7 shows the countries which were found from annual

reports to represent internationalization.

Table 7 Keywords for capabilities

FINANCE

Account* Alloca* Borrow* Capital* Cost* Currenc* Debt* Depreciat* Discount* Dividend* Earning* Equit* Expen* Financ* Fund* Income* Invest* Liquidity Loan* Money Profit* Receivab* Revenue* Saving* Securit* Sharehold* Tax*

RESPONSIBILITY

Accident* Audit* Award* Carbon* Certif* Climat* Damag* Dispos* Emission* Environ* Ethic* Gases Green* Guideline* Health* Law* Legal* Legislat* Oxygen* Polic* Principl* Regulat* Responsib* Safe* Social* Societ* Stakehold*

STRATEGY

Action* Aim* Capab* Choice* Competen* Core Decid* Decisi* Direction* Focus* Future* Goal* Growth Implement* Intend* Intent* Key Mission* Objectiv* Organic* Perform* Plan Planned Planning Portfolio* Position* Primar*

MONITORING

Analy* Anticip* Asses* Believ* Challeng* Circumstan* Compar* Competit* Condition* Crisis Declin* Demand* Economy Estima* Evaluat* Expect* Favorab* Fluctuat* Forecast* Identify Industr* Investig* Market Markets Monitor* Opportun* Potential*

INTERNAL

DEVELOPMENT

Bonus* Compensat* Cultur* Educat* Employ* Expert* HR Human* Incentiv* Intang* Intellect* Job* Know* Labor Labour Learn* People* Person* Profession* Recruit* Retire* Reward* Salar* Secretar* Skill*

173

Turnover* Unallocated

Sustainab* Transparen* Waste*

Priorit* Risk* Seek* Strateg* Strength* Target* Vision*

Projected Recogni* Sector* Situation* Studies Study Trend* Uncertain* Volatil*

Specialist* Staff* Task* Team* Train* Wage* Worker* Workplace* Workforce*

MANAGE-

MENT

CEO Executive* Leadership Manag* MBA Senior

EXTERNAL

SOURCING

Acqui* Agreemen* Associate Associates Collabo* Contract* Coopera* Counterpart* Joint* Lease* Merg* Negotiat* Outsourc* Partner* Relation* Universit* Ventur*

MARKETING & SALES

Advertis* Brand* Consumer* Customer* Marketing* Sale* Sell* Sold Product Products

PROCUREMENT & LOGISTICS

Chain* Deliver* Distrib* Inventor* Location* Logistic* Procur* Purcha* Sourcing Transport* Shipp*

STRUCTURING

Department* Divest* Division* Integrat* Parent* Restruct* Struct* Subsid* Synerg* Unit Units

INTERNA-TIONALIZATION

Country names Geograph* Countr* Internatio* Global* World*

OPERATIONS

Effective* Efficien* Maintenance Manufactur* Optimi* Produce* Producti* Quality Utilis* Utiliz*

SERVICES

Servic*

CHANGE

Adjust* Chang* Flexib* Prevent* Proactiv* Renewal* Renewed Respond*

INNOVATIVENESS

Innovat* Patent* R&D Research*

174

Table 8 Countries in annual reports

Algeria Argentina Australia Austria Bahamas Bahrain Bangladesh Barbados Belarus Belgium Belize Bermuda Bhutan Bolivia Brazil Bulgaria Cameroon Canada Chile China Colombia Croatia Cuba Cyprus Czechoslovakia Denmark Ecuador Egypt

Estonia Ethiopia Finland France Georgia Germany Ghana Greece Guatemala Guinea Guyana Haiti Honduras Hungary Iceland India Indonesia Iran Iraq Ireland Israel Italy Jamaica Japan Jersey Jordan Kazakhstan

Kosovo Kuwait Laos Latvia Lebanon Liechtenstein Lithuania Luxembourg Malawi Malaysia Malta Mexico Mongolia Montenegro Morocco Namibia Nepal Netherlands Nicaragua Nigeria Norway Oman Pakistan Panama Paraguay Peru Philippines

Poland Portugal Qatar Romania Russia Serbia Singapore Slovenia Spain Sudan Suriname Sweden Switzerland Syria Taiwan Tanzania Thailand Turkey Ukraine Uruguay Uzbekistan Venezuela Vietnam Yugoslavia Zambia Zimbabwe

Note: Because individual words were counted, multi-part country names were not found. Because of this, for example, United States was not found. On the other hand occurrences for ‘Jersey’ resulted from ‘New Jersey’ where Ciba has a production site and M-real has a subgroup M-real New Jersey Service Co.

175

7.2 Self-organizing map

This chapter presents the self-organizing map based on the content analysis and data

preparation discussed in the preceding chapter. First, a label map that MATLAB produced

indicating annual reports’ positions is presented. That is followed by maps in which firm-

specific positions are marked. Also year-specific maps are presented. Last, U-matrix and the

component planes illustrating capabilities are shown.

Figure 11 Self-organizing map showing labels

AN = Andritz

CI = Ciba

KE = Kemira

MB = Metsä-Botnia

ME = Metso

ML = Metsäliitto

MR = M-real

PO = Ponsse

SE = Stora Enso

TA = Tamfelt

UPM = UPM-kymmene

Positions by firm

Next maps show annual reports’ positions for every firm. They are derived from the preceding

map. Arrows represent changes in firms’ positions year after year. If a node includes more than

one dot, several annual reports for that firm are positioned in that same node indicating similar

configurations. Nodes that are next to each other also represent similarity but the level of it

cannot be interpreted without the U-matrix (see chapter 6.3.2 for reasoning).

176

Figure 12 Positions of annual reports by firm

Positions by year

These maps show annual reports’ positions year after year without paying attention to what

firms they are. Arrows represent firms’ movement by indicating where they have moved from

to their current position.

Figure 13 Positions of annual reports by year

177

What comes to classification by year, only little collective movement can be found in 1999-

2008. In the beginning and at the end of the period firms tend to position in the upper part of

SOM. These periods involve much attention for financial issues, change, external sourcing, and

structuring. In the midway of the period the dominant area is in the bottom part of SOM which

is dominated by attention for innovativeness, internal development, responsibility, strategy,

management, and marketing.

7.2.1 U-matrix and component planes

The component planes show variables’ values in different nodes of SOM. Red indicates higher

values (relatively more attention) whereas blue represents lower values. U-matrix indicates

distances between the nodes.

Figure 14 U-matrix and component planes

178

7.3 Interpretations of the maps

As discussed in the literature part of the study, a firm is able to achieve a high level of capability

accumulation with consistent development. Therefore, the main interest is in which

capabilities have received relatively much attention consistently. If a company has retained its

position on the map, its portfolio’s focus has been consistent but movement does not mean that

development of individual capabilities’ could not be consistent – as long as firm moves to an

area representing high attention. Firm-level

inferences are presented next. Figure 18 shows an

example how Andritz has given consistent

attention for services. In 1999 firm’s attention for

services was small but for the next nine years the

attention was high. For simplification the levels of

attention are from now on classified as low,

moderate, or high.

Figure 15 Andritz’s change of emphasis on service capability

Andritz

BEING STRATEGIC “…overall strategic goal of the Andritz Group...” “… the strategic direction is based…“ MONITORING “Market conditions in Austria for the business area’s main product…” EXTERNAL SOURCING “A key factor to this success is our strategy of complementary acquisitions…”

High attention Being strategic Monitoring External sourcing Structuring Internationalization Operations Services

Moderate attention Marketing & sales Innovativeness

Andritz has developed consistent portfolio of capabilities since 2000. Orientation

towards innovativeness has been moderate but increasing. Company’s R&D

expenditure tripled in six years from 1994 to 2000, and after that the attention for

innovation has just been increasing. Andritz’s innovativeness is manifested in applied

patents and pilot testing. According to Andritz, all of the development has not been

directed to product innovation, but innovation for own processes and systems has

been important as well. Company has set up research facilities and pilot plants over

the world as internationalization has been heavily present in the recent years. Andritz

aims to be, in its words, ‘globally present’. It has service centers worldwide and has

179

sought for foothold especially in the world’s growth areas. All of Andritz’s business

areas have consistently expanded their aftermarket services which are now an

essential part of the company’s business. For example, in 2000 one fourth of total sales

were services and most of the growth came from them. What comes to structuring,

year 1999 saw changes in group's structure following the change of ownership as

Andritz AG was integrated into its affiliate Andritz Internationale Technologie AG.

And later, acquisition of Ahlström involved restructuring as target company’s

structure was aligned with the acquirer. Andritz declares to actively restructure and

evaluate the productivity of its facilities. Operations development has received

attention also as Andritz emphasizes its quality management system in which Andritz

has allocated resources increasingly in order to cope with company’s growth.

Ciba

BEING STRATEGIC “…effectively links our core competencies with the consumer market…” MARKETING & SALES “…to address the needs of our customers...” INNOVATIVENESS “A new approach to innovation will be driven by targeted resource allocation…” RESPONSIBILITY “…products that help to mitigate the effects of climate change…”

High attention Responsibility Being strategic Monitoring Marketing & sales Innovativeness

Moderate attention Internal development Internationalization Operations Services

Company distinguishes itself from other sample companies with its notably strong

attention towards customers. Over the years Ciba has sought to be more and more

customer-oriented organization. Another thing gathering attention is Ciba's thrive for

product innovation. As a chemicals company Ciba pays particular attention also to

environmental responsibility. This has manifested in e.g. participation in non-profit

organization Carbon Disclosure Project.

180

Kemira

MARKETING & SALES “We adapt to our customers’ changing needs…” INTERNAL DEVELOPMENT “…give our people a greater say in how things are done at our company…” MANAGEMENT “The management development program PEAK addresses strategy…”

Moderate attention Responsibility Finance Being strategic Monitoring Internal development Management

External sourcing Structuring Marketing & sales Internationalization Operations Change Innovativeness

Kemira has had a sort of two modes in its capability development. Inconsistency has

resulted in only moderate attention. Other mode is dominated by structuring, finance

and change whereas other is focused more on internal development, responsibility,

and marketing. In 1999 company announced structural changes as the group

sharpened the focus of its business mix. Similarly, ten years later Kemira announced

structural changes when it decided to focus on water and fiber management

chemistry. In the mid-period Kemira’s attention shifted. Responsible business turned

from necessity into a way to stand out from competition. This was called

‘responsibility-driven growth’. Internal development was also in the focus. For

example, Kemira established a program called ‘Kemira from good to great’ to develop

corporate culture and to promote participation.

M-real

FINANCE “…closures of production capacity, a new cost savings programme, a working capital reduction programme and divestments of assets…” STRUCTURING “We will specify the other divestment targets later…” EXTERNAL SOURCING “M-real is participating in a number of cooperation projects.”

High attention Finance Change External sourcing Structuring

Moderate attention Responsibility Internal development Management Procurement & logistics Internationalization Operations

181

M-real’s financial difficulties have dominated company’s attention for the whole

decade, except in the midway following company’s profitable period. Company started

emphasizing responsible business by forming official guidelines for its

implementation. In 2006 company’s attention turned to finance again, as it reported

its record losses. This was followed by asset divestment and restructuring programs.

Large divestments were made and programs for cost savings were implemented.

Metso

BEING STRATEGIC “According to the strategy, Metso will be developed into…” MONITORING “the market situation is clearly improving…” STRUCTURING “business restructuring of recent years, which have helped to streamline the cost structure…” MANAGEMENT “The Corporation runs two programs for developing general management capabilities…”

High attention Being strategic Monitoring Services Management

Moderate attention Responsibility Internal development External sourcing Marketing & sales Internationalization Operations Innovativeness Structuring

Metso was born at the beginning of the observation period from a merger of Rauma

and Valmet. Therefore the first year involved structuring, which has been visible also

later e.g. in the form of Metso Automation’s and Metso Mineral’s restructuring in year

2006. Services have received attention as they have been a significant part of Metso’s

business. Newborn company’s strategy emphasized services from the start when

company launched its service concept ‘future care’. Nearly ten years later half of

Metso’s operating profit came from services. At times Metso has given relatively much

attention to internal development and management, which have become concrete in

company’s own management programs - in internal doctoral program ‘Metso

Academy’ and in ‘Metso Forum’ which is meant to help cooperation between

personnel and management.

182

Metsä-Botnia

RESPONSIBILITY "The principles of corporate social responsibility apply to all of Botnia’s activities…” INTERNAL DEVELOPMENT “We provide our key personnel with training in all matters relating to strategy implementation…” PROCUREMENT & LOGISTICS “The cornerstones of Botnia’s competitiveness are - - - and optimized logistics.”

High attention Responsibility Internal development Procurement & logistics Operations

Moderate attention Being strategic External sourcing Management Marketing & sales

Metsä-Botnia stands out from other sample companies with its distinctive attention

towards internal development. Personnel receive plenty of attention, for example, in

forms of training and job rotation. According to the company, it has in most cases

managed to keep a satisfactory level in its personnel surveys. Company emphasizes

responsibility by being one of the best environmental performers in the world, and in

the last years company has considerably increased its environmental investments.

Metsä-Botnia differs from most of the sample firms also in terms of attention given for

logistics. As a pulp producer company declares to put effort on correct and timely

delivery of pulp to its customers with the development of its own logistics planning and

reporting systems. Matters behind company’s distinctive attention towards sourcing

and operations include establishments of programs for improving the process

availability rate, which is one key metric for a pulp producer.

Metsäliitto

INTERNAL DEVELOPMENT “The annual employee questionnaire is also used to gauge progress in job satisfaction…” PROCUREMENT & LOGISTICS “Logistics processes are developed in close cooperation with customers…”

High attention Procurement & logistics Internal development

Moderate attention Finance Responsibility Management External sourcing Operations Change

183

Metsäliitto is one of the least consistent companies in capability development as its

attention has shifted the most. Logistics is naturally in the focus for a wood

procurement organization throughout the whole observation period. Also internal

development has received much attention, for example, in the form of programs

promoting wellbeing at work. Cost-effectiveness have been much in the focus, but in

the recent years marketing has shown signs of rising in priority as company has sought

to take more customer-focused approach. In 2006 company’s CEO identified it as the

next big trend in the forest industry and saw marketing and sales as necessities for their

survival in the future.

Ponsse

EXTERNAL SOURCING “Ponsse has several cooperation partners among forestry institutions around the world…” STRUCTURING “Two new subsidiaries were established during the accounting period.” FINANCE “To maintain financial flexibility and balance seasonal fluctuations, the company uses finance credit agreements…”

High attention Finance External sourcing Structuring Change

Moderate attention Monitoring Logistics & procurement Services

Ponsse’s dominating characteristic is its attention given for external sourcing - mostly

in the form collaborations. For example, in addition to those in Finland, collaborations

have been established in Baltic countries, where Ponsse especially benefited from new

sales networks. In Brazil Ponsse has made contracts with several retailers and other

cooperators to strengthen its position in South America. In the middle of the period

Ponsse started to give increasingly much attention to services. In 2004 Ponsse launched

the world’s first remote diagnostics system for forest machines, and as a part of its

maintenance service development plan the company made large investments in its

service centers around Finland.

184

Stora Enso

BEING STRATEGIC “The very challenging goals the Group has set for itself will be achieved only by…”’ MONITORING “…but uncoated fine papers are suffering from increased competition and the weak US dollar.” STRUCTURING “Another key to raising profitability is our asset restructuring programme…”

High attention Being strategic Monitoring Structuring

Moderate attention Internal development Finance Marketing & sales Internationalization Innovativeness Change Management Operations

The integration of new-born Stora Enso involved structuring at the end of the 20th

century. Structuring took place also when Stora Enso made a major acquisition of US-

based paper producer Consolidated Papers. Attention then shifted more towards

strategy, monitoring and marketing. Company set a goal of building a distinctive brand

for the new company and aimed to increase customer-orientation. Lately Stora Enso’s

attention has shifted towards finance, which can be explained with the on-going

recession.

Tamfelt

MANAGEMENT “In China, much can be done to improve productivity through management actions.” STRUCTURING “…the procedure will make for a uniform Group structure in operative and legal terms…” FINANCE “Regular monitoring helps extend the life of the clothing and to improve cost efficiency.”

High attention Management Structuring Finance Change

Moderate attention Responsibility Being strategic Monitoring Internal development Innovativeness

Finance gets much of Tamfelt’s attention. Company informs that in the beginning of

the millennium for many years its investment expenses were more than industry

average, differing from investments for improved efficiency to environmental

185

improvements. Tamfelt is rather cost-oriented with its target in being the cost-leader of

the industry. Cost efficiency has been pursued with new investments for operations and

waste treatment in addition to moving some operations to countries where cost levels

are lower.

UPM

FINANCE “During the last four years, we have reduced costs significantly and are now better prepared…” EXTERNAL SOURCING “In new growth markets, UPM aims to grow through investments, as well as acquisitions and joint ventures.” STRUCTURING “UPM has focused its activities by investing in its core businesses and by divesting noncore assets and activities”

High attention Finance External sourcing Structuring Change

Moderate attention Monitoring Procurement & logistics Internationalization

A few capabilities stand out in UPM’s capability portfolio. Company’s attention has

been dominated by finance and structuring. UPM declares that it always aims to make

investments in newest technology to continually improve its cost-effectiveness. What

comes to structuring, UPM itself declares that its strengths are in investing in its core

business and divesting its non-core assets and activities, and strong vertical integration.

Although acquisitions were visible for the whole observation period, in the latter part

external sourcing started to receive increasingly more attention.

7.4 Cluster-level inferences

Cluster-level inferences about capability development are presented next. First, some chosen

pairs of capabilities are discussed. Particularly cluster’s attention towards innovativeness is of

interest.

186

Figure 16 Attention towards innovativeness in relation to other capabilities

Figure 19 shows firms’ orientation towards innovation in relation to other selected capabilities.

It can be seen from the top-left matrix how strategic firms are seeking to be innovative whereas

top-right matrix shows how financially-oriented companies tend to be less innovation-oriented.

Especially paper producers are financially-oriented instead of innovativeness - with industry’s

typical focus on cost competition and large investments. When cost-competition eats

resources, companies’ mind-set might turn away from innovation. For example, UPM has

hardly had any major innovations in the period. Some new materials made from surplus

materials have been invented, but innovations have mostly been incremental and related to

processes - for example including pine use for mechanical pulping, river transportation system,

and technical innovations to reduce costs in cutting fiber. M-real at the same spectrum has

only incrementally improved its products for better printability or brighter pulp resulting in

cost savings. At the other end, Ciba is showing the biggest thrive for product innovation. It has

introduced a variety of new products from sun screen products to flame retardants and

187

radically new types of textile dyes. Though, Ciba is operating in rather different industry than

paper companies.

It is also visible that market-oriented companies tend to seek for innovation (bottom-left

matrix). As paper has practically remained the same for decades with only incremental

improvements in it and as customer-relationships with printing houses and publishers have

been fairly stable, paper companies tend to be least market-oriented. The industry might be

‘stabilized’ in a state where customers do not expect any new innovations from paper producers

and on the other hand producers don’t therefore feel a great need to put effort on being

market-driven.

Those firms that put effort on monitoring also seek to be more innovative. Sensing changes in

the market and following competitors can motivate firms towards innovativeness, and on the

other hand those companies that seek for innovation are most willing to sense their

environment. Again, on average paper and pulp companies monitor least actively their

environment.

Figure 17 Inferences about change

What comes to the attention given to ‘change’ (figure 20), a couple of notions can be made.

Firstly, attention towards finance and change have gone much hand in hand, which is also

visible in figure 16 where the two component planes of SOM show correlation with each other

(although they correlate, there is not necessarily causality between them). What could partly

explain this relationship is that generally forest industry is very stable but times of change

become apparent in finance. For instance, in the last years M-real started a major change in its

188

organization, which manifested in asset sales, profit improvement programme, and so on. On

the other hand, those companies that innovate more don’t necessarily feel that great need for

major changes.

Figure 18 Profitability of Finnish pulp and paper industry (Metsäntutkimuslaitos 2008)

Figure 19 Attention for finance, innovativeness and monitoring

Figures 21 and 22 show how firms’ attention towards some chosen capabilities has changed

when financial performance has changed. It is visible how after the most profitable year 2001

companies’ attention shifted away from finance and turned to innovation whereas attention to

monitoring remained unchanged. And when financial performance declined, especially

-20

-15

-10

-5

0

5

10

15

20

25

30

35

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Tota

l pro

fit

(% o

f tu

rno

ver)

40

60

80

100

120

140

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Index (1999 level = 100)

FINANCE

MONITORING

INNOVATIVENESS

189

innovativeness was left aside. Attention towards finance seems to follow performance with a

delay. It started to decrease when profitability was at its highest level and attention towards

innovation started to increase. And when profitability started to decrease, it took a while until

attention towards finance was triggered. It makes sense that firms are focused on finance when

they are performing less well, which leaves less ‘room’ for innovation. At those times also

internal development receives less attention (figure 23) whereas at the better times firms rely

on their own personnel rather than external sourcing.

Figure 20 Internal & external development in relation to innovativeness

8 DISCUSSION AND CONCLUSIONS

This chapter starts with a brief summary of the study and discussion about the research results.

That is followed by discussion about the methodology. The purpose is to give insights to this

relatively new approach and to identify weaknesses and limitations, and offer

recommendations of how similar research could be improved in the future. Also how methods

could be applied for different types of management studies is discussed.

8.1 Summary and results

The main objective of this study was to find out how Finnish forest cluster firms have

developed their capabilities in the last ten years and find links between the development and

cluster’s innovativeness. To be able to do this, another goal was to identify the capabilities to

assess, and to define criteria on how to measure them.

60

80

100

120

140

160

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Index(1999 level = 100)

INTERNAL

EXTERNAL

INNOVATIVENESS

190

So far practically anything that a firm has been good at has been called a capability and

frameworks of generic capabilities or such for identification are still missing. There are no

established methods for empirical research for identifying and measuring capabilities, although

different approaches have been used (Wang and Ahmed 2007). So far most of the studies have

been qualitative one firm case studies (ibid.). Research with larger samples is still rare. That is

not surprise as the concept of capabilities is still much on an abstract level and a bit unclear

even to academics (Dosi et al. 2000; Jacobides 2006). Especially, identifying capabilities that

can act as unique sources of competitive advantage is difficult. Collis (1994) notes that

competitive advantage lies in every single activity that a firm does and thus listing them is

impossible.

Most of the empirical capability research so far has been focused on measuring some chosen

capabilities (Peltoniemi et al. 2010). Dutta et al. (2005) proposed that capability level could be

measured as efficiency of turning inputs into outputs by comparing the achieved level to the

theoretical maximum. This requires that a capability has first been identified. And how can one

define the maximum and can everything be measured with inputs and outputs? The difficulty is

that knowledge and skills are hard to measure. For example R&D capabilities have been

measured with R&D expenditures as inputs and number of patents as outputs (George 2005),

but that data does not tell whole truth as there are also other dimensions of R&D capability

such as quality of products. Capabilities are by definition firm-specific, context-dependent, and

a source of heterogeneity in organizations. That makes them also difficult to compare between

firms.

To overcome these difficulties of measurement and identification, an alternative framework

was built based on the extant literature. Thereafter, the framework was applied in the empirical

part of the study. Capabilities to be included were chosen iteratively in the empirical part based

on capabilities that have been discussed in the prior literature. An objective was to define

capabilities that can be compared across the sample firms. The measurement adopted from

literature for assessing the level of capabilities was attention that is given for them and the data

used for measurement were corporate annual reports. It was assumed in the framework that

high attention in annual reports represents high development.

Even attention for errors does represent development. For example, in the case of Stora Enso’s

acquisition of Consolidated Papers learning took place in the form of trial-and-error learning

(Gavetti and Levinthal 2000; Patel and Pavitt 2000) and thus it can be thought of as

development. Because having a capability should be understood as relative to others (Winter

191

2000; Dutta et al. 2005), other firms’ attention could be used for benchmarking. This should be

kept in mind when interpreting the results. What results tell us is which firms have given

attention more than others. The best firms set the level which interpreted as high. That does

not necessarily imply general superiority (or inferiority) in that capability. That is, results could

look very different if firms were taken out of the sample and compared with other firms. In

addition, attention does not indicate how things are done in organizations. But studying that

was not in the scope of this study.

Fifteen capabilities were mapped at a firm level for eleven firms to be either high, moderate, or

low according to how much they were showing distinctive accumulation compared to others

(D'Este 2002), and some matters behind the attention were presented. Next, answers to the

initial research questions are summarized.

What capabilities have they developed?

The capabilities that emerged from the literature are present in differing levels for different

firms. Altogether fifteen capabilities were identified: financial capability, being responsible,

being strategic, monitoring, internal development, managerial capabilities, external sourcing,

marketing and sales, procurement and logistics, structuring, operational capabilities, service

capability, ability to change, innovative capabilities, and internationalization.

During the period under study capabilities’ relative importance at the cluster level changed. In

the beginning of the period, financial capabilities, change, and structuring were dominating.

But in the beginning of the millennium attention shifted. Firms became more market-oriented,

strategic, and innovation-oriented. What partially explains the shift away from finances is that

this period was the most profitable for pulp and paper industry. At the same time cluster firms

also shifted from external sources of knowledge to internal development. But a few years later

as profitability increasingly declined, companies started to move back to the state they were in

the beginning, with finance, structuring and change as dominating capabilities.

Every company in the sample has its own portfolio of capabilities. Paper and pulp companies

generally stand out as being the most financially oriented. But surprisingly, they don’t

distinguish themselves with attention towards operations. Innovation is typically not in their

portfolio. What connects chemical producers is that they give fairly much attention to

responsibility which is natural as they are probably the ones who are the most confronted by

192

environmental issues. Ciba is perhaps the most different from other firms. It has consistently

distinguished itself with its strong market-orientation, and attention towards strategy and

innovation. What was surprising is that international giants Stora Enso and UPM are not those

who give attention to internationalization the most, but Andritz clearly differs in this sense.

Have they developed capabilities consistently?

The results indicate that firms differ in terms of how consistent they have been in their

capability development. Regarding the whole portfolio, particularly Andritz, Ciba, M-real, UPM

have been consistent ones. Also Metsä-Botnia and Tamfelt have been relatively consistent.

Kemira can be described as semi-consistent as it has operated in two consistent ‘modes’ of very

different portfolios of capabilities.

But being inconsistent regarding the portfolio does not mean that individual capabilities could

not be given attention consistently. These capabilities are indicated by ‘high attention’ in the

results. For example, Metsäliitto has been one of the least consistent regarding the whole

portfolio but logistics have received regularly more attention than in other companies. Out of

all capabilities particularly finance, change, and structuring have been under regular attention

from a number of companies, whereas responsibility has in most cases been under attention

only for shorter periods of time. Overall, forest cluster firms seem to have rather stable

resource positions relative to others.

How are developed capabilities linked to innovation?

At cluster level capabilities’ links to each other were analyzed and three characteristics

promoting orientation towards innovation were identified:

Being strategic

Monitoring the external environment

Being market-oriented

What comes to the first one, it has been argued that strategic management skills boost

innovation (Teece et al. 1997), and that the most important characteristic making a firm

innovative is its strategic posture (Özsomer et al. 1997). Regarding monitoring, Day (1994) has

argued that the two most important capabilities for successful market-driven organizations are

193

customer-linking and market-sensing capabilities. Day’s argument resembles the results listed

above. Dutta et al. (1999) found that marketing capability is one of the most important

capabilities for an innovative firm – especially together with strong R&D. In this study it was

found that firms oriented towards marketing give also more attention to innovation.

A characteristic suppressing innovativeness was also identified. Those firms that are relatively

more concerned about financial issues were less oriented towards innovation, and at the times

when finance is dominating, firms are constraining their innovation. One reason behind this

might lie in ‘efficiency-thinking’. Many of the sample firms are operating in fields where

competition has so far been driven by costs and ever-increasing investments in new equipment

rather than by innovative new products.

To conclude, for an organization to be truly market-driven and innovative, emphasis should be

turned from internal processes and cost-control more to external environment (Day 1994). In

an old industrial district such as Finnish forest cluster customer relationships have been stable

for long and organizations have not necessarily felt a great pressure to be market-driven which

can constrain cluster’s renewal (Grabher 1993).

194

8.2 Weaknesses and limitations

The method of text-mining and SOM proposed by Janasik et al. (2009) is a promising addition

to content analysis and text mining. However, multiple weaknesses were confronted during the

empirical research process.

Automated content analysis was found to include some problems. Some of the chosen

keywords have biased the results because all the keywords’ contexts were not verified in the

data. Problems arise from words with multiple meanings and words in different contexts. For

example ‘personnel’ was interpreted as internal development but it can be questioned if, for

example, laying off personnel is internal development. Or if a company sells its plant it is

difficult to justify its marketing and sales skills with the keyword ‘sold’. What is problematic in

text-mining is that at the same time it is good to be used for large amount of data but on the

other hand verifying the context is difficult if the amount of data is large. The best keywords

are those that have few meanings and are rarely used in other contexts. For example companies

don’t like to mention ‘mission’ or ‘ethic’ in a negative sense. There are words that companies

like to mention a lot. They might claim to be innovative without any concrete proof about it.

Choosing the keywords require exhaustive beforehand knowledge about the companies, data,

and the industry. Emphasis really should be put on forming a good dictionary, and I would

recommend keeping it simple. There are also ready-made dictionaries varying from psychology

to business studies (for example by Harvard University).

Although annual reports are important documents for companies’ management, the language

written in them is not necessarily ‘from their pen’. Considering that written language has a

meaning (Janasik et al. 2009), also who has written the text must have a meaning. Annual

reports are usually written by the public relations department and accepted by the

management. They can also be written by the same people year after year, and they can be

based on the previous years’ annual reports. These factors can have an effect on the language

and can position the annual reports systematically near each other year after year. To better

represent management attention, it would perhaps be wise to use CEO letters in annual reports

for content analysis like Eggers and Kaplan (2009) instead of including whole annual reports.

Including the whole reports might still represent what is called whole organization’s attention

leading to organizational action (Sonpar and Golden-Biddle 2008).

Miller (1996) notes that configurational approaches often have a weakness where changing a

single variable or grouping algorithm can result in totally different results. The same weakness

195

is present in this study concerning self-organizing maps. If a few variables are for one firm

significantly different year after year compared to other firms, its annual reports tend to group

into same area although other variables would not differ that much from other firms. On the

other hand, deviations in variables can cause unexpectedly big changes in annual reports’

positions. For example finance, which naturally gets most of the attention, can vary a lot and

cause big leaps on the map. This can happen because of the structure of annual reports can

change. When attention for one capability increases it means less attention for other

capabilities in the used framework. Therefore deviations do not only affect the changed

capabilities but the whole portfolio when the values are calculated as shares of total attention.

8.3 Contribution, future research and managerial implications

The main contribution of this study is using the relatively new approach of combining content

analysis and self-organizing map in the field of management studies. Similar approach of

handling textual data has been used before, for example, for organizing document collections

on the Internet (comprehensive review in Janasik et al. 2009). Quantitative content analysis is

a traditional research method in this field, and is still widely in use, but using it together with

SOM has not been done. Even self-organizing map alone has been used only little for business

studies.

In addition to being a promising addition to content analysis, self-organizing map is a

convincing alternative for multidimensional scaling techniques presently used in many

management studies. For example illustrating organization's search in a rugged landscape (e.g.

Gavetti and Levinthal 2000) with different kinds of data could be one application. With a large

and accurate data set results could be very interesting. For example patents or monetary

resource allocation could be used as a more representative data. And with a larger number of

samples the results of SOM usually improve.

One of the most popular applications of SOM is finding clusters from data. So far in the field of

business it has been used for market segmentation (Kiang et al. 2006; Huang et al. 2007), but it

would be also an ideal method for finding strategic groups among industries (e.g. Cool and

Schendel 1988; Reger and Huff 1993; Osborne et al. 2001; D'Este 2002). However, less variables

(or much more samples) should be used than in this study if clustering would be an objective.

This is because every increase in the number of variables give samples a new dimension of how

to differentiate from other samples, which makes clustering more difficult. Investigating the

relationship between firm performance and strategic groups could also be rewarding.

196

Implications for managers mostly derive from the literature part of this study. Firstly,

everything in organizations cannot be given attention. Thus managers have to divide their

attention to an optimal set of capabilities for firm’s current situation. Secondly, different

communication channels in an organization can affect the firm behavior. That is, annual

reports and other channels for external communication can also serve as important channels

for internal communication.

197

REFERENCES

ADNER, R. & HELFAT, C. E. 2003. Corporate Effects and Dynamic Managerial Capabilities.

Strategic Management Journal 24(10): 1011-1025.

AMIT, R. & SCHOEMAKER, P. J. H. 1993. Strategic Assets and Organizational Rent. Strategic

Management Journal 14(1): 33-46.

ATHREYE, S. S. 2005. The Indian Software Industry and Its Evolving Service Capability.

Industrial and Corporate Change 14(3): 393-418.

AUGIER, M. & TEECE, D. J. 2008. Strategy as Evolution with Design: The Foundations of

Dynamic Capabilities and the Role of Managers in the Economic System. Organization

Studies 29(8-9): 1187-1208.

AUGIER, M. & TEECE, D. J. 2009. Dynamic Capabilities and the Role of Managers in Business

Strategy and Economic Performance. Organization Science 20(2): 410-421.

BACK, B., TOIVONEN, J., VANHARANTA, H. & VISA, A. 2001. Comparing Numerical Data and

Text Information from Annual Reports Using Self-Organizing Maps. International

Journal of Accounting Information Systems 2(4): 249-269.

BARNEY, J. 1991. Firm Resources and Sustained Competitive Advantage. Journal of Management

17(1): 99-120.

BARNEY, J. B. 2001. Resource-Based Theories of Competitive Advantage: A Ten-Year

Retrospective on the Resource-Based View. Journal of Management 27(6): 643-650.

BARR, P., STIMPERT, J. & HUFF, A. 1992. Cognitive Change, Strategic Action, and

Organizational Renewal. Strategic Management Journal: 15-36.

BARRON, D. 2003. Approaches to Strategy. In Faulkner, D. and Campbell, A. (eds.) The Oxford

Handbook of Strategy, pp. 80-103. New York: Oxford University Press.

BECERRA, M. 2008. A Resource-Based Analysis of the Conditions for the Emergence of Profits.

Journal of Management 34(6): 1110-1126.

BEINHOCKER, E. D. 1999. Robust Adaptive Strategies. Sloan Management Review 40(3): 95-+.

BELL, M. & PAVITT, K. 1995. The Development of Technological Capabilities. In ul Hague, I.

(eds.) Trade, Technology, and International Competitiveness, pp. 69-101. Washington

D.C.: World Bank Publications.

BLACK, J. A. & BOAL, K. B. 1994. Strategic Resources - Traits, Configurations and Paths to

Sustainable Competitive Advantage. Strategic Management Journal 15: 131-148.

BLACK, L. D. 2006. Corporate Social Responsibility as Capability. Journal of Corporate

Citizenship 23: 25-38.

BOEKER, W. & WILTBANK, R. 2005. New Venture Evolution and Managerial Capabilities.

Organization Science 16(2): 123-133.

198

BOWMAN, E. 1984. Content Analysis of Annual Reports for Corporate Strategy and Risk.

Interfaces: 61-71.

BRUDERER, E. & SINGH, J. V. 1996. Organizational Evolution, Learning, and Selection: A

Genetic-Algorithm-Based Model. Academy of Management Journal 39(5): 1322-1349.

CAPRON, L. & MITCHELL, W. 2009. Selection Capability: How Capability Gaps and Internal

Social Frictions Affect Internal and External Strategic Renewal. Organization Science

20(2): 294-312.

CAVUSGIL, S., CALANTONE, R. & ZHAO, Y. 2003. Tacit Knowledge Transfer and Firm

Innovation Capability. Journal of business & industrial marketing 18(1): 6-21.

CHEN, J. J. W., PECK, K., HONG, T. M., YANG, S. C., SHER, Y. P., SHIH, J. Y., WU, R., CHENG,

J. L., ROFFLER, S. R., WU, C. W. & YANG, P. C. 2001. Global Analysis of Gene

Expression in Invasion by a Lung Cancer Model. Cancer Research 61(13): 5223-5230.

CHETTY, S. & PATTERSON, A. 2002. Developing Internationalization Capability through

Industry Groups: The Experience of a Telecommunications Joint Action Group. Journal

of Strategic Marketing 10(1): 69-89.

CLARKE, G. 1997. Messages Fromceos: A Content Analysis Approach. Corporate

Communications: An International Journal 2(1): 31-39.

COHENDET, P. & LLERENA, P. 2003. Routines and Incentives: The Role of Communities in the

Firm. Industrial and Corporate Change 12(2): 271-297.

COLLIS, D. J. 1994. Research Note - How Valuable Are Organizational Capabilities. Strategic

Management Journal 15: 143-152.

CONGER, J. 2004. Developing Leadership Capability: What’s inside the Black Box. Academy of

Management Executive 18(3): 136-139.

CONTRACTOR, F. 2007. Is International Business Good for Companies? The Evolutionary or

Multi-Stage Theory of Internationalization Vs. The Transaction Cost Perspective.

Management International Review 47(3): 453-475.

COOL, K. & SCHENDEL, D. 1988. Performance Differences among Strategic Group Members.


D'ESTE, P. 2002. The Distinctive Patterns of Capabilities Accumulation and Inter-Firm

Heterogeneity: The Case of the Spanish Pharmaceutical Industry. Industrial and

Corporate Change: 847-874.

DAY, G. S. 1994. The Capabilities of Market-Driven Organizations. Journal of Marketing 58(4):

37-52.

199

DELERY, J. E. & DOTY, D. H. 1996. Modes of Theorizing in Strategic Human Resource

Management: Tests of Universalistic, Contingency, and Configurational Performance

Predictions. Academy of Management Journal 39(4): 802-835.

DIERICKX, I. & COOL, K. 1989. Asset Stock Accumulation and Sustainability of Competitive

Advantage. Management Science 35(12): 1504-1511.

DOSI, G. 1982. Technological Paradigms and Technological Trajectories - a Suggested

Interpretation of the Determinants and Directions of Technical Change. Research Policy

11(3): 147-162.

DOSI, G., FAILLO, M. & MARENGO, L. 2008. Organizational Capabilities, Patterns of

Knowledge Accumulation and Governance Structures in Business Firms: An

Introduction. Organization Studies 29(8-9): 1165-1185.

DOSI, G., NELSON, R. R. & WINTER, S. G. 2000. Introduction: The Nature and Dynamics of

Organizational Capabilities. In Dosi, G., Nelson, R.R. and Winter, S.G. (eds.) The

Nature and Dynamics of Organizational Capabilities, pp. 1-22. New York: Oxford

University Press.

DRAULANS, J., DEMAN, A. & VOLBERDA, H. 2003. Building Alliance Capability: Management

Techniques for Superior Alliance Performance. Long Range Planning 36(2): 151-166.

DUTTA, S., NARASIMHAN, O. & RAJIV, S. 1999. Success in High-Technology Markets: Is

Marketing Capability Critical? Marketing Science 18(4): 547-568.

DUTTA, S., NARASIMHAN, O. & RAJIV, S. 2005. Conceptualizing and Measuring Capabilities:

Methodology and Empirical Application. Strategic Management Journal 26(3): 277-285.

EASTERBY-SMITH, M., LYLES, M. A. & PETERAF, M. A. 2009. Dynamic Capabilities: Current

Debates and Future Directions. British Journal of Management 20: S1-S8.

EGGERS, J. & KAPLAN, S. 2009. Cognition and Renewal: Comparing Ceo and Organizational

Effects on Incumbent Adaptation to Technical Change. Organization Science 20(2): 461.

EISENHARDT, K. M. & MARTIN, J. A. 2000. Dynamic Capabilities: What Are They?

FIGUEIREDO, P. N. 2002. Does Technological Learning Pay Off? Inter-Firm Differences in

Technological Capability-Accumulation Paths and Operational Performance

Improvement. Research Policy 31(1): 73-94.

FIGUEIREDO, P. N. 2003. Learning, Capability Accumulation and Firms Differences: Evidence

from Latecomer Steel. Industrial and Corporate Change 12(3): 607-643.

FISS, P. C. 2007. A Set-Theoretic Approach to Organizational Configurations. Academy of

Management Review 32(4): 1180-1198.

FLYNN, B. & FLYNN, E. 2004. An Exploratory Study of the Nature of Cumulative Capabilities.

Journal of Operations Management 22(5): 439-457.

200

GALBREATH, J. 2005. Which Resources Matter the Most to Firm Success? An Exploratory

Study of Resource-Based Theory. Technovation 25(9): 979-987.

GALUNIC, D. C. & Rodan, S. 1998. Resource Recombinations in the Firm: Knowledge Structures

and the Potential for Schumpeterian Innovation. Strategic Management Journal 19(12):

1193-1201.

GAVETTI, G. 2005. Cognition and Hierarchy: Rethinking the Microfoundations of Capabilities'

Development. Organization Science 16(6): 599-617.

GAVETTI, G. & LEVINTHAL, D. 2000. Looking Forward and Looking Backward: Cognitive and

Experiential Search. Administrative Science Quarterly 45(1): 113-137.

GEORGE, G. 2005. Learning to Be Capable: Patenting and Licensing at the Wisconsin Alumni

Research Foundation 1925-2002. Industrial and Corporate Change 14(1): 119-151.

GOTTSCHALK, L. 1994. The Development, Validation, and Applications of a Computerized

Measurement of Cognitive Impairment from the Content Analysis of Verbal Behavior.

Journal of clinical psychology 50(3): 349-360.

GRABHER, G. 1993. The Weakness of Strong Ties. The Lock-in of Regional Development in the

Ruhr Area. In Grabher, G. (eds.) The Embedded Firm: On the Socioeconomics of

Industrial Networks, pp. 255-277. London: Routledge.

GRANT, R. M. 1991. The Resource-Based Theory of Competitive Advantage - Implications for

Strategy Formulation. California Management Review 33(3): 114-135.

GRANT, R. M. 1996a. Prospering in Dynamically-Competitive Environments: Organizational

Capability as Knowledge Integration. Organization Science 7(4): 375-387.

GRANT, R. M. 1996b. Toward a Knowledge-Based Theory of the Firm. Strategic Management

Journal 17: 109-122.

HAMBRICK, D. C. & MASON, P. A. 1984. Upper Echelons - the Organization as a Reflection of

Its Top Managers. Academy of Management Review 9(2): 193-206.

HEIDENREICH, M. 2005. The Renewal of Regional Capabilities Experimental Regionalism in

Germany. Research Policy 34(5): 739-757.

HELFAT, C. E. 1997. Know-How and Asset Complementarity and Dynamic Capability

Accumulation: The Case of R&D. Strategic Management Journal 18(5): 339-360.

HELFAT, C. E., FINKELSTEIN, S., MITCHELL, W., PETERAF, M. A. & WINTER, S. G. 2007.

Dynamic Capabilities: Understanding Strategic Change in Organizations. Oxford, U.K.:

Blackwell Publishing.

HELFAT, C. E. & PETERAF, M. A. 2003. The Dynamic Resource-Based View: Capability

Lifecycles. Strategic Management Journal 24(10): 997-1010.

201

HUANG, J. J., TZENG, G. H. & ONG, C. S. 2007. Marketing Segmentation Using Support Vector

Clustering. Expert Systems with Applications 32(2): 313-317.

JACOBIDES, M. G. 2006. The Architecture and Design of Organizational Capabilities. Industrial

and Corporate Change 15(1): 151-171.

JANASIK, N., HONKELA, T. & BRUUN, H. 2009. Text Mining in Qualitative Research

Application of an Unsupervised Learning Method. Organizational Research Methods

12(3): 436-460.

JONES, M. A. 2006. A Content Analysis of Customer Satisfaction in Annual Reports. Journal of

Consumer Satisfaction, Dissatisfaction and Complaining Behavior 19: 59.

KALE, P., DYER, J. & SINGH, H. 2002. Alliance Capability, Stock Market Response, and Long-

Term Alliance Success: The Role of the Alliance Function. Strategic Management

Journal: 747-767.

KAPLAN, S., MURRAY, F. & HENDERSON, R. 2003. Discontinuities and Senior Management:

Assessing the Role of Recognition in Pharmaceutical Firm Response to Biotechnology.

Industrial and Corporate Change 12(2): 203-233.

KAUFFMAN, S. 1993. The Origins of Order: Self Organization and Selection in Evolution. New

York: Oxford University Press.

KIANG, M. & FISHER, D. 2008. Selecting the Right Mba Schools–an Application of Self-

Organizing Map Networks. Expert Systems with Applications 35(3): 946-955.

KIANG, M. Y., HU, M. Y. & FISHER, D. M. 2006. An Extended Self-Organizing Map Network for

Market Segmentation - a Telecommunication Example. Decision Support Systems 42(1):

36-47.

KOCHHAR, R. 1997. Strategic Assets, Capital Structure, and Firm Performance. Journal of

Financial and Strategic Decisions 10(3): 23-36.

KOGUT, B. & KULATILAKA, N. 2001. Capabilities as Real Options. Organization Science 12(6):

744-758.

KOHONEN, T. 2001. Self-Organizing Maps. Berlin, Heidelberg: Springer.

KOHUT, G. & SEGARS, A. 1992. The President's Letter to Stockholders: An Examination of

Corporate Communication Strategy. Journal of Business Communication 29(1): 7.

KRIPPENDORFF, K. 2004. Content Analysis: An Introduction to Its Methodology. Thousand

Oaks, London, New Delhi: Sage Publications, Inc.

KYNGÄS, H. & VANHANEN, L. 1999. Sisällön Analyysi. Hoitotiede 11(1).

LAAMANEN, T. & WALLIN, J. 2009. Cognitive Dynamics of Capability Development Paths.

Journal of Management Studies 46(6): 950-981.

202

LAI, K. 2004. Service Capability and Performance of Logistics Service Providers. Transportation

Research Part E 40(5): 385-399.

LAMBERG, J.-A. & OJALA, J. 2006. Evolution of Competitive Strategies in Global Forestry

Industries: Introduction. In Lamberg, J.-A., Näsi, J., Ojala, J. and Sajasalo, P. (eds.) The

Evolution of Competitive Strategies in Global Forestry Industries: Comparative

Perspectives, pp. 1-29. Dordrecht: Springer.

LAMBERG, J. A., TIKKANEN, H., NOKELAINEN, T. & SUUR-INKEROINEN, H. 2009.

Competitive Dynamics, Strategic Consistency, and Organizational Survival. Strategic


LAVIE, D. 2006. Capability Reconfiguration: An Analysis of Incumbent Responses to

Technological Change. Academy of Management Review 31(1): 153-174.

LAWRENCE, S., GILES, C. L., TSOI, A. C. & BACK, A. D. 1997. Face Recognition: A

Convolutional Neural-Network Approach. Ieee Transactions on Neural Networks 8(1):

98-113.

LEE, C., LEE, K. & PENNINGS, J. M. 2001. Internal Capabilities, External Networks, and

Performance: A Study on Technology-Based Ventures. Strategic Management Journal

22(6-7): 615-640.

LEONARD-BARTON, D. 1992. Core Capabilities and Core Rigidities - a Paradox in Managing

New Product Development. Strategic Management Journal 13: 111-125.

LEVINTHAL, D. & MARCH, J. G. 1981. A Model of Adaptive Organizational Search. Journal of

Economic Behavior & Organization 2(4): 307-333.

LEVINTHAL, D. & MYATT, J. 1994. Co-Evolution of Capabilities and Industry: The Evolution of

Mutual Fund Processing. Strategic Management Journal 15: 45-62.

LEVINTHAL, D. A. 2000. Organizational Capabilities in Complex Worlds. In Dosi, G., Nelson,

R.R. and Winter, S.G. (eds.) The Nature and Dynamics of Organizational Capabilities,

pp. 363-376. New York: Oxford University Press.

LEVITT, B. & MARCH, J. G. 1988. Organizational Learning. Annual Review of Sociology 14: 319-

340.

LITZ, R. 1996. A Resource-Based-View of the Socially Responsible Firm: Stakeholder

Interdependence, Ethical Awareness, and Issue Responsiveness as Strategic Assets.

Journal of Business Ethics 15(12): 1355-1363.

LORENZONI, G. & LIPPARINI, A. 1999. The Leveraging of Interfirm Relationships as a

Distinctive Organizational Capability: A Longitudinal Study. Strategic Management

Journal 20(4): 317-338.

203

MAYER, K. J. & SALOMON, R. M. 2006. Capabilities, Contractual Hazards, and Governance:

Integrating Resource-Based and Transaction Cost Perspectives. Academy of


MCCARTHY, I. P. & TAN, Y. K. 2000. Manufacturing Competitiveness and Fitness Landscape

Theory. Journal of Materials Processing Technology 107(1-3): 347-352.

MCEVILY, B. & ZAHEER, A. 1999. Bridging Ties: A Source of Firm Heterogeneity in

Competitive Capabilities. Strategic Management Journal 20(12): 1133-1156.

Metsäntutkimuslaitos 2008. Metsätilastollinen Vuosikirja 2008. Vammala:

Metsäntutkimuslaitos.

MEYER, A. D., TSUI, A. S. & HININGS, C. R. 1993. Configurational Approaches to

Organizational Analysis. Academy of Management Journal 36(6): 1175-1195.

MILGROM, P. & ROBERTS, J. 1995. Complementarities and Fit - Strategy, Structure, and

Organizational-Change in Manufacturing. Journal of Accounting & Economics 19(2-3):

179-208.

MILLER, D. 1986. Configurations of Strategy and Structure - Towards a Synthesis. Strategic


MILLER, D. 1992. Environmental Fit Versus Internal Fit. Organization Science 3(2): 159-178.

MILLER, D. 1996. Configurations Revisited. Strategic Management Journal 17(7): 505-512.

MILLER, D. & SHAMSIE, J. 1996. The Resource-Based View of the Firm in Two Environments:

The Hollywood Film Studios from 1936 to 1965. Academy of Management Journal 39(3):

519-543.

MINTZBERG, H. & WATERS, J. A. 1985. Of Strategies, Deliberate and Emergent. Strategic


MOLITERNO, T. & WIERSEMA, M. 2007. Firm Performance, Rent Appropriation, and the

Strategic Resource Divestment Capability. Strategic Management Journal 28(11): 1065.

MOORMAN, C. & SLOTEGRAAF, R. J. 1999. The Contingency Value of Complementary

Capabilities in Product Development. Journal of Marketing Research 36(2): 239-257.

NELSON, R. R. & WINTER, S. G. 1982. An Evolutionary Theory of Economic Change. Cambridge,

MA: Harvard Business School Press.

NEU, D., WARSAME, H. & PEDWELL, K. 1998. Managing Public Impressions: Environmental

Disclosures in Annual Reports. Accounting, Organizations and Society 23(3): 265-282.

OCASIO, W. 1997. Towards an Attention-Based View of the Firm. Strategic Management

Journal 18: 187-206.

204

OSBORNE, J. D., STUBBART, C. I. & RAMAPRASAD, A. 2001. Strategic Groups and Competitive

Enactment: A Study of Dynamic Relationships between Mental Models and

Performance. Strategic Management Journal 22(5): 435-454.

PANAYIDES, P. 2006. Enhancing Innovation Capability through Relationship Management and

Implications for Performance. European Journal of Innovation Management 9(4): 466.

PARK, H., GARDNER, T. & WRIGHT, P. 2004. Hr Practices or Hr Capabilities: Which Matters?

Insights from the Asia Pacific Region. Asia Pacific Journal of Human Resources 42(3):

260.

PATEL, P. & PAVITT, K. 2000. How Technological Competencies Help Define the Core (Not

the Boundaries) of the Firm. In Dosi, G., Nelson, R.R. and Winter, S.G. (eds.) The

Nature and Dynamics of Organizational Capabilities, pp. 313-333. New York: Oxford

University Press.

PELTONIEMI, M., LAAKSONEN, O. & TIKKANEN, U. 2010. Operationalization of capabilities:

A review. In Lamberg, J-A. & Peltoniemi M. Capabilities and competitive advantage in

the forest industry. Research report. Aalto University School of Science and Technology.

PENROSE, E. T. 1959. The Theory of the Growth of the Firm. Oxford: B. Blackwell.

PETERAF, M. A. 1993. The Cornerstones of Competitive Advantage - a Resource-Based View.


PISANO, G. P. 2000. In Search of Dynamic Capabilities: The Origins of R&D Competence in

Biopharmaceuticals. In Dosi, G., Nelson, R.R. and Winter, S.G. (eds.) The Nature and

Dynamics of Organizational Capabilities, pp. 363-376. New York: Oxford University

Press.

PORTER, M. 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors,

New York: Free Press.

PORTER, M. & SIGGELKOW, N. 2008. Contextuality within Activity Systems and Sustainability

of Competitive Advantage. Academy of Management Perspectives 22(2): 34-56.

PORTER, M. E. 1998. Clusters and the New Economics of Competition. Harvard Business

Review 76(6): 77-+.

REGER, R. K. & HUFF, A. S. 1993. Strategic Groups - a Cognitive Perspective. Strategic


RIFFE, D., LACY, S. & FICO, F. 1998. Analyzing Media Messages. New Jersey: Lawrence Erlbaum

Associates, Inc.

ROBINS, J. & WIERSEMA, M. F. 1995. A Resource-Based Approach to the Multibusiness Firm -

Empirical-Analysis of Portfolio Interrelationships and Corporate Financial Performance.


205

ROTHAERMEL, F. & DEEDS, D. 2006. Alliance Type, Alliance Experience and Alliance

Management Capability in High-Technology Ventures. Journal of business venturing

21(4): 429-460.

RUEF, M. 1997. Assessing Organizational Fitness on a Dynamic Landscape: An Empirical Test

of the Relative Inertia Thesis. Strategic Management Journal 18(11): 837-853.

SAATY, T., VARGAS, L. & DELLMANN, K. 2003. The Allocation of Intangible Resources: The

Analytic Hierarchy Process and Linear Programming. Socio-Economic Planning Sciences

37(3): 169-184.

SCHROEDER, R., BATES, K. & JUNTTILA, M. 2002. A Resource-Based View of Manufacturing

Strategy and the Relationship to Manufacturing Performance. Strategic Management

Journal 23(2): 105-117.

SHARMA, S. & VREDENBURG, H. 1998. Proactive Corporate Environmental Strategy and the

Development of Competitively Valuable Organizational Capabilities. Strategic


SIERILÄ, P. 1989. Strategioita Vuosikertomuksista Tulkittuna. Paperi ja puu 70: 220-225.

SIGGELKOW, N. 2001. Change in the Presence of Fit: The Rise, the Fall, and the Renaissance of

Liz Claiborne. The Academy of Management Journal 44(4): 838-857.

SIGGELKOW, N. 2002. Evolution toward Fit. Administrative Science Quarterly 47(1): 125-159.

SIRMON, D. G. & HITT, M. A. 2009. Contingencies within Dynamic Managerial Capabilities:

Interdependent Effects of Resource Investment and Deployment on Firm Performance.


SIRMON, D. G., HITT, M. A. & IRELAND, R. D. 2007. Managing Firm Resources in Dynamic

Environments to Create Value: Looking inside the Black Box. Academy of Management

Review 32(1): 273-292.

SONG, M., DI BENEDETTO, C. & NASON, R. 2007. Capabilities and Financial Performance:

The Moderating Effect of Strategic Type. Journal of the Academy of Marketing Science

35(1): 18-34.

SONPAR, K. & GOLDEN-BIDDLE, K. 2008. Using Content Analysis to Elaborate Adolescent

Theories of Organization. Organizational Research Methods 11(4): 795-814.

STANTON, P. & STANTON, J. 2002. Corporate Annual Reports: Research Perspectives Used.

Accounting, Auditing & Accountability Journal 15(4): 478-500.

STUART, T. & PODOLNY, J. 1996. Local Search and the Evolution of Technological

Capabilities. Strategic Management Journal 17: 21-38.

TALLMAN, S., JENKINS, M., HENRY, N. & PINCH, S. 2004. Knowledge, Clusters, and

Competitive Advantage. Academy of Management Review 29(2): 258-271.

206

TEECE, D. J. 1986. Profiting from Technological Innovation - Implications for Integration,

Collaboration, Licensing and Public-Policy. Research Policy 15(6): 285-305.

TEECE, D. J. 2007. Explicating Dynamic Capabilities: The Nature and Microfoundations of

(Sustainable) Enterprise Performance. Strategic Management Journal 28(13): 1319-1350.

TEECE, D. J., PISANO, G. & SHUEN, A. 1997. Dynamic Capabilities and Strategic Management.


TEECE, D. J., PISANO, G. & SHUEN, A. 2000. Dynamic Capabilities and Strategic Management.

In Dosi, G., Nelson, R.R. and Winter, S.G. (eds.) The Nature and Dynamics of

Organizational Capabilities, pp. 334-362. New York: Oxford University Press.

TRACEY, M., LIM, J. S. & VONDEREMBSE, M. A. 2005. The Impact of Supply-Chain

Management Capabilities on Business Performance. Supply Chain Management-an

International Journal 10(3-4): 179-191.

TRIPSAS, M. & GAVETTI, G. 2000. Capabilities, Cognition, and Inertia: Evidence from Digital

Imaging. Strategic Management Journal 21(10-11): 1147-1161.

TUOMI, J. & SARAJÄRVI, A. 2009. Laadullinen Tutkimus Ja Sisällönanalyysi. Helsinki: Tammi.

ULTSCH, A. & SIEMON, H. 1990. Kohonen’s Self Organizing Feature Maps for Exploratory Data

Analysis. (eds.) Proceedings of International Neural Network Conference (Innc’90), pp.

305–308. Dordrecht, Netherlands: Kluwer.

VAN DEN BOSCH, F. & VAN WIJK, R. 2001. Creation of Managerial Capabilities through

Managerial Knowledge Integration: A Competence-Based Perspective. In Sanchez, R.

(eds.) Knowledge Management and Organizational Competence, pp. 159-176. New York:

Oxford University Press.

WANG, C. L. & AHMED, P. K. 2007. Dynamic Capabilities: A Review and Research Agenda.

International Journal of Management Reviews 9(1): 31-51.

VENKATRAMAN, N. & CAMILLUS, J. C. 1984. Exploring the Concept of Fit in Strategic

Management. Academy of Management Review 9(3): 513-525.

WERNERFELT, B. 1984. A Resource-Based View of the Firm. Strategic Management Journal: 171-

180.

VESANTO, J. 2002. Data Exploration Process Based on the Self-Organizing Map. PhD thesis.

Espoo, Finland: Helsinki University of Technology.

WINTER, S. G. 2000. The Satisficing Principle in Capability Learning. Strategic Management

Journal 21(10-11): 981-996.

WITTEN, I. H. & FRANK, E. 2002. Data Mining: Practical Machine Learning Tools and

Techniques with Java Implementations. ACM SIGMOD Record 31(1): 76-77.

207

VORHIES, D. & MORGAN, N. 2005. Benchmarking Marketing Capabilities for Sustainable

Competitive Advantage. Journal of Marketing 69(1): 80-94.

WRIGHT, S. 1931. Evolution in Mendelian Populations. Genetics 16(2): 97-159.

YAM, R. C. M., GUAN, J. C., PUN, K. F. & TANG, E. P. Y. 2004. An Audit of Technological

Innovation Capabilities in Chinese Firms: Some Empirical Findings in Beijing, China.

Research Policy 33(8): 1123-1140.

ZAHRA, S., SAPIENZA, H. & DAVIDSSON, P. 2006. Entrepreneurship and Dynamic

Capabilities: A Review, Model and Research Agenda. Journal of Management Studies

43(4): 917-955.

ZOLLO, M. & SINGH, H. 2004. Deliberate Learning in Corporate Acquisitions: Post-Acquisition

Strategies and Integration Capability in Us Bank Mergers. Strategic Management

Journal 25(13): 1233-1256.

ZOLLO, M. & WINTER, S. G. 2002. Deliberate Learning and the Evolution of Dynamic

Capabilities. Organization Science 13(3): 339-351.

ZYGLIDOPOULOS, S. 1999. Initial Environmental Conditions and Technological Change.

Journal of Management Studies 36(2): 241-262.

ÖZSOMER, A., CALANTONE, R. & DI BENEDETTO, A. 1997. What Makes Firms More

Innovative? A Look at Organizational and Environmental Factors. Journal of business &

industrial marketing 12(6): 400-416.

208

CHAPTER 7

DYNAMIC CAPABILITIES OF EXPLOITATION AND EXPLORATION IN

COMPETITIVE SETTINGS

ARNE KÖHLER


1 INTRODUCTION

Important contributions of extant research on competitive actions aside – addressing the

importance of issues such as the amount (Young et al., 1996), speed (Miller and Chen, 1994;

Chen, 1996; Ferrier, Smith and Grimm, 1999), sequence (Ferrier, 2001), and scope of actions

(Miller and Chen, 1996) – little is still known about the nature of actions firms take in such

dynamic competitive environments. Two separate and contradicting views emerge from

literature regarding how incumbents cope given fierce rivalry. On the one hand, to stay ahead

in highly competitive markets, firms can exploit existing capabilities and take actions to

enhance the “current ways of doing things” (Kirzner, 1973). Faced with stiff competition, firms

have been suggested to increase focus on the core activities (Porter, 1980; Harrigan and Porter,

1983; Ocasio, 1995), cost efficiency (Singh, 1986; Castrogiovanni, 1991) and short term

competitiveness (Cameron, Kim and Whetten, 1987; Lumpkin and Dess, 2001) to protect the

current business and in order to remain competitive. On the other hand, faced with high

competition firms take action to “create new things to do” (Kirzner, 1973), i.e. to innovate and

explore market opportunities. A popular management literature maxim has been to innovate

and escape direct competition whenever rivalry escalates (Kim and Mauborgne, 1999). Firms

have been argued to pursue innovations in order to create uncontested market space or

barriers to competition through differentiation (Haveman, 1993; Dobrev and Kim, 2006; Jansen,

209

Van den Bosch and Volberda, 2006). Although competition is broadly agreed to be a key

stimulant of firm action (e.g. Chen, Su and Tsai, 2007) there is still little consensus on what

kind of action competition gives rise to.

The main objective of this paper is to explore how competitive intensity in an industry affects

the nature of the competitive actions of incumbents. This objective is divided into four distinct

parts. First, the existing knowledge of the implications of competition on firm behavior is

mapped. This is done in order to understand the mechanisms by which rivalry drives action.

Second, viewing competitive actions from an exploitation-exploration perspective the theory

development separates between actions designed to increase efficiency – labeled exploitation

actions – and actions employed to innovate and avoid competitive pressure – exploration

actions. Third, applying theory on organizational adaption and competition the research

addresses the differences in the competitive and performance implications of exploitation and

exploration actions. Drawing on the outlined theory and rational economic behavior, a causal

model of the effect of competitive intensity on both the exploitation and the exploration

actions of firms is put forth; formalized in three distinct hypotheses. Fourth, the model is

subsequently tested and verified using a unique sample of the largest 150 firms in the pulp and

paper industry. This is done through content analysis of newsfeed data, recording 12 790

initiated competitive actions over a period of 20 years.

The aim is to expand knowledge on how companies act in relation to their environment by

focusing on the apparent differences in the characteristics of responses. This potentially not

only enriches theory on competitive interaction, but also gives new insight adding to the

debate on organizational adaption as a consequence of hostile environments. Moreover, this

research attempts to answer recent calls for analyzing the effects of competitive actions on an

industry level (Ferrier, 2001; Chen et al., 2007), and to add much called for longitudinal

empirical input on antecedents of actions (e.g. Lewin et al., 1999). Methodologically, the study

intends to advance the use of content analysis as a creative approach to studying

organizational actions (Uotila et al., 2009) and to demonstrate best practice on analyzing

overdispersed count data (see Norton et al., 2004; Echambadi et al., 2006; Hoetker, 2007). The

full pattern of results suggests that current research on competition would be complimented by

adding the dimension of action nature – incorporating the applied exploitation and exploration

perspective.

The remainder of the chapter proceeds as follows. First, the theoretical background is

introduced by reviewing the two key concepts forming the theoretical base of this research –

210

namely competition and exploration vis-à-vis exploitation. The theoretical part is concluded by

forming hypothesis concerning the expected outcomes of competitive intensity. The empirical

part, in turn, is introduced in with a presentation of the sample industry, its characteristics and

suitability for valid analysis, and continues to present the applied methodology. Finally, the

results are presented followed by a discussion of the findings.

2 THEORETICAL BACKGROUND

2.1 Defining competitive intensity

Competition and competitor analysis has gained a central role in strategic management

literature in the past decades (e.g. Porter, 1985; Chen 1996). Within this research, the impact of

the environmental context – in particular the impact of the moves of competitors – on firm

performance and as a driver of firm actions has become widely accepted (Porter, 1980;

Castrogiovanni, 1991; McGahan, 2004). Below, the extant theory and research on competition is

summarized to understand the mechanisms of competition and its implications for firms and

firm actions.

As noted by Ferrier (2001) incumbents experience pressure by the sum of all competitive

actions in its market environment. Firms first discern what information they can from their

entire competitive environment and then respond, as firms that fail to do so are outperformed

(Chen et al., 2007). On an industry level the propensity to react is driven by the awareness of

the attack, the motivation to respond and the capability to do so (Smith et al., 1992). The

awareness of an action has been found to come from the amount of firms affected by that

action, often measured in the amount of actions an attacker takes, which on the industry plane

translates to the sum of rival action (Chen et al., 1992). Motivation, in turn, stems from the

threat that such actions give rise to, again commonly modeled as amount of and focus of

competitive actions. Hence, more actions create more awareness and motivation, while the

capability to contest remains unaffected. Ceteris paribus, increased rivalry, defined as the

aggregate of industry competitive actions (Young et al., 1996), should increase the competitive

tension experienced by a focal firm.

However, following arguments from industrial organization (IO) theory the amount of actions

taken in an industry may not be enough to capture the intensity of competitive pressure a focal

firm is under. Indeed, competitive pressure also depends on the degree of interdependence

211

between rivals. This corresponds with Castrogiovanni’s (1991) proposition that environmental

munificence is dependent not only on the competition for resources in that environment, but

also on the scarcity of those resources. Then the capacity of the environment merely sets the

upper limit for activity in the short term – the degree of mutual dependence (Cameron and

Zammuto, 1985; Castrogiovanni, 1991). Accordingly, it is the activity of the industry incumbents

that determine to what extent this capacity is utilized. In other words, when resources become

scarcer competition becomes a zero sum game and the interdependence of firms increase

directly or indirectly (Barnett, 1997). Hence, resource availability moderates the effects of

competitive actions but does not imply high rivalry or competitive pressure in itself.

Consequently, a change in competitive pressure on incumbents in an industry is dependent on

two dimensions, namely the change in incumbents’ competitive activity aimed at acquiring

available resources moderated by the size of the resource base (Castrogiovanni, 1991). This

relationship between the amount of competitive intensity in a market, the market resource

growth and activity is shown in figure 1. The term competitive intensity7 is henceforth applied

to describe the pressure exerted by the competition – the competitive actions taken – for a

limited set of resources.

7 Note that the term competitive intensity has previously been applied in a somewhat different context by Barnett

(1997) – as the influence one company has on other incumbents, dependent on industry density and the degree of competing for scarce resources. In contrast, this thesis views competitive intensity as the competitive pressure exerted by all industry incumbents, not a singular firm.

High competitive

intensity

Medium

competitive

intensity

Medium

competitive

intensity

Low competitive

intensity

Increasing Decreasing

Increasing

Decreasing

Competitive

activity

(Rivalry)

Resource availibility

(moderating effect)

Figure 1 - Competitive intensity as a function of competitive activity and resource availability (growth/decline).

212

2.2 Implications of competitive intensity

Competition and resulting pressure has been posited to be the stimulating process causing

firms to act (Gresov et al., 1993; Barnett and Hansen, 1996), because it is the actions a firm takes

in relation to that of competitors that determines its success (Porter, 1980). In other words,

competitive intensity (i.e. the actions firms take and their degree of interlinkedness) induces

firm actions through a mechanism of threat to performance.

When the amount of moves and countermoves escalate all competitors in an industry might

suffer, as any potential above-normal earnings are eroded (Porter, 1980). This is the result when

the high competition for scarce resources drives up the costs of acquiring such resources until a

point where actions merely recoup initial investments (Chen and Hambrick, 1995). Empiric

studies have found that, when rivalry is high, above normal returns tend to be short lived and

the variation of such returns greater (D’Aveni, 1994; Wiggins and Ruefli, 2005). Ferrier (2001)

found that the longer periods of attack and the higher number of attack actions the lower focal

firm performance. Accordingly, the basic view on rivalry is that lower rivalry is good whereas

high rivalry is an unwanted state8 (Bain, 1956). The negative impact on performance of high

rivalry makes competitive intensity perhaps best viewed as a threat: “an environmental event

that has impeding or harmful consequences for the entity” (Staw, 1981: 502). Previous scholars

have argued that as competition exerts pressure of impeding losses to a firm, it is as threatening

as realized losses (Miller and Chen, 1994). In turn, Child even views competitive intensity

directly as a materialized threat, describing competitive intensity as “the degree of threat that

faces decision makers in the achievement of their goals from external competition, hostility or

indifference.” (1972: 4).

This threat from increased competitive intensity has been found induce firms to respond (Chen

et al., 1992; Gresov et al., 1993) and to increase the level of responses (Chen et al., 2007). Indeed,

research suggests that the longer the time lag before a focal firm responds to competitive

intensity the lower the returns (Smith et al., 1991; Miller and Chen, 1994) and/or the higher the

losses in market share (Chen and MacMillan, 1992; Ferrier et al., 1999). Moreover, findings

show that the greater the propensity to respond (Smith et al., 1991) and the greater the number

of responses in the face of increased competitive intensity the better the performance (Young

8 Scholars have also argued the opposite, that rivalry carries also beneficial effects through its impact on effectiveness

and industry evolution as a stimulant for progress and legitimacy (Porter, 1991; Hannan and Freeman, 1984). However, these arguments generally pertain to lower levels of competition or situations where effectiveness that rivalry has forced competitors to learn can be exploited in new less competitive domains (e.g. roots in a competitive home market when the industry is globalizing).

213

et al., 1996). Hence, a rival action is a threat which is minimized by swift and voluminous

counteraction. Firms are thus motivated to respond, which in turn increases industry rivalry

further having a detrimental effect on performance levels.

Even if incumbents have been suggested to respond there is ambiguity in how firms respond to

increased rivalry. Moreover, responding quickly and in great numbers is not always beneficial,

as found in research on competition, if the action taken is not optimal (Hopkins, 2003).

Further, the characteristics of a competitive response may have profound effects as recognized

by Young and colleagues: “some set…of competitive actions may yield more performance

benefits or evoke more rivalrous countermoves” (1996: 252). Then the chosen response becomes

pivotal when, as stated above, the greater the responses any action provokes, i.e. the more it

accelerates rivalry, the worse the performance outcomes (Miller and Chen, 1994).

2.3 The concept of exploitation and exploration and competition

Two separate organizational responses emerge from literature regarding the manner in which a

firm acts when under external competitive pressure. Kirzner contended that to stay ahead in

highly competitive markets firms must either “create new ways of doing things” or create “new

things to do” (1973: 20). The first option entails traditional competition-based strategy –

exploiting current knowledge in improving efficiency, costs and coverage (March, 1991;

Ghemawat and Ricart i Costa, 1993; Greve, 2007), in other words doing better what the firm

already does. The other less common option pertains to avoiding direct competition in existing

domains, i.e. doing new things, through exploring new less contended areas with innovations

and developing new capabilities (Porter, 1980; Ghemawat and Ricart i Costa, 1993; Jansen et al.,

2006). March (1991) captured this in his concepts of exploration and exploitation, defining the

exploitation of existing capabilities and the exploration of new opportunities as the central

learning mechanism by which organizations adapt to their environment9.

Exploitation actions. Exploiting processes have been translated into business activities such as

the incremental improvement of current capabilities, processes, and technologies, as well as

9 Though learning processes are endogenous to the firm and observed actions by definition are external to the

organizations they may, following arguments by Barnett and Hansen (1996), be best viewed as the visible extensions of firm’s internal learning processes. Organizational search is evoked from poor performance, potentially brought on by competitive intensity. Learning attempts are ultimately manifested in actions tested in the market, i.e. in competitive interaction, which in turn affects competitors experienced performance for one thing, and for the other dependent on if the outcome is satisfactory or unsatisfactory determines the focal firms need to carry on searching (exploiting) or broaden its search parameter (exploring).

214

rationalizing and reducing costs (Lewin et al., 1999; Anand et al., 2009). The action outcomes of

such competitive actions can be embedded in the organizations targets as specific goals (Lewin

et al., 1999). Investment in machinery, equipment as well as the processes with which these are

operating has been argued to be the rudimentary approach for maximizing return to current

knowledge in production intense industries (McGahan, 2004; Ghosal and Nair-Reichert, 2009).

Furthermore, process innovation has similar implications for productivity and efficiency

leading firms to improve on current ways of doing things (McGahan and Silverman, 2001).

Accordingly, competitive actions such as (1) investments in new machinery or equipment,

machinery rebuild, upgrade and modifications, as well as (2) similar actions at a larger scale,

namely investing in new mills and factories, and (3) process innovations can be seen as

exploitation actions.

Exploration actions. An innovation is defined as the development, production and launch of

new products or services as well as the communication thereof (Kim and Pennings, 2009).

Innovations have been contended to be the underlying force driving strategic and industrial

evolution through introducing variety and disrupting existing market paradigms, inasmuch as

being analogous to the definition of exploratory processes (Schumpeter, 1934; Porter, 1980;

March, 1991; Brown and Eisenhardt, 1997). Hence, the term exploration actions has previously

been attached to competitive actions such as (1) launching new innovations or (2) aiming at

developing (or successfully developing) new innovations and novel knowledge (Greve, 1998;

Koza and Lewin, 1998; Vermeulen and Barkema, 2001; Greve, 2007; Anand et al., 2009).

Competitive action research has examined product innovations mainly by, including the

introduction, unveiling and roll out of new products (Ferrier et al., 2002). However, increased

search and experimentation activities search, capture the total organizational focus on

exploration more than the mere successful launch of innovations, as the outcomes of

exploration activity is risky and unknown, and thus often leads to failed efforts (Katila and

Ahuja, 2002; He and Wong, 2004). In other words, observable product launches do not fully

represent innovative efforts. To address the issue, some empirical examination of exploration

search has encompassed intensifying and diversifying research and development investments

as well as increased innovation efforts (Koza and Lewin, 1998). Henceforth, exploitation action

and exploration actions will be applied in the meaning of competitive actions that are

exploitative and explorative respectively.

215

2.4 Characteristics of exploitation and exploration

Both processes in the twin concept have been argued to be essential for firm success. In the

words of March: long term survival is dependent on an organizations ability to “engage enough

in exploitation to ensure it’s the organizations current viability and enough in exploration to

ensure its future viability” (1991: 105). Hence, firms need to reach a balance between the right

degree of exploration and exploitation activity. Focusing solely, or excessively, on exploration

leads to “too many undeveloped ideas and too little distinctive competence” (March, 1990:105).

On the other hand, exploitation on the expense of exploration can ossify. The organizational

performance (and ultimately survival) in the long term is thus threatened, as the organization

continues to refine capabilities, resources and knowledge that are gradually becoming obsolete

or non-essential (Uotila et al., 2009). In the short run a focus on exploitation may not be

detrimental, on the contrary it might allow the firm to outcompete rivals and maximize returns

from current capabilities (March, 1991). However, excessive focus on exploitation reduces the

organizations ability to find new opportunities or adapt to environmental changes (Benner &

Tushman, 2002).

As stated by Teece, Pisano and Shuen (1997) a firm’s ability to compete over time may lie in its

ability both to integrate and build upon its current competences while at the same time

developing fundamentally new capabilities. However, these business activities compete for the

same scarce firm resources, for e.g. funds or managerial attention (Levinthal and March, 1993;

Uotila et al., 2009). Moreover, a balance has been argued as difficult as managers are biased for

exploitation (Lewin et al., 1999), both action types are self reinforcing (Holmqvist, 2004; Gupta

et al., 2006), and they require widely different organizational capabilities, routines and

mindsets (Nelson and Winter, 1982).

Moreover, exploitation actions capitalize on fairly certain and proximate revenue streams

(March, 1991; Lewin et al., 1999; Anand et al., 2009), while the performance outcomes related to

exploration activities are both distant and uncertain. The outcome of, for instance, capacity

increasing actions can often be exactly planned and their impacts on costs assessed, and they

are seen in the short term, as the returns from for e.g. increasing capacity are all but

simultaneous to that capacity coming effectively online (Ghosal and Nair-Reichert, 2009). In

contrast, the future returns of launched innovations, for example, are uncertain inasmuch as

resulting demand projections are unreliable (Anand et al., 2009; Griffin, 1997), and even more

so the outcome of investments in early R&D projects.

216

Exploitation is hence associated with incremental improvements and low variance of these

returns, while exploration in turn increases the variance of expected returns (March, 1991;

Lewin et al., 1999). Hence, exploitation is likely to maintain or at best increment the previous

level of performance, while the variation in performance associated with exploration means

that outcomes from exploration can be either high above or far below current returns.

2.5 Effects on exploitation and exploration on competitive intensity

When engaging in exploitation, firms act and respond in existing domains aiming to match and

neutralize any advantage gained by competitors (Schumpeter, 1934; Porter, 1980; Kim and

Mauborgne, 1999). Exploitation, as previously outlined, allows the firm to compete more cost

effectively, but it does not expand the market base for which it competes (Vermeulen and

Barkema, 2001; Anand et al., 2009). Hence, as available market resources remain unchanged

gains from exploitation come only at the expense of competitors. As not responding to

competitors’ incremental improvement actions poses a threat to firm performance there is an

incentive to respond in order not to be outcompeted in the short term (Ferrier, 2001; Chen et

al., 2007). Moreover, as exploitation actions solely targets existing domains, competitors are

likely to become very aware of any exploitation move (Miller and Chen, 1996a). Also, the

relative certainty and immediateness of the outcome of an exploitation action leads to

increased awareness as managers have been found to react more to close than to distant threats

(Hambrick and D’Aveni, 1988).

Innovation, and other exploration action, on the other hand, has been described as

sidestepping involvement in direct competition (Kim and Mabourgne, 1999, 2005), and as a

non-threatening move in general (Porter, 1980). Although to some extent rivalrous behavior,

exploration can create new strategic assets through expanding markets, or introducing new

technology and know-how; which may be beneficial for long term survival of the industry in

general (McGahan, 2004; Mentzel and Fornahl, 2009; Kim and Penning, 1999). Further, even if

markets remain unexpanded, the degree of differentiation in an industry, an outcome of

innovations, is argued to increase competitive barriers through resulting customer loyalty and

increased customer switching costs (Jansen et al., 2006). Successful differentiation acts as a

mobility barrier, shielding from competition on cost/price to the extent that it creates

customer switching costs allowing possible price premiums (Porter, 1980).

As exploratory activities are characterized by a high degree of uncertainty in outcomes as well

as mostly long term effects, they do not represent direct materialized competitive threats

217

(Anand et al., 2009). With any innovative effort there are doubts about their potential benefits

due to technical uncertainties as well as to whether there exists a sufficient market for the

products at a sufficient price (Kim and Pennings, 2009). Even when launched, innovative

products have been found to generate uncertainty in the market leaving firms doubtful if and

how to respond (Tushman and Anderson, 1986; D’Aveni, 1994). Faced with uncertainty

competitors might, instead of responding, employ a wait and see approach to if an innovations

is able to break commercially (Banbury and Mitchell, 1995; Kim and Pennings, 2009).

The possibility for a firm to contest the different actions also varies. Mainly, this is because of

the difference in the ease of the imitation of the action (Barney, 1991). Exploitation actions are

mostly imitable, as firms in the same industry poses similar resources, capabilities and routines

(Porter, 1985; Miller and Shamsie, 1996; Chen, 1996), and as exploitation action only involve the

extension of the current knowledge base (March, 1991). Hence, any exploitation action can be

responded to and they are thus unlikely to be the source of long term competitive advantage.

In contrast, exploratory actions are the outcome of experiential search generating novel

knowledge and capabilities beyond those currently held. The capability to respond to

exploration actions is thus limited as there is ambiguity both in the capabilities needed for the

actions and the process of generating such capabilities (Barney, 1991; Teece et al., 1997). In

other cases, competitors’ innovations are even protected by proprietary rights (Teece, 1986).

Hence, rivals may simply lack the capabilities to respond, thus withstanding retaliation. Not

surprisingly then, Caves and Ghemawat (1992) found that differences in differentiation played a

significant role in explaining intraindustry superior performance over the long term.

Table 1 above summarizes the outlined theoretical arguments regarding exploration and

exploitation of significance in a competitive context. The conclusion is that, exploitation

actions can, in the words of D’Aveni (1994), be seen as direct accelerators of competition while

exploration actions are unlikely to exert much pressure, and hence not affect competitive

intensity. Of paramount importance is also the difference in risk and proximity of returns

relating to exploration and exploitation, as well as the organizational tendency to favor

exploitation to exploration. Exploration is essentially riskier and has returns more far off in the

future than exploitative actions. Keeping this in mind we proceed to examine and theorize on

the different effects of competitive intensity on firm exploitation and exploration actions.

218

Table 1 - Summary of theory on exploration and exploitation: Original definitions, applied competitive action definitions, impacts on outcome as well as impact on the industry competitive intensity.

Action (Activity)

Definition

(March, 1991)

Competitive action equivalent

Performance distribution

Effects on competitive

intensity

Exploitation Utilizing and refining current capabilities focusing on production, efficiency, selection, implementation and execution.

(1) Investments in new or old machinery/ equipment, (2) new mills and factories, and (3) process innovations.

Proximity in returns

Certain returns

Incremental improvements

High – Direct attack on current markets, imposes a real short term threat, strong signals ensures awareness. And actions are highly imitable.

Exploration Searching for new opportunities. Associated with variation, risk taking, experimentation, play, flexibility, discovery and innovation

(1) Launching new innovations and (2) efforts to develop innovations or novel knowledge.

Distant returns

Uncertain returns

Potential for large performance improvement

Low – Differentiates, expands markets, imposes only uncertain and distant threat. And actions are inimitable.

3 HYPOTHESIS

Notwithstanding the broad consensus that competitive intensity leads to firms taking action

there is less agreement on the nature of the actions firms take when under threat from rising

competitive intensity. In fact two contrasting views on the expected outcome of competitive

intensity exist. One claims that increasing competition will force competitors to focus on core

activities (Porter, 1980; Ocasio, 1995) and respond to increasing competitor moves by similar

measures (Chen, 1996), in other words exploitation. The other, in turn, postulates that

increasing competitive pressure will lead firms to explore new less competed avenues

(Haveman, 1993; Dobrev and Kim, 2006; Kim and Mauborgne, 1999). The theoretical

development (illustrated in figure 2), rests on arguments from theory on organizational

learning and rational economic behavior. It moves towards a reconciliation of the seemingly

contradictory views presented above, through accounting for the nature of responses and

distinct implications of the differences in the degree of threat they represent.

219

Repeating arguments on competition; increased competitive pressure induces actions (Porter,

1980; Chen et al., 1992; Gresov et al., 1993). Firms are driven by performance incentives, as firms

that are slow to respond to rising competitive intensity or fail to respond completely, are likely

to lose relative market positions and perform worse financially (Young et al., 1996; Ferrier, 2001;

Ferrier et al., 1999; Miller and Chen, 1994). Put differently, problems or dissatisfactions with the

current state or threats due to increased competitive intensity induce firms to take action, and

start to search for opportunities close to existing domains (Dutton and Jackson, 1987;

Holmqvist, 2001).

Prospect theory suggests that the propensity of firms to take risks increase with greater threat,

thus firms take more risky actions (Kahneman and Tversky, 1979). In addition, managers

become risk takers in the face of potential losses (March and Shapira, 1992), and firms might

even display hyper active behavior when threatened (Hambrick and D’Aveni, 1988). Advocating

the opposite, rigidity theory proposes that the threat from competitive intensity will lead a firm

to take on less risk and become rigid to change, but that it will increase activity along past

trajectories (Staw et al., 1981). Threatened firms should hence strengthen past behavior relying

on dominant and tested responses. In spite of unresolved views on the nature of responses,

theory thus suggests that increased competitive intensity will lead to increased activity – either

in new domains or along past trajectories.

Hypothesis 1: Increasing competitive intensity will have a positive effect on the overall level of

firm actions

Beyond indicating higher overall activity, threat-rigidity theory implies that firms respond to

threats with exploitation actions, through relying on less risky actions and focusing on past

H3

Competitive

intensity Actions

Exploitation

actions

Exploration

actions

H1

H2

Figure 2 - The hypothesized relationships and directionality of effects

220

strengths. According to theory on competition, increased competition and threat has likewise

been argued to lead to focus on core activities (Porter, 1980; Zammuto and Cameron, 1985;

Harrigan and Porter, 1983; Prahalad and Hamel, 1990), cost efficiency (Singh, 1986;

Castrogiovanni, 1991), short term competitiveness (Cameron, Kim and Whetten, 1987; Lumpkin

and Dess, 2001), incremental adaption (DiMaggio and Powell, 1983), risk aversion and

skepticism about noncore activities (Cameron, Kim and Whetten, 1987; Walsh, 1995). The

underpinning argument is that increased competitive intensity is a direct threat on the firm’s

resource domain and failing to become competitive in the short term has clear and immediate

consequences; leading to poor performance and even reducing slack resources that can be used

for exploration and long term success.

Exploitation is hence the likely first response to competitive threat as it is associated with

reliable and quick results (March, 1991), which are preferred by managers (Lewin et al., 1999;

Beckman, 2006). High competitive intensity will further divert managerial attention away from

exploring other opportunities and more towards competing effectively in chosen domains as

strategy becomes more like fire-fighting: dictated by the need to focus on immediate issues

(Miller and Chen, 1994; Nadkarni and Barr, 2008). Furthermore, managerial perception has also

been argued to be biased to cling to heuristics of past success in the face of a changed

environment (Lant, 1992) and Ocasio (1995) found that managers faced with threats were more

likely to act within their existing domain.

However, as competitive intensity increases, at some point exploitation actions will start to

become a rather unattractive response. The higher competitive intensity erodes competitive

advantage of imitable moves and shortens the sustainability of above normal returns (D’Aveni,

1994). The cost of undertaking exploitative actions will still be the same as in the low intense

environment. Then, at some extreme the cost of engaging in an action will outweigh the

potential benefit that can be gained, as competition is likely to nullify any gains in high rivalry

(D’Aveni, 1994). Given that expected payoff to an action is a strong predictor of competitive

response (Grimm and Smith, 1997), high competitive intensity thus reduces the profit incentive

underlying exploitation action. Consequently, firms will start to lessen the degree to which they

engage in direct competitive behavior, i.e. exploitation action. Indeed, firm tend to refrain from

further action once the threat of to hard retaliation becomes apparent (Anand et al., 2009;

Baum and Korn, 1999; Chen, 1996).

Moreover, exploitation actions result in incremental improvements and increased reliability in

the expected mean performance of the firm in comparison to rivals (March, 1991). However,

221

increased rivalry reduces the industry performance mean, thus when competition becomes

more intense, efforts that only lead to higher reliability of achieving the industry mean return

become less attractive. In sum, firms are likely to initially respond to increased competitive

intensity by increasing their exploitation actions. However, as competitive intensity increases

further, the marginal benefits of exploitation actions is eroded to a point where they no longer

offset their costs, potentially even having negative returns, and firms become likely to avoid

further exploitation action. Ergo,

Hypothesis 2 Increased competitive intensity will have an inverted curvilinear relationship

(inverted U-shape) with firm exploitation actions

Given low levels of competitive intensity firms can afford to focus less on being efficient short

term. As a result firms tend to allocate more resources to experiment with various ways of

operating and delivering value (Klepper, 1996). Accordingly, Schumpeter (1950) argued that

reduced rivalry actually promotes innovation. This he attributed to the fact that innovation

needs slack in the form of high cash flows as innovation investments are sizeable and their

returns uncertain. Firms hence need large cash buffers and low threat to their core business in

order to engage in exploration action. The propensity of firms to engage in exploration actions

may thus be higher given low competitive intensity.

When competitive intensity starts to increase, the tendency for firms to engage in exploration

actions can be argued to decrease for several reasons. First, as explicated earlier, increasing

competitive intensity shifts industry and firm focus towards short term survival, hence actions

enhancing efficiency, namely exploitation action. A firm cannot survive in the long run, if it

fails to survive in each of the short runs along the way. Second, attempting change through

exploration strategies is inherently risky and gains accrue only in the long term (Hannan and

Freeman, 1984; March, 1991). As managers are generally risk-averse (Singh, 1986; Levitt and

March, 1988), they tend to avoid introducing innovations as long as other competitive moves

still secure satisfactory performance levels. Owing to this, as well as the scarceness of firm

resources, and the difficulties described previously in pursuing exploitation and exploration

simultaneously, the suggested increase in exploitation with rising competitive intensity will

limit the amount of exploration firms engage in.

However, as competitive intensity increases further, the potential benefits from exploring

become increasingly attractive in comparison to exploitation actions. The returns to

exploitation become obsolete, when firms are trapped in endless circles of efficiency battles

(Levinthal and March, 1993; Kim and Mauborgne, 2005). Hence, exploitative action can no

222

longer be the source of long nor short term advantage and managers need to expand their

search into new less contended product/market areas (Bowman, 1982; Lewin et al., 1999; Uotila

et al., 2009). Indeed, the only way to escape decreasing margins might be to develop new

products or services (Kim and Mauborgne, 2005; Jansen et al., 2006). Furthermore, exploratory

actions are highly inimitable and can, contrary to exploitation actions, be effective in

competitive markets as they allow firms to differentiate from the rivals creating barriers to

competition. Hence, exploration can result in sustainable and above average returns even when

competition is high (D’Aveni, 1994; Lewin et al., 1999). Lastly, exploratory action having a low

effect on competitive intensity, contrary to exploitation, might not risk increasing already high

competitive intensity, which if increased would have grave consequences for industry returns.

In conclusion, exploration actions prosper in low competitive intensity but become crowded

out by exploitation as intensity escalates. However, exploration generates more sustainable

advantages than exploitation; hence, its returns become increasingly attractive as the

competitive intensity increases in an industry. Exploration, although containing higher risk,

generates in comparison higher expected returns as the returns to exploitation actions become

more short-lived. Thus, the probability of firms taking exploration action increases when

competitive intensity increases further. Ceteris paribus,

Hypothesis 3: Increased competitive intensity will have a curvilinear relationship (U-

shaped) with firm exploration actions

From the theory development it is evident that increasing competition has mirrored effects on

exploration and exploitation, which is attributable to their opposite performance distributions

and effects on competition. Modeling exploration and exploitation as inversely related to

competitive intensity satisfies partially the theorized constraints of exploitation and

exploration, i.e. the requirement of balance.

4 THE EMPIRICAL SETTING – THE PAPER INDUSTRY 1989-2009

4.1 Survey of the paper industry

The paper industry (often termed the pulp and paper industry) refers to organizations that

manufacture various paper and board products from pulp. The industry has traditionally

accounted for significant amounts of employment and capital investments on both sides of the

Atlantic (Ghosal and Nair-Reichert, 2009). In 2005, paper industry provided direct employment

223

for half a million people and production amounted to a net worth of 650 billion USD in Europe

alone (Diesen, 2007). End product sales are dispersed over several industries – in 2007 45% of

paper industry production sales were consumed by communication (newsprint, printing,

writing and other papers), 40 % by packaging (packaging paper, linerboard and boxboard) and

15 % by other sources, mostly hygiene and health care tissue.

Paper has grown to become one of the most important global commodities. Paper production

was at its emergence a very local industry, situated at natural occurrences of raw material and

energy (hydropower) in all corners of the literate world. Growth surged in the beginning of the

century, as printing shifted from rags to cheaper and more abundant sulfite material (Lamberg,

2005). This has lead to the paper industry being among the fastest growing industries in the

20th century (Lamberg and Ojala, 2006). Since, the 1950 the paper industry has settled a

compound annual growth rate of just around 3-4 %, correlating closely the global economic

growth exceeding it by a factor of 1-1.5 on average (Laestadius, 1998; Diesen, 2007). Figure 3

displays this industry growth per market region from the 1960 onward including the period of

study. In the focal period the paper industry reached maturity (average growth in 21st century

around 2%), driven mostly by growth in population and general economic trends. As seen,

European growth retarded while North American demand has stagnated and begun decline in

the late 1990’s with the emergence of substituting technology, digitalization and shifting

consumer preferences. However, the global demand for paper products is still growing, pushed

by significant growth in Asia. Concurrently, the relative importance of the South American

market area has also grown. The industry appears to have experienced a global shift, both in

production and consumption, towards rising Asian nations during the period of examination.

This trend is of growth in Asia is expected to continue strong while the consumption in Europe

and North America is forecasted to further decline (Diesen, 2007; Stora Enso annual report

2008).

224

Figure 3 – Paper and paperboard annual regional consumption (production added imports less exports, assuming no storage) in continent regions. Source: Food and Agriculture Organization of the United Nations statistics (FAOstat)

The history of rapid growth has stimulated use of new technology in order to fulfill new

demand. Through the years capacity has mostly been expanded through increasing machine

speed and width (Pesendorfer, 2003) while new plants only account for a small share of

capacity growth. From 1960 to 2005 the production capacity of a newsprint machine rose

fivefold – from 75 000-400 000 ton per year. Generally the larger throughput has resulted in

large economies of scale through lower investment and manufacturing cost per ton of output.

Firms are all but forced to build factories close to current possible maximum scale and with

most recent technology to be competitively efficient (Pesendorfer, 2003). Moreover, huge

investments and long depreciation times – investments recovered in over 20 years – leads to

firms pre-empting coming growth and adding capacity in chunks (Diesen, 2007). Consequently,

the industry has become highly capital intensive as short term competitive advantage has been

in larger and more complex equipment. Today, the industry average is over 100, 000 USD of

equipment for every employee (Ghosal and Nair-Reichert, 2009). The resulting large fixed

costs means firms strive to operate close to capacity– often levels above 90 % utilization are

cited as minimal for achieving satisfactory returns (Pesendorfer, 2003).

0

20

40

60

80

100

120

140

160

180M

illi

on

to

nn

es Africa

Northern America

South America

Asia

Europe

Oceania

Window of study

225

Figure 4 - Weighted average rate of net profits/losses for top paper industry firms based on domestic continent. Source: Pulp and Paper international (PPI).

High competition combined with high investments grades has lead to poor returns on capital.

As seen in figure 4, the long term earning levels in the industry appears to have compressed

significantly, especially in developed continents (Europe and North America), and in Africa,

while Asian and South American paper manufacturers have fared better . The study window

begins at a time when the performance of incumbents has decreased in tandem with the

financial crises of late 80’s and early 90’s. Thereafter, industry characteristic price cyclicality

explains a large part of performance troubles (Lamberg, 2005; Laestadius, 1998). Paper industry

cyclicality arises from multiple sources: general economic swings (PWC annual reports);

unbalanced capacity and demand, as capacity can only be added in large increments; currency

and price fluctuations; and ensuing customer price speculation exacerbating the situation

through stock piling and destocking trends (Sajasalo, 2003; Laestadius, 1998; Diesen, 2007).

As said, the industry has grown in conjunction with the growth of the average size of

production capacities and, hence, companies. The number of paper industry firms peaked the

first decade in the 20th century at around 4000. Since then the population has gradually

diminished with technology advances, increasing efficiency requirements and requirements of

economies of scale. Consequently, waves of mergers and acquisitions have consolidated the

field. By the 1950’s firms had reduced by a fourth and by the year 2000 only half remained (1500

companies). In comparison to typical population decline in mature industries the rate has been

modest (see Klepper, 1996). The paper industry still remains very fragmented with a high

number of companies, most of which are small and medium sized, in spite of the large returns

to scale. In the beginning of the 21st century the top ten companies only represented about one

-10 %

0 %

10 %

20 %

30 %N

et

Pro

fit/

loss

Africa

America

Asia

Australasia

Europe

South America

226

fourth of the top 1000 companies aggregated production. De facto, as a result of a rise in Asian

producers, the industry is no more concentrated today than in the 1950’s (Lamberg and Ojala,

2006).

Traditionally a globalization ratio in the paper industry has been the driven by centrality of

nearness to a raw material base and hydropower as well as strong regional protectionism. In

spite of new technology allowing recycled paper as material making new locations feasible, as

well as the opening up of markets, standards and the reduced costs for freights this still

remains the case (Ghosal and Nair-Reichert, 2009). Internationalization has been limited to

geographically close domains with few expansion attempts beyond continent boarders

(Sajasalo, 2003; Diesen, 2007). In 2000 only ten percent of companies had production facilities

abroad (Lamberg and Ojala, 2006). And, this percentage was substantially lower at the start of

the observation period (in 1974 it was around 2 %). More than local production, the trade flows

between regions is still marginal. Hence, companies are still largely local - of the ten largest

producers all operate at over 80% in domestic continental regions from 1989-2008 (Diesen,

2007).

In conclusion, the paper industry provides a suitable research site owing to its stable and

interlinked environment, high activity and regional separation. First, both the structure – low

levels of growth, commodity product, high costs of unused capacity, high capital intensity,

existence of scale economies, fragmentation – and the low performance under the focal period

of study, indicate a high mutual dependence between incumbents. Competitive actions and

reactions can thus be studied as industry incumbents are highly interlinked, i.e. they feel the

pressure of each others’ moves. Second, the time interval of 1989-2008 is further especially

fruitful as it represents time period of high activity of paper industry firms (Lamberg and Ojala,

2006). Third, the high regional separation offers a possibility to analyze an all but identical

industry setting over separate regional markets.

4.2 Competition in the paper industry

Traditionally competition within the paper industry has focused on efficient production,

namely exploitation, and technologically has evolved incrementally rather than by radical

shifts. Firms have employed a two-pronged strategy in exploiting their capabilities to the fullest

and reaching scale (1) investments in modernization, and (2) mergers and acquisitions (M&A)

actions.

227

Typically this has implied huge investments in production technology; mostly on larger and

more efficient machines in order to achieve scale economies and lower production costs

(Laestadius, 1998; Lamberg and Ojala, 2006). Giving evidence of the importance of such

exploitation action is the constant evolution of larger, faster and higher paper quality machines

(read less defects). Since the 1960 there existing processes have also been improved through

applying automation and control processes (Laestadius, 1998), and more recently the

introduction of information systems and energy conserving solutions, making processes more

efficient (Diesen, 2007). Exploitation of current capabilities in such modernizations and

incremental efficiency improvements are of paramount importance in the paper industry for

achieving cost gains compounding to short term competitive advantage (Pesendorfer, 2003).

For instance American firms that refrained from investment in the mid 1990’s forecasting a

drop in demand have become trapped as they have similarly fallen behind European companies

in efficiency and grade concentration (Ghosal and Nair-Reichert, 2009). Yet, Lamberg (2005)

contend that little long term advantage had come from efficiency enhancing actions in the

paper industry. Moreover, the high focus on production capabilities and scale has been argued

to lead to overcapacity, and hence, to be the source behind poor performance in the industry

(Diesen, 2007).

Although breakthrough innovations and exploratory action are not the mainstay way of

competing in the industry, firms have throughout the years brought new grades, materials and

products to the market (Ghosal and Nair-Reichert, 2009). For e.g., innovations such as the

introduction of sulfate pulp replacing sulfite pulp; the innovation of machine coated papers;

and thermo and mechanical pulp innovations in recyclable paper recovery have been radical

from a competition viewpoint (Diesen, 2007). Whereas the common bulk commodities have

competed on price set by demand and supply – not production costs - specialty products (new

grades, materials or products) in the paper industry have competed on quality and attributes

(Diesen, 2007). The erosion of price premiums for specialty products has also been slow, for e.g.

coated paper has gradually become a bulk product but this took decades (Diesen, 2007). More

recently exploratory action has been taken in the introduction of bio-energy products,

biotechnology products, plastic-fiber mixing materials and eucalyptus as higher quality

material (Laestadius, 1998; PWC annual industry reports).

In sum, the nature of the paper industry is well documented and there is a basic understanding

of the effects of industry actions, however how and when such actions are employed remain

unanswered, which in part creates the motivation for this study.

228

5 METHODOLOGY

5.1 Sample

The sampling of firms was conducted using the following three criteria: (1) the firm is from the

paper industry, (2) the firm was one of the 100-150 largest companies in the years 1989-2009, (3)

pulp and paper made up more than 70 % of the firm sales.

The sample was constructed of the largest 100 paper industry firms, measured by sales of paper

and pulp in dollars, between the years 1989-2008 (before the year 2001 this list included the 150

largest firms). Industry and period suitability is discussed above. The non-random sampling

was chosen for both efficiency and analysis reasons. First, utilizing large firm data allowed for

effective gathering of firm-year financial and production data otherwise difficult to reconstruct.

Other sampling techniques would have restricted sample size and limited the quality of the

data. Second, the chosen sample covers over 70 % of global paper and pulp sales which makes

for a significantly broad scope of the whole industry. Further, the sampling was convenient for

the analysis as larger firms contribute the material part of industry actions (Chen and Miller,

1994; Lamberg, 2005) and the long time period permits discerning individual firm effects.

Following Rumelt (1974) only firms with over 70% of their business coming from the paper

industry are considered representative of the industry and have been included. These firms are

theorized to be sufficiently responsive to within industry changes and similar in structures,

cognition, hence, behavior (Chen, 1996). This allows for credible assessment of the impact of

industry specific factors.

This resulted in a sample of 1608 observations from 152 firms. However, the sample panel is

highly unbalanced. Most firms are present in the initial years 1990-1996, but there is slow panel

attrition throughout the study period, mainly ascribable to merger and acquisition activity but

also to randomly occurring bankruptcies. Entries are also present, as new companies form or

grow, appearing among the 100 largest industry companies. A minor part of the balance pattern

is due to omitted cases because of missing data, which results in firms having shorter time

spans. Listwise deletion was used for missing values at the start or end of the observation

period. In turn, data gaps in the middle of available firm-years were interpolated to take full use

of available information and avoid selection bias (Little, 1992).

229

5.2 Sources and collection methods

The data used was compiled and consolidated from several secondary data sources. Apart from

the main sources detailed below; firm annual reports, general industrial databases (Hoovers)

and investor information was utilized to search for missing and erroneous data points, as well

as firm incorporation dates.

The firms list, including financial and production items, for each sample year was collected

from the Paper and Pulp International (PPI) journals annual list of top 100 companies by paper

and pulp sales. PPI is a professional journal covering all areas of the industry and has been

previously applied as a reliable source by strategic management scholars (e.g. Ojala et al., 2006;

Ghosal and Nair-Reichert, 2009). The sourced data included paper and pulp sales, earnings,

total sales, employees and production quantities. The list incorporated both public and private

companies, but excluded companies for which no data was public in the measured year.

FAOstat, a statistics service provided by the United Nations, which offers longitudinal data on

exports, imports and production quantities for countries and regions by several paper industry

product categories, was used to collect market growth rates across regions.

Firm actions were gathered through the PIRA (Packaging, Paper, Printing and Publishing,

Imaging and Nonwovens Abstracts) database, a joint venture updated by industry analyst

companies. The database stores abstracts and main information from all published items

related to the paper industry in 1000 journals, newspapers, books, market reports, conference

proceedings, standards, and technical reports. Coverage is international and extends from 1975

to present including over 600 000 articles. To avoid duplicates because of using multiple

sources the database incorporates an automatic filter. Industry experts have attached

descriptors, article codes, publishing years and involved firm names to each abstract. The

descriptors are one or two-word indicators of the content of the news item. Usually abstract

descriptors include geographic region, business and nature of the event (e.g. Asia, paperboard,

new product or company takeover). Previously, the coverage of one main publication has been

standard procedure for event coding in competitive dynamics research (Chen et al., 1992;

Ferrier et al., 1999). The broader coverage of the PIRA database and significant overlap in the

news items reported suggests that the PIRA database is sufficient for valid collection of firm

actions.

The data from each source was linked using firm specific codes connected to company names.

Company name changes and alternative name forms were checked, confirming and manually

230

bridging company codes across occurrences of mergers or divestments. A search for missing

data points from additional sources was then conducted and outlier values were cross-checked

with company reports. Listwise deletion was used in cases of missing data when manual

augmentation was unsuccessful. For analysis purposes the data was structured as panel data –

where each firm-year was listed with corresponding financial, production and firm action

information. A panel consisting of a one year lag for variables was used (except for age, size and

diversification). The lag in panel structure was used as both exploitation and exploration

actions take a considerable time to plan and implement (Chen et al., 1992). Hence, observed

action as reaction to changing competitive intensity comes with a delay. For robustness a three

year lagged panel structure was also employed, but it yielded similar results and for sake of

parsimony it is not reported. Further details on how the panel model variables were

constructed, is presented below.

5.3 Dependent variables – Coding of exploration and exploitation

A central part of the research design was to identify firm’s actions. Actions were identified

using quantitative content analysis of news items, categorizing relevant firm actions into

different classes of actions. This is consistent with approaches utilized in previous studies of

firm level events or actions (MacMillan et al., 1985; Chen et al., 1992; Young et al., 1996). News

feeds are chosen as the source for actions as prior research has found that incumbents react to

the announcement of actions rather than the actions themselves (Westphal and Zajac, 1998).

Methods of structured content analysis as outlined by Krippendorff (2004) were applied to

classify the actions, constructing a coding manual and then testing the validity of the

automated categorization process.

The coding process itself was executed by proprietary software, specifically designed for the

purposes of this study. The software automatically compared manually assembled keyword lists

to newswires and attached descriptors allocating actions to categories. Computer aid has been

argued to allow for error free content analysis of the large amounts of data (Krippendorff, 2004)

and research has shown that automated content coding utilizing key words generates as

accurate results as manual coding (King and Lowe, 2003). Moreover, recent study suggests that

single keyword analysis is able to match results from coding schemes using complex context

specific coding rules (Laver, Benoit and Garry, 2003). A simplified structural model of the

coding process is pictured in figure 5 below.

231

As stated, the textual data used for the content analysis was sourced from the PIRA database.

The standard approach in events coding in previous research (e.g. Miller and Chen, 1994) has

been to use a single news source, most commonly one credible industry journal. Conversely,

PIRA covers all relevant industry reporting, hence, avoiding bias resulting from the choice of

journal. All news items containing any action relating to the sample firm were collected for the

years 1989-2008, resulting in approximately 30 000 collected news items.

Determining the recording units is the main step in content analysis. Following the theoretical

background, to ensure empiric definition is consistent with the conceptual definition, the

categories were exploitation; (1) investments in new machinery /equipment, machinery rebuild,

upgrade and modifications, (2) as well as similar actions at a larger scale, namely investing in

new mills and factories, and (3) process innovations; and exploration; (1) the launch of new

innovations, (2) or actions aimed at developing new innovations and novel knowledge. In order

to construct as specific coding rules as possible for the different actions all descriptors with a

frequency over 10 in the collected newswires (over 7000 descriptors) were manually examined.

The descriptors or combinations of descriptors were then assigned to extensive key word lists

for each action type which served as the coding manual for the software. Keyword’s included

the direct definition terms presented above, and their synonyms as well as related terms. Note

that the keyword list also featured descriptors identifying irrelevant news items for exclusion.

Descriptive examples of different news items classified into exploration and exploitation action

classes and main keywords used can be seen in table 2 below.

All

act

ion

s

Exploitation

actions

Exploration

actions

Installations

Modifications

Process innovations

Introduced product/service

innovations

Launched development

projects

Allocation

principles

Classification

based 7000

available

descriptors

Newsfeed data - 30 000

firm relevant items

Figure 3 – Illustration of the automated classification process from newsfeed abstracts, through coding filter based on descriptors to subcategories of main exploitation/exploration actions

232

Table 2 – Action categories, main keywords and examples news items listed in each category

Category Main Keywords (Descriptors) Example excerpt from news item abstract

Exploitation New mill, new plant, new equipment, new installation, modification, automation, investment machine, upgrade, rebuild

The Matussiere et Forest group is investing FFr150m in its main paper production facility at Turckheim in Alsace raising production from 112,000 to 150,000 tones, chiefly newsprint.

Exploration Innovation product/material/pack, new product/material/pack, new technology

Kimberly-Clark Corp. has developed new Kimdura synthetic paper for in-mould label applications. It is recyclable, has a smooth look, and handles like paper for printing, sheeting and die-cutting.

Finally, the level of exploitation action was operationalized as the count of exploitation actions

and exploration action, in turn, as simply the count of all explorative actions. Consequently, ‘all

actions’ was measured as the sum of both action types. The resulting category

operationalization is not wholly industry as it has been chosen for measurement precision in

the sample setting, hence compromising general applicability.

The retrieved news items were analyzed matching attached descriptors to keyword lists while

matching firm names listed as active in the news item. The software recorded the action type

by firm, as well as the location of the action and the date. News items were allowed to be

classified into multiple categories if descriptors contained key terms for both exploratory and

exploitative actions, for e.g. when a firm rolls out a new product and simultaneously install new

paper machines. Finally, all ambiguous category listings were re-examined manually (e.g. in

cases where newsfeeds were attributed to several categories or more than one company was

listed as active). To avoid the risk of reporting the same action twice times a separate code was

used. It scanned all listed news items listing amount of overlap in actors, dates and abstract key

nouns (places, figures, actions, names). Articles with the any same actor, action, date (absolute

difference of action timing less than 3 years) or general abstract content similarity exceeding

40% were considered possible duplicates and re-examined manually. When duplications were

encountered, only the first chronological appearance of a particular action was then retained.

The remaining unique dataset included 11 428 news items and 12 790 actions.

To establish the validity of both key term coding and the computerized coding process, two

academic experts in strategic management with setting industry experience were asked to

separately read a subset of randomly selected abstracts (n=100) and classify the abstracts into

the action categories. A short description of each category was provided otherwise coders were

233

instructed to use personal judgment in categorizing. Subsequently, coding reliability of manual

coding versus machine coding was tested using Cohen’s kappa (Cohen, 1960). The test yielded a

kappa of 0.82 indicating high validity in the coding process.

5.4 Independent variables – Competitive intensity

Following arguments from theory, competitive intensity induced by competitor activity in a

market is conditioned on market growth. To avoid losing any information on competitive

intensity as a threat a measure that covers both is applied (Castrogiovanni, 1991). As exploration

action is by definition not directly targeted at existing domains, and argued not to exert

competitive pressure, market activity is measured only as the aggregate of the exploitation

actions in a market. Market growth has simply been operationalized as firm home market

consumption growth, i.e. change in paper and board consumption for each time period.

Competitive intensity has subsequently been modeled as the inverse of growth rate times the

amount of direct competitive actions taken in that market, i.e. the sum of all exploitation

actions excluding those of the focal firm:

(1)

Following Echambadi (2006) centering of variables was avoided. Alternative weighed measures

were applied as well. A measure using the growth in market demand (in tons) less the increases

in capacity (tons) associated with exploitation actions was introduced to capture the

competition over resources more pragmatically. The measure ads a tangible dimension to

competitive intensity, measuring the added production capacity versus the added market

demand, i.e. the market the markets’ ability to absorb the new capacity. This measure was

similar manner to the initial measure, co-aligning with changes in the initial measure and

having the same scale variance. However, the measure lacked in relativity, e.g. a decrease in

demand (x tons) would certainly have a different effect on competitive intensity depending on

the total volume of the market to begin with, which is not captured in the measure. Moreover,

the measure is error prone. In spite of differences, results were largely homogenous. This in a

sense partly validates the viability of the direct unweighted inverse interaction measure. For its

simplicity the initial measure is used in reporting.

234

To capture the curvilinear effects in the hypothesis a squared variant of the measure was

included. The competitive intensity variables have, as in similar research, been lagged as firm

actions require time to implement, due to planning, bureaucracy and general difficulties in

mounting the sufficient resources for response (Chen, 1996). An average of the competitive

intensity for the past three years was used as well as other lags (two, four etc.). The measures

yielded similar results as the lagged growth and are hence not reported.

5.5 Control variables

Several other factors have been identified to affect company actions (Lewin et al., 1999).

Accordingly, a baseline model to control for other hypothesized influencing effects is

developed. Adding sufficient control variables further justify the applicability of random effects

models, as firm heterogeneous effects are accounted for. Each control variable has been

included for at least one of the following reasons: (1) the variable has previously been found to

have a significant effect on company actions, (2) there is strong theoretical motivation

suggesting that the variable influences actions, or (3) the variable is strictly firm specific and

has potential in capturing firm specific effects. The educated reader may wish to skip to the

final list of included variables and their anticipated effects shown in table 3 at the end of this

sub-chapter.

First, firm size is controlled for as researchers have found that the proposed nature of response

(Anand et al., 2009), the propensity to act when faced with threats (Cyert and March, 1963;

Levinthal, 1991; Chen and Hambrick, 1995), and the overall activity of a firm (Haveman, 1993;

Lamberg, 2005) vary with organization size. Firm size is controlled by including the natural

logarithm of the number of employees in the model (cf. Ferrier et al., 1999; Uotila et al., 2009)

and by dividing the sample into a subset of large and small firms, the divider being the median

amount of employees. Employees were chosen rather than sales as the amount of employees

fluctuate less than sales and provide a more robust measure of company size. Including sales as

a control variable will also alleviate potential bias from missed data as firm size is expected to

be correlated with the missing data.

Second, as similar effects as those caused by size have been argued to exist for firm age as well a

measure of firm age is included (Nelson and Winter, 1982; Hannan and Freeman, 1984; Miller

and Chen, 1996a). The effect of firm age is controlled by including the measured of time passed

since the funding of the firm. Alternatively, the age was measured as the time passed since the

firm entered the paper industry for diversified conglomerates. Third, the threat of poor

235

performance is a main driver behind the theorized effects of the independent variable, hence

there is a need to control for a direct measure of performance, which is measured as the firm’s

profits at the previous time interval, t-1. Fourth, acknowledging concepts such as inertia

(Hannan and Freeman, 1984), strategic paths (Lamberg, 2009), and momentum (Miller and

Friesen, 1982) previously taken actions have been found to influence future actions

substantially. Previous actions are therefore included, as the firms’ actions at t-1, to control for

firm heterogeneity in the propensity for either action.

Fifth, mergers and acquisitions (M&A’s) or divestments can have similar effects as exploitation,

namely increasing efficiency and, hence, be motivated by achieving economies of scale or scope

(Ghosal and Nair-Reichert, 2009). M&A’s can also be knowledge sourcing, substituting partly

explorative activity (Anand et al., 2009), requires firm resources, constraining the ability to

engage in other action, and influences the level of competitive intensity (Pesendorfer, 1998).

Sixth, as argued earlier the greater the overlap of two firms businesses the greater the

competitive pressure that is felt from the rival’s actions. Though the sample includes only firms

with over 70% of sales from the paper industry, the degree of involvement can still affect the

way a company would react to the market and peer company actions. Involvement is modeled

through the rate of paper industry sales in relation to overall sales by a firm, which signals how

dependent the firm is on the examined industry. Seventh, although market growth acts as a

moderator to competitive intensity, as theorized earlier, it has been found to have effects in

isolation as well (e.g. Greve, 2007; Jansen et al., 2006). Hence, potential individual effects of

market growth are controlled by the including market growth at t-1.

Lastly, there are special market idiosyncrasies beyond mere growth and competitive activity

that can influence actions. For instance, regulatory and stakeholder requirement, governance

structures, capital market differences and other institutional exceptions can influence firm

actions (Lewin et al., 1999). As shown in the setting description (Section 4) firms are

dominantly dependent on their domestic continental markets. Firms’ main markets, i.e.

continental regions, are thus controlled. The home continent of each firm (except one to avoid

multicollinearity) was dummy-coded as either 1 or 0 and included in the model. The final

included continents were North America, Africa, Europe, South America, Asia and Australasia.

236

Table 3 - Operationalization of control variables and expected directionality of effects

Control variable Measure Expected effect

Size The natural logarithm of the number of company employees

Proposed effects of competitive pressure will be more pronounced for larger firms

Age Years since founding or years since entering the paper industry

Negative effect on exploratory action and positive effect on exploitation action

Performance Profit margin (t-1) Negative effect on actions. Negative effect on exploration action and weakly positive on exploitation actions.

M&A and divestment actions

Count of M&A and divestment actions

Negative effect on both exploration and exploitation action

Previous actions Exploration and exploitation actions (t-1)

Exploration(/Exploitation) acts positively on subsequent exploration but negatively on exploitation (/exploration)

Industry involvement

Paper sales/total sales Positive; Enhances the number of expected actions and reaction to industry changes

Market growth Consumption growth (t-1) Positive effect on overall actions and exploitation. Negative effect on exploration actions.

Home market factors

Categorical dummy assigned according to company’s main markets

None proposed

5.6 Model building and testing

First, all variables were tested for within and between group pair wise cross-correlations. This

was simply done to reveal intervariable relationships degrees and directions, and to discover

possible impediments to the succeeding regression analysis due to multicollinearity.

The dependent variables subject to hypothesis testing, i.e. the strategic actions, are counts of

type of actions for each firm-year in the sample. Naturally, a count of actions is cut off at zero

and non-linear, displaying only positive integer values. Further, the analyzed data is panel data,

i.e. pooled cross-sectional data, resulting in a cross-section of firms (i) and their actions over

several years (t). Panel models incorporate both cross sectional and time varying variables, and

are hence more advantageous for the understanding of the dynamics of competition than other

discrete choice methods (Anand et al., 2009). Following the panel structure of the observations,

the process has been modeled with random effects Poisson regression applying cluster-robust

237

estimators of standard errors. Poisson regression assumes that the dependent variable

observations arise from Poisson distribution, as below:

(2)

Normal Poisson regression requires the sample mean to equal the variance, however

dependents violated this assumption indicating overdispersion which can cause unobserved

heterogeneity. Due to this cluster robust estimators of the variance-covariance matrix was

used. Utilizing cluster-robust estimators with Poisson regression allows for computation of

Poisson regression while correcting for overdispersion. Furthermore, negative binomial

regression, which accommodates by default for overdispersion, was also applied. Poisson

regression is however used as the main model in this thesis as it allows for analysis of time

invariant factors. Moreover, as Poisson estimators rely on weaker distributional assumptions, in

correcting specifications of the mean, it has been found to be more robust to use Poisson

estimators with cluster-robust standard errors estimators than negative binomial regression

(Cameron and Trivedi, 2009). A random effects model was applied in order to compute the

effect of time-invariant factors (see equation 3 below).

(3)

The random effects model assumes that there are firm specific attributes that affect outcomes,

which are resultant from random variance, but that these effects are uncorrelated with

regressors (xit) and with the time (contrary to fixed effect models which assumes that αi

correlates with the regressors). The model thus assumes within-firm observational

independence when computing standard errors, thus allow drawing inferences about a broader

population than the sample. Further, the random effects model was chosen as when suitable it

gives better indication of standard errors and yields estimates of marginal effects for all

coefficients. The suitability of the random effects model and exogeneity of variables was tested

through an instrumental approach (Echambadi et al., 2006); running fixed effects regression on

models, and comparing the coefficients and significances with initial random effect results. For

reasons parsimony, only random effects models and full fixed effects models are reported in the

results section. Furthermore, unobserved heterogeneity was controlled by including the lagged

variable of actions both exploratory and exploitative, to capture unexplained firm propensity to

238

engage in either action. Lastly, the models aimed for including several control variable proxies

for potentially significant firm specific attributes.

Model overall fit was finally estimated with Wald Chi-square estimation for random effects

Poisson regression and log likelihood estimation for negative binomial regression (Hausman,

Hall and Griliches, 1984). The results are reported in incidence rate ratios (IRR) and as marginal

effects.

6 RESULTS

6.1 Descriptive statistics

Figure 6 shows the geographical distribution of the sample firms and sample actions. As seen

most organizations are operating in Europe (including Scandinavia) and America, followed by

Asian companies. South America, Africa and Australasia (here referring to broad Oceania) only

represent around 10 % of companies. The action observations are similarly dispersed, with the

exception of Scandinavian companies being highly active hence increasing the share of

observations from the European market area. Moreover, African (3%) and Australasian (3%)

companies were relatively active, while American (31%) and Asian (14%) firms have been

relatively less active over the sample time frame.

239

Figure 6 – The distribution of examined companies between market segments and examined actions

These actions are not balanced and equally present for the research time periods. As seen in

figure 7 (A and B), more actions were taken by the sample firms per year at the beginning of the

observation period. It is evident that the amount of exploitation actions had a diminishing

trend, falling abruptly from 1989-1994; thereafter experiencing a temporary increase followed

again by a more moderate decreasing trend. Only for South America and Africa have the

amount of exploitation actions increased throughout the study (A). A similar trend is visible for

exploration actions (B). The overall rate has remained steady – decreasing slightly - from 1993

to 2001, after which a jump in activity is apparent followed by a slump. Recently activity is on

the rise again pushed mainly by increased exploratory activity in Europe.

2 % 3 %

35 % 31 %

19 %

14 %

2 %

3 %

22 %

12 %

13 %

35 %

6 %3 %

0 %

10 %

20 %

30 %

40 %

50 %

60 %

70 %

80 %

90 %

100 %

Share of organizations Share of actions

South America

Scandinavia

Europe

Australasia

Asia

America

Africa

240

Figure 7 – Cumulative exploitation (A) and exploration actions (B) by region 1989-2009

To address concerns of collinearity table 4 below presents the full variable descriptive data and

bivariate correlation table. The mean level exploitation action for the paper industry firms is as

expected higher than the level of exploration actions, being 2.11 and 0.85 respectively. The mean

of all actions, 2.96, thus follows directly. Exploitation actions also tend to have less variance

compared to the mean than exploration action suggesting higher shifts in exploration activity

0

50

100

150

200

250

300

350

400

450

500

Am

ou

nt

of

ex

plo

ita

tio

n a

ctio

ns

South America

Europe

Australasia

Asia

America

Africa

0

50

100

150

200

250

Am

ou

nt

of

ex

plo

rati

on

act

ion

s

South America

Europe

Australasia

Asia

America

Africa

A.

B.

241

on average. In general, sample firms are old (mean of 75 years) and large (have around 4500

employees and 1.2 billion dollars in sales), earn a low 7 % profit margin, and are highly involved

in the industry. Moreover, within group variances are typically below between group variances.

Observe that the mean cumulative amount of exploitative activity in a market area is high with

high variance, while the growth variable shifts between low values and has lower variance. This

carries implications for the competitive intensity operationalization. In the sample setting the

competitive activity is far more important than growth as a component of the competitive

intensity measure (equation 1). Accordingly, competitive activity mean and variation have

similar values as for competitive intensity and the variables display a high correlation, to the

extent that they cannot as such be entered into the same regression models. Still, incorporating

both variables in the competitive intensity measure (as in equation 1) resulted in stronger

models than when using simply competitive activity (though statistical differences were small).

Sales and employees also correlate substantially and employees have thereby been left in the

model for reason explained earlier. Correlation from region control-dummies have also been

controlled in the fixed effects models. Otherwise correlations are not particularly significant

(compare to Katila and Ahuja, 2002; Stuart, 2000), suggesting low threat of multicollinearity

and variance inflation. Nevertheless, multicollinearity test models were run, however failing to

indicate multicollinearity and variance inflation to an extent that would cause concern.

Nonetheless, to control the threat of distortion from multicollinearity, models were ran with

the least correlating regressors first verifying that signs and effects did not change radically

upon adding new more correlated variables. The effect of the sizeable sample further mitigates

the loss of power associated with collinearity (Echambadi et al., 2006).

242

Table 4 - Within firm means, standard deviations and between firm standard deviations

Variable Mean S.d. within group

S.d. between group

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

1 Exploitation action 2.11 3.03 3.37

2 Exploration action 0.85 1.37 1.68 0.62

3 Exploitation actions t-1a 2.59 2.04 3.37 0.77 0.64

4 Exploration actions t-1 a 0.88 1.49 1.87 0.58 0.75 0.67

5 Competitive activity t-1 99.38 31.81 66.13 0.19 0.08 0.15 0.07

6 Market growtht-1 1.03 0.02 0.03 -0.01 -0.04 -0.04 -0.04 -0.15

7 Competitive intensity t-1 97.45 31.38 65.31 0.19 0.08 0.15 0.08 0.98 -0.19

8 Other actions (M&A's or consolidative action)

1.30 1.81 1.68 0.48 0.53 0.47 0.46 0.08 -0.07 0.08

9 Age 74.22 4.40 40.13 0.11 0.18 0.14 0.16 0.12 -0.10 0.13 0.06

10 Ln employees 8.48 0.28 1.07 0.38 0.42 0.39 0.39 -0.10 -0.12 -0.09 0.49 0.14

11 Ln sales 7.06 0.38 0.95 0.41 0.44 0.44 0.42 -0.14 -0.07 -0.14 0.48 0.17 0.81

12 Profitability (%)t-1 0.08 0.07 0.06 0.09 0.02 0.05 0.00 -0.01 0.11 -0.01 0.02 0.00 0.09 -0.02

13 Involvement (% of sales from P&B)

0.94 0.15 0.09 -0.11 -0.09 -0.11 -0.08 0.01 -0.02 0.01 -0.09 -0.08 -0.14 -0.12 -0.04

14 North America (1=Yes, 0=No)

0.28 0.00 0.46 -0.14 0.02 -0.10 0.03 -0.25 -0.26 -0.24 0.08 0.11 0.30 0.22 -0.09 -0.05

15 South America (1=Yes, 0=No)

0.68 0.00 0.25 -0.04 -0.09 -0.06 -0.09 -0.32 0.03 -0.32 -0.11 -0.08 -0.02 -0.15 0.30 0.04 -0.17

16 Europe (1=Yes, 0=No)

0.41 0.00 0.49 0.22 0.06 0.15 0.04 0.83 -0.12 0.83 0.09 0.11 -0.16 -0.14 -0.04 0.08 -0.52 -0.22

17 Asia (1=Yes, 0=No)

0.21 0.00 0.39 -0.08 -0.06 -0.05 -0.04 -0.42 0.42 -0.42 -0.17 -0.18 -0.20 -0.04 -0.05 -0.06 -0.32 -0.14 -0.42

18 Australasia (1=Yes, 0=No)

0.01 0.00 0.16 -0.04 -0.01 -0.03 -0.03 -0.16 -0.01 -0.16 0.03 0.01 0.00 0.02 0.00 0.01 -0.08 -0.03 -0.10 -0.06

19 Africa (1=Yes, 0=No)

0.02 0.00 0.11 0.01 0.04 0.06 0.03 -0.20 0.01 -0.20 0.05 -0.09 0.20 0.14 0.04 -0.01 -0.09 -0.04 -0.12 -0.08 -0.02

20 Missing value dummy (1=Yes, 0=No)

0.03 0.16 0.07 -0.05 -0.06 -0.07 -0.06 0.04 -0.01 0.04 -0.07 -0.09 -0.16 -0.17 -0.01 0.00 -0.04 0.01 0.07 -0.03 -0.02 -0.03

a The difference in mean and standard deviations between the lagged terms and their simple versions is explained by the differences in the included action years

243

6.2 Incidence rate ratios

Table 5 shows the results from the action, exploitation- and exploration action analysis. Note

that the effects are multiplicative and that the coefficients are reported as incidence rate ratios

(hereinafter denoted IRR: not to be confused with the common abbreviation for internal rate of

return). IRR’s indicate the multiplicative effect on the dependent of a change in a variable by an

increment of 1 ( ). The partial effect of a variable can thus be

understood as a multiplier, and the effect of larger changes in the variable (X) than one are

simply calculated as the IRR in the potency of the change ( ).

Models 1-3 contain the effect of covariates on the total level of actions a company engages in,

hence adhering to hypothesis 1. In turn models 4-6 test hypothesis 2, while models 7 through 9

relate to hypothesis 3. Models 1a, 4a and 7a include only the control variables, and serve as

baseline models for comparison of the improved fit when including the independent variables.

Model 2a, 5a and 7a add the variables corresponding to the developed theory, but do not

include the second order terms. The models thus show the stepwise change of including

competitive intensity in its single form. The second order terms have in turn been included in

the full models 3a, 6a and 9a. All the prior models test the proposed relationships using Poisson

regression, reporting Wald-Chi-square fit (Cameron and Trivedi, 2009). Finally, models 3b, 6b

and 9b replicate the full models applying negative binomial fixed effects regression to verify the

robustness of the results.

Hypothesis 1 – stating that competitive intensity has a positive effect on actions – is partly

supported by the findings. Increasing competitive intensity has a positive effect on the

predicted level of firm actions (the sum of exploitation and exploration), but only up to a

certain point after which the positive impact of increasing competitive intensity decreases. As

seen in model 3a both the IRR of competitive intensity and its squared term are significant.

Their multiplicative effects (1.011) and (0.999) are however opposite indicating that at low levels

of competitive intensity the effect of an increase will be positive while at higher levels this

effect will de facto turn negative. While the IRR implies the constant percentage increase

independent of the value of the variable this does not hold for higher order terms. Consider for

instance the multiplicative effect of an increment of one in the independent on the dependent,

all actions:

(4)

244

Table 5 – Random effects Poisson panel regression with cluster-robust estimators and fixed effects negative binomial regression full modela

All actions

Exploitation actions

Exploration actions

Model 1a Model 2a Model 3a Model 3b



Market growth t-1 4.602 (2.505)**

7.381 (4.07)***

6.697 (3.704)**

5.894 (3.974)**

8.842 (6.805)**

14.658 (11.782)**

14.902 (11.669)**

12.859 (9.604)**

0.843 (0.779)

1.175 (1.075)

1.014 (0.982)

0.874 (0.979)

Age 1.001 (0.002)

1.003 (0.002)

1.003 (0.002)

1.003 (0.002)

1.001 (0.002)

1.002 (0.002)

1.002 (0.002)

1.003 (0.003)

1.004 (0.003)

1.005 (0.003) †

1.005 (0.003) †

1.007 (0.004) †

Size (LN employees) 1.668 (0.104)***

1.696 (0.117)***

1.706 (0.115)***

1.375 (0.1)***

1.57 (0.095)***

1.594 (0.105)***

1.604 (0.101)***

1.492 (0.12)***

1.849 (0.191)***

1.878 (0.196)***

1.887 (0.195)***

1.256 (0.154) †

Profitabilityt-1 4.703 (1.502)***

4.029 (1.347)***

3.118 (1.03)**

3.441 (1.234)**

4.329 (1.517)***

3.789 (1.351)***

2.934 (1.013)**

3.618 (1.392)**

5.462 (3.902)*

4.127 (2.837)*

3.379 (2.396)+

2.814 (2.012)

Involvement (% of sales from P&B)

0.841 (0.256)

0.8 (0.257)

0.812 (0.249)

0.844 (0.258)

0.582 (0.212)

0.549 (0.207)

0.556 (0.205)

0.645 (0.25)

1.258 (0.719)

1.225 (0.679)

1.237 (0.676)

1.641 (0.975)

Other actions 1.019 (0.008)*

1.02 (0.007)**

1.017 (0.007)*

1.021 (0.006)**

1.004 (0.008)

1.005 (0.009)

1.002 (0.009)

1.001 (0.007)

1.031 (0.016)*

1.032 (0.015)*

1.031 (0.015)*

1.034 (0.01)**

Exploitation actions t-1 1.016 (0.007)*

1.014 (0.007)*

1.013 (0.006)*

1.018 (0.004)***

1.018 (0.009)*

1.016 (0.008) †

1.015 (0.007)*

1.016 (0.004)***

1.002 (0.009)

1 (0.009)

1 (0.009)

1 (0.007)

Exploration actions t-1 1.05 (0.008)***

1.05 (0.008)***

1.047 (0.009)***

1.044 (0.009)***

1.03 (0.009)**

1.032 (0.011)**

1.029 (0.011)**

1.026 (0.01)*

1.056 (0.016)***

1.055 (0.016)***

1.054 (0.016)**

1.045 (0.013)***

Exploitation actions Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. 1.017 (0.013)

1.017 (0.012)

1.016 (0.012)

1.024 (0.007)**

Exploration actions Inappl. Inappl. Inappl. Inappl. 1.043 (0.011)***

1.042 (0.011)***

1.04 (0.01)***

1.043 (0.011)***

Inappl. Inappl. Inappl. Inappl.

Home market dummy Inc. Inc. Inc. Inappl. Inc. Inc. Inc. Inappl. Inc. Inc. Inc. Inappl.

Missing value dummyb 1.125 (0.227)

1.165 (0.235)

1.166 (0.222)

1.098 (0.21)

1.162 (0.246)

1.195 (0.255)

1.182 (0.24)

1.13 (0.231)

0.779 (1.747)

0.839 (1.842)

0.857 (2.025)

0.82 (0.404)

Competitive intensity t-1 1.003 (0.001)**

1.011 (0.003)***

1.006 (0.002)**

1.002 (0.001)**

1.013 (0.003)***

1.008 (0.002)***

1.003033 (0.001)*

1.006909 (0.006)

1.005 (0.003) †

Competitive intensity2 t-1 0.999971

(0)** 0.999985 (0)*

0.999964 (0)***

0.999978 (0)**

0.999986 (0)

0.999992 (0)

Wald Chi-Square 298.51 340.12 447.24 250.61 240.1 237.5 342.4 234.720 115.4 138.0 146.9 127.54

Degrees of freedom 12 13 14 11 13 14 15 12 13 14 15 12

Number of firmsc 1608 1608 1608 136 1608 1608 1608 134 1608 1608 1608 98

Number of firm yearsc 152 152 152 1535 152 152 152 1521 152 152 152 1199

Log-likelihooda Inappl. Inappl. Inappl. -2264.338 Inappl. Inappl. Inappl. -1976.72 Inappl. Inappl. Inappl. -1036.21

***p<0.001, **p<0.01, *p<0.05, †p<0.1 aModels (a) are Poisson models for which fit have been assessed with Chi-Square test, while models (b) are fixed effects negative binomial models for which fit have been assessed with log likelihood; bFinal controlled market dummy’s were Europe, America, Asia and Other (Australasia, Africa and South America); cDifference in firms to Poisson models and between models explained by fixed effects models needing more than one observation within groups, hence dropping firms with too few observations.

245

Figure 8 – Predicted multiplicative effect of increasing competitive intensity all other variables kept constant (The deciles correspond to the actual degrees of competitive intensity in the sample, i.e. 0 equals zero, while 10 equals 257).

The term in the last bracket (4) is not independent of the size of the variable. In fact, the size

and sign of the effect is dependent on the size of the independent variable (X), and hence varies

over its range. The effect of competitive intensity on the predicted count of actions over its

entire range is illustrated in figure 8. The red line indicates all actions and the competitive

intensity reported as deciles of the observed intensity range. As seen the effect of increasing

competitive intensity is curvilinear on all actions when holding all other variables constant.

Recapitulating, the effect of competitive intensity is positive on the amount of actions firm take

at low levels of competitive intensity. However, as competitive intensity increases beyond the

7-8 observed deciles of competitive intensity the effect becomes negative.

Moving on to the exploitation action models, findings are consistent with hypothesis 2:

competitive intensity has an inverse curvilinear relationship with exploitation actions.

Interpreting coefficients from the full model 6a, the coefficients of competitive intensity and its

squared variant have opposite, and very significant, multiplicative effects. The effects are thus

similar as in the models regarding all actions (1a-3a). Likewise, this implies a curvilinear

relationship between competitive intensity and exploitation actions; though as coefficients are

larger than in model 3a the observed changes in the incremental multiplicative effect becomes

more pronounced. As seen in figure 8, competitive intensity has increasing effects on

exploitation action up until its seventh decile. This indicates, as hypothesized, that on most

0

0,5

1

1,5

2

2,5

3

3,5

0 1 2 3 4 5 6 7 8 9 10

Mu

ltip

lica

tiv

e e

ffe

ct a

ll o

the

r re

ma

inin

g

con

sta

nt

Deciles of competitive intensity

Actions exploitation actions exploration actions

246

levels of competitive intensity the expected count of focal firm’s exploitation actions rises with

increasing intensity. However, for very high levels competitive intensity this no longer holds

and firms’ expected exploitation activity becomes a decreasing function of further increasing

intensity. Moreover, the economic significance of competitive intensity’s effect on exploitation

action is material. On low levels an increase of one standard deviation in competitive intensity

(equal to approximately 2 deciles in figure 8) implies an increase in expected exploitation

actions by approximately 80 %, whereas it in already high competition decreases the expected

actions by 20% (see figure 8).

Turning to the final hypothesis, tested in models 7 to 9 we observe the effect of competitive

intensity on exploration actions. Hypothesis 3 concerning the inverse relationship between

competitive intensity and the effect on firm engagement in exploration actions is in line with

findings, albeit not entirely to the theorized extent. Including a second order term in order to

explain the extent of exploration action proved unsuccessful, as seen in the difference of fit and

significances between model 8a and 9a. The second order effect of competitive intensity on

exploration actions is thus not found. The effects of competitive intensity on exploration

actions may then best be assessed from model 8a. As seen the coefficient for competitive

intensity is significant and positive. An increase by one standard deviation of competitive

intensity will increase the expected amount of exploration actions by a focal firm by 10%

(=1.00303331.81). The effect of competitive intensity on exploration action is then increasing –

the absolute effect will be greater depending on the degree of previous competitive intensity –

and curvilinear (exponential). However, the failure to find a significant second degree

relationship between competitive intensity and exploration action implies that the proposed

positive effect of low competitive intensity on exploration activity was not supported by the

findings. Figure 8 illustrates how the effect of increasing competitive intensity on exploration

actions grows exponentially.

Comparing these findings regarding competitive intensity’s effect on exploration actions to that

of exploitation actions, and the descriptive statistics, it suggests that the observed effect on all

actions is driven mostly by the large influence of exploitation actions. As exploitation actions

have a much higher presence across the observations than exploration actions, the effects of

exploitation actions in their sum, namely the all actions variable, is dominant. Hence, the

observed effect of competitive intensity on exploitation actions appears to also explain the

unexpected curvilinear relationship found between all actions and competitive intensity.

247

Models 3b, 6b and 9b further give evidence that the findings are not at all sensitive to the

estimator applied. Additionally, the models confirm that findings hold irrespective of whether

random or fixed effects are assumed. Differences in coefficients and significance are marginal at

most, having similar effect on proposed relationships, hence negating suspicions of distorting

effects of endogeneity. Finally, the inclusion of the two variables, competitive intensity and

competitive intensity squared, improved upon the baseline model in all preceding models.

The above analysis focuses on multiplicative effects and individual variable effects are assessed

through holding all other variables constant. However, this might not hold for any of the

observations. The other variables are in fact highly unlikely to remain constant in any of the

sample firms. Hence, although informative, the IRR can in this sense be misleading as its

premises can be violated for all observation. To assess the additive effect of variable changes

and whether the hypothesized relationships actually hold for the observed data it is preferable

to further assess the marginal effects of changes in the variables.

6.3 Marginal effects of competitive intensity

To increase the robustness of results, following Norton, Wang and Ai’s (2004) and Hoetker’s

(2007) best practices for the analysis of non-linear models, the average marginal effects (AME)

of model 1 through 9 are reported in table 6. These models show the effect an infinitely small

change in the independent variable has on the dependent variable on average across the

observed data points. Keep in mind that the calculated marginal effect at any point is

dependent on the values of the covariates at that point, and hence indicates the real observed

variation at the sample data points in contrast to the IRR ratios.

The resulting average effect of competitive intensity on all actions, exploitation actions and

exploration actions can be seen in model 3a, 6a and 8a (note that model 9 was inferior to model

8). On average the effect of an increment in competitive intensity has been positive for all

dependents; all actions, exploitation- and exploration actions. However, only for exploration

action is this average effect statistically significant, and even meaningful. The positive average

marginal effect of competitive intensity in model 8 corroborates previous findings, showing

that growing competitive intensity has increased the propensity of firms to engage in

exploration actions across observations. On average an infinitely small change in competitive

intensity has increased exploration actions by 0.002. As said, the average marginal effect of the

competitive intensity in models 3a and 6a which include a squared competitive intensity term

have little interpretive meaning and hence warrants further analysis.

248

Table 6 – Average marginal effects for models 1a-9aa,b

All actions


Exploration actions

Model 1a Model 2a Model 3a



Market growth t-1 3.858 (1.427)**

5.118 (1.498)**

4.866 (1.5)**

4.159 (1.52)**

5.174 (1.638)**

5.164 (1.588)**

-0.11 (0.594)

0.105 (0.598)

0.009 (0.633)

Age 0.003 (0.005)

0.007 (0.005)

0.007 (0.004)

0.001 (0.003)

0.003 (0.003)

0.003 (0.003)

0.002 (0.002)

0.003 (0.002) †

0.003 (0.002) †

Size (LN employees)

1.292 (0.212)***

1.352 (0.238)***

1.367 (0.23)***

0.861 (0.153)***

0.898 (0.168)***

0.904 (0.158)***

0.397 (0.096)***

0.41 (0.097)***

0.415 (0.097)***

Profitabilityt-1 3.912 (0.833)***

3.568 (0.863)***

2.91 (0.854)**

2.796 (0.709)***

2.567 (0.711)***

2.058 (0.681)**

1.096 (0.449)*

0.922 (0.435)*

0.795 (0.457) †


-0.437 (0.777)

-0.572 (0.836)

-0.533 (0.795)

-1.033 (0.719)

-1.156 (0.757)

-1.123 (0.73)

0.148 (0.368)

0.132 (0.359)

0.139 (0.355)


0.051 (0.018)**

0.043 (0.018)*

0.007 (0.015)

0.01 (0.017)

0.003 (0.017)

0.02 (0.01)*

0.02 (0.01)*

0.02 (0.01)*

Exploitation actions t-1

0.04 (0.018)*

0.037 (0.017)*

0.034 (0.015)*

0.033 (0.017) †

0.031 (0.017) †

0.029 (0.014)*

0.001 (0.006)

0 (0.006)

0 (0.006)

Exploration actions t-1

0.123 (0.022)***

0.126 (0.022)***

0.118 (0.026)***

0.057 (0.018)**

0.061 (0.02)**

0.055 (0.02)**

0.035 (0.011)**

0.035 (0.011)**

0.035 (0.012)**


Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. 0.011 (0.009)

0.011 (0.008)

0.011 (0.008)

Exploration actions

Inappl. Inappl. Inappl. 0.081 (0.022)***

0.078 (0.022)***

0.074 (0.019)***

Inappl. Inappl. Inappl.

Home market dummy

Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc.

Missing value dummy

0.314 (0.575)

0.421 (0.606)

0.424 (0.572)

0.309 (0.471)

0.375 (0.494)

0.348 (0.463)

-0.143 (1.126)

-0.105 (1.197)

-0.093 (1.322)

Competitive intensityc

t-1 0.006

(0.002)** 0.010997 ( .013948)

0.004 (0.002)*

.012641 (.01484)

0.002 (0.001)*

.01654 (.01711)

***p<0.001, **p<0.01, *p<0.05, †p<0.1 aThe validity of the models are equal the validity of the Poisson model validities reported in table 5 bAME gives the average marginal effect, i.e. the average of the change in the dependent given an infinitely small change in the independent, at each observation in the sample. Note that the real marginal effect is not constant throughout the sample.

cAverage marginal effect of the interaction term has been calculated separately

To analyze the marginal effect of the competitive intensity across the sample data the point

estimates of the predicted marginal effect in model 3a and 6a were calculated. Figure 9 shows

the point estimates of competitive intensity in model 6a (A), with the respective significance of

each observation, measuring whether the marginal effect is statistically different from zero (B).

Note that the significance levels are pointwise – they are not adjusted to reflect any

simultaneous testing over the observations in the data – and are hence lower on average than

in model 6a displayed in table 5.

The plotted predicted marginal effects of observations further corroborates hypothesis 2. The

observed marginal effects in figure 9 (A), shift from being positive at low levels of competitive

intensity to becoming negative at higher levels. The positive marginal effects at low levels of

competitive intensity are further significantly different from zero (B), whereas this no longer

holds for the medium levels. And, the negative effect of high levels of competitive intensity

249

again becomes significantly different from zero. This implies that on low levels of competitive

intensity an increase in competition increases firms’ propensity to engage in exploitation action

and at high levels of competitive intensity this is reversed. However, given a medium range of

competition it cannot be statistically shown that competitive intensity has an effect – which is

in accordance with hypothesis 2 as when the effect shifts from positive it first levels off and

then turns negative. The point estimates of competitive intensity in model 3a concerning the

marginal effect on all actions are all but identical to the presented figure below (potentially for

reasons outlined earlier linking exploitation to all actions), having only marginally lower effects

and significance, and is hence not been reported.

-0,06

-0,04

-0,02

0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0 50 100 150 200 250 300Pre

dic

ted

le

ve

l o

f M

arg

ina

l e

ffe

ct

Competitive intensity

A.

250

Figure 9 –Predicted marginal effects of competitive intensity on exploitative actions at each sample observations organized by competitive intensity (A) and the statistical significance of each predicted marginal effect (B).

6.4 Control variables

As expected size correlates positively with actions, both for exploration and exploitation, as

seen in all models 1-9a and in corresponding fixed effects negative binomial models. A shift of

one within standard deviation in size implies a predicted increase in 14% of exploitation actions

and nearly 20% for exploration actions. To control for the proposed hypothesis potentially

being more strongly present in large firms as theorized, models 1 through 9 were replicated

separately for subsamples of large and small firms, divided at the median amount of employees

(= 1953) . The results of the test are reported in table 7. The findings validate expected effects of

size on firm actions. First, the effects of size even within the subsets are positive and

significant. Hence, size increases the propensity of firms to take action. Second, hypothesized

effects are for the most part more strongly present in the larger population, insofar as Model 6

d shows that hypothesis 2 does not even hold for smaller firms. Instead, for smaller firms

exploitation actions grow constantly as a function of competitive intensity and never reach

culmination. In turn, there is no visible difference for the effects in exploration actions across

the large and small subsets.

Other expected effects of the control variables are also supported by the analysis. Prior period

market growth has as expected a positive effect on the amount exploitation actions (and

0

0,2

0,4

0,6

0,8

1

1,2

0 50 100 150 200 250 300

Sta

tist

ica

l si

gn

ific

an

ce o

f p

red

icte

d m

arg

ina

l e

ffe

cts

(p-v

alu

e)

Competitive intensity

B.

251

subsequently on all actions) in every model. For instance in model 6a (market growth IRR =

14.902), an increase in market growth by one standard deviation increases the expected count

of exploitation actions by approximately 5 %. However, the proposed opposite effect on

exploration activity is not found.

Firm age, though proposed to constrain exploratory actions, is found to have weak positive

effect on exploration, while no effect on exploitation actions is seen. Older firms thus take

more exploration action. In grave contrast to anticipations, previous profitability has a

unanimous positive relationship on actions, though with large variance for exploration actions.

The unexpected finding negates the theory that low performance induces exploitation and

exploration action directly. Actions may be driven more by the availability of resources

resulting from good past behavior. Or possibly, the effect of low performance threat on

behavior becomes visible only at very low levels of profitability, thus not testable with the

constructed models, and as such warrants further research. In turn, Industry involvement had

no significant impact, perhaps due to including only largely involved companies, in which case

the sampling criteria would be motivated.

Concerns relating to the effect of firm other actions and past action trajectories proved

significant. Unexpectedly, engaging in other actions such as M&A or divestments has a positive

but weak effect on exploration actions as well as on the sum of actions (see models 1-9 in tables

5 and 6). Hence, the suggested repressing effect that competition over internal resources has on

a firms’ ability to engage in exploitation and exploration is not observable. Conversely, active

firms in other actions were even more likely to engage in exploration actions simultaneously.

However, as the amount of other actions was scant over the researched sample, caution should

be taken in interpreting these findings.

In turn, prior exploitation action at (t-1) does show weak positive correlation with the extent of

firm exploitation at (t), while no discernible effect on exploration actions is apparent. Thereby,

results indicate path dependences in firm competitive behavior. The similar effect is evident

from past exploratory activity (t-1) in model 8a as well – having previously taken exploration

action strongly increases the predicted amount of exploration actions taken by a firm.

Further, as visible from model 6a, and somewhat unexpectedly, previous exploration actions at

(t-1) carry implications for current exploitation actions as there is a significant positive

relationship between prior exploration and exploitation (the reverse is also visible in the

negative binomial fixed effects models 8b, but note that it is not in the Poisson model and

hence not discussed here). Action patterns suggested by arguments regarding temporal shifts

252

between focusing on the development of new capabilities and shifting to exploiting these

capabilities post their competitive introduction are hence visible from the analysis. In turn,

evidence of the mutually exclusive effects of both types of actions is not found – not in model

6a nor in 8a (or in their corresponding fixed effect models). Hence, previously found crowding

out effects of exploitation actions on exploration actions and vice versa are not replicated in

this study. In fact, firms active in exploration actions are more likely to also engage in

exploitation action although the reverse link is not indicated. The finding may be attributable

to the low levels of exploration actions in the sample firms in the first place. As when

exploration is rarely practiced firms already highly active may have a higher probability of

engaging in at least sporadic exploration. Note that the economic impacts of prior actions

though small are still meaningful; a standard deviation in previous exploration increases

expected exploration actions by roughly 8 % and exploitation actions by 4 %.

Lastly, the included control dummy variables have limited impact on outcomes. First, of the

home market dummy controls only the North America dummy carries a statistically significant

impact, and then only on exploitation actions. The average marginal effect of being a North

American company decreases the amount of exploitation actions by 1.67 in model 6a (barely

statistically significant), whereas being European (nearly statistically significant) decreases the

expected action count on average by 0.93. Finally, supporting the viability of the interpolation

technique the interpolation dummy fails to reach significance in all constructed models.

253

Table 7 – Subset large versus small firm random effects Poisson panel regression models with cluster-robust estimators for actions (divided at median average employees = 4106)

a

Large Firms – Actions

Small Firms – Actions

Large Firms –

Exploitation actions Small Firms –

Exploitation actions Large Firms –

Exploration actions Small Firms –

Exploration actions

Model 1c Model 2c Model 3c Model 1d Model 2d Model 3d Model 4c Model 5c Model 6c

Model 4d

Model 5d

Model 6d

Model 7c

Model 8c

Model 9c

Model 7d

Model 8d

Model 9d

Market growth t-1 4.09 (2.471)*

6.277 (3.904)**

5.304 (3.337)**

7.2 (9.489)

13.496 (18.907) †

12.549 (17.3) †

9.755 (9.112)*

15.625 (15.611)**

14.827 (14.538)**

8.258 (11.386)

15.31 (21.99) †

14.504 (20.532) †

0.582 (0.604)

0.783 (0.798)

0.635 (0.677)

2.265 (5.936)

3.48 (9.583)

3.188 (8.725)

Age 1.001 (0.003)

1.003 (0.002)

1.003 (0.002)

1.003 (0.002)

1.003 (0.003)

1.002 (0.003)

1.001 (0.002)

1.002 (0.002)

1.001 (0.002)

1.003 (0.003)

1.002 (0.003)

1.002 (0.003)

1.003 (0.004)

1.004 (0.004)

1.004 (0.004)

1.006 (0.005)

1.006 (0.005)

1.006 (0.005)

Size (LN employees) 1.6 (0.185)***

1.673 (0.219)***

1.712 (0.207)***

1.515 (0.253)*

1.499 (0.245)*

1.513 (0.256)*

1.431 (0.165)**

1.493 (0.194)**

1.536 (0.177)***

1.484 (0.27)*

1.484 (0.269)*

1.503 (0.282)*

1.776 (0.311)**

1.875 (0.344)**

1.905 (0.346)***

1.782 (0.586)

†

1.731 (0.543)

†

1.738 (0.558)

†

Profitability t-1 6.113 (2.426)***

5.524 (2.333)***

3.857 (1.601)**

2.635 (1.394)+

2.332 (1.214)

2.407 (1.235)+

5.866 (2.564)***

5.421 (2.449)***

3.824 (1.708)**

2.192 (1.35)

2.037 (1.25)

2.137 (1.286)

4.661 (4.084)

†

3.747 (3.155)

2.802 (2.398)

6.512 (6.656)

†

5.896 (5.753)

†

6.045 (6.025)

†


1.089 (0.383)

1.003 (0.383)

1.07 (0.369)

0.386 (0.468)

0.339 (0.409)

0.332 (0.399)

0.72 (0.246)

0.675 (0.262)

0.741 (0.272)

0.239 (0.322)

0.22 (0.293)

0.22 (0.291)

1.56 (1.284)

1.453 (1.153)

1.511 (1.181)

1.074 (1.524)

1.08 (1.563)

1.077 (1.571)


1.02 (0.009)*

1.016 (0.009) †

1.03 (0.051)

1.029 (0.052)

1.024 (0.052)

1.002 (0.009)

1.003 (0.009)

0.999 (0.009)

1.058 (0.045)

1.056 (0.046)

1.052 (0.045)

1.033 (0.016)*

1.032 (0.016)*

1.03 (0.016) †

0.96 (0.105)

0.959 (0.106)

0.956 (0.106)

Exploitation actionst-1

1.017 (0.01) †

1.016 (0.01)

1.014 (0.008) †

1.012 (0.043)

1.01 (0.039)

1.01 (0.037)

1.019 (0.012)

1.018 (0.011)

1.017 (0.01)

† 1.012 (0.052)

1.01 (0.049)

1.01 (0.048)

0.998 (0.014)

0.996 (0.014)

0.995 (0.014)

1.019 (0.032)

1.018 (0.034)

1.019 (0.034)

Exploration actions t-1 1.05 (0.011)***

1.049 (0.011)***

1.044 (0.012)***

1.049 (0.03)+

1.051 (0.032)

1.054 (0.034)

1.033 (0.013)**

1.034 (0.013)*

1.027 (0.013)*

1.024 (0.052)

1.03 (0.054)

1.033 (0.056)

1.054 (0.021)**

1.052 (0.02)**

1.05 (0.022)*

1.058 (0.116)

1.062 (0.115)

1.062 (0.112)

Exploitation actions Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. 1.019 (0.016)

1.02 (0.015)

1.019 (0.014)

1.023 (0.069)

1.022 (0.066)

1.022 (0.066)

Exploration actions Inappl. Inappl. Inappl. Inappl. Inappl. Inappl. 1.048 (0.012)***

0.515 (0.1)**

1.043 (0.011)***

1.022 (0.112)

1.018 (0.11)

1.016 (0.109)

Inappl. Inappl. Inappl. Inappl. Inappl. Inappl.

Home market dummy a Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc. Inc.

Missing value dummy b Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl. Excl.

Competitive intensity t-1 208.480 1.011 (0.003)***

1.004 (0.001)**

1.008 (0.004)*

1.002 (0.001)*

1.013

(0.003)***

1.003 (0.001)*

1.008 (0.004) †

1.003 (0.002) †

1.007 (0.006)

1.004 (0.002)*

1.006 (0.006)

Competitive intensity2 t-1 0.99997

(0)** 0.99998

(0) 0.99996

(0)*** 0.9999842

(0) 0.999983

(0) 0.99999

(0)

Wald Chi-Square 176.83 208.480 255.740 28.19 35.220 38.100 123.87 139.320 234.000 27.83 32.030 52.220 62.030 72.180 23.150 33.810 32.720

Degrees of freedom 10 11 12 10 11 12 11 12 13 11 12 13 11 12 13 11 12 13

Number of firms 798 798 798 810 810 810 798 798 798 810 810 810 798 798 798 810 810 810

Number of firm years 76 76 76 76 76 76 76 76 76 76 76 76 76 76 76 76 76 76

***p<0.001, **p<0.01, *p<0.05, †p<0.1 a Due to lack in overlap the region control variable has been reduced to western and other regions, b The Missing value dummy has been excluded due to low overlap between variables

254

7 DISCUSSION

7.1 Conclusions

The main objective of this research was to increase the understanding on how firms react to

competition. This study advanced and tested the notion that competitive intensity, a function

of firm competitive activity in a market moderated by market resource growth, influences the

nature of competitive actions taken by incumbents. Hence, the level of competitive intensity

alters a firms’ propensity to compete directly within established domains – taking exploitation

actions – and attempting to create barriers and avoid direct competition – taking exploration

action. Prior work has theorized that rivalry spurs reaction by industry incumbents (Gresov et

al., 1993), even to the extent that competition is ever accelerating (D’Aveni, 1994; Barnett and

Hansen, 1996), as companies failing to respond promptly to competitors actions are

outperformed, eventually resulting in the demise of the organization (Hannan and Freeman,

1977).

However, as increasing competitive intensity changes the expected performance outcomes of

different actions, it may change the very nature of actions taken by firms to sustain their

performance levels. Three hypotheses regarding the impact of competitive intensity on the

nature of firm competitive actions was put forth in this study. These hypotheses are

recapitulated alongside findings in table 8 below.

Table 8 – Summary of developed hypothesis and empirical tests

Hypothesis Results Empirical Findings

H1: Increasing competitive intensity will have a positive effect on the overall level of firm actions (Ch. 3.1)

Partial support The sum of actions rises only to a certain point after which actions start decreasing with increased competitive intensity (Increasing tendency for exploration does not outweigh decrease in exploitation).

H2: Increased competitive intensity will have an inverted curvilinear relationship (inverted U-shape) with firm exploitation actions (Ch. 3.2)

Yes (but not for smaller firms)

Competitive intensity has a positive impact on exploitation actions at low levels. The effect diminishes with increasing competition finally turning negative at high levels of competitive intensity.

H3: Increased competitive intensity will have a curvilinear relationship (U-shaped) with firm exploration actions (Ch. 3.3)

Partial support An increasing exponential relationship between competitive intensity and exploration actions is found. The hypothesized increasing effect of competitive intensity is then supported but positive effects of low competitive intensity cannot be found.

255

Increasing competitive intensity was argued to focus firm attention towards being competitive

in its existing domains over the short term, and hence increase the exploitation actions taken.

Owing to the decreasing returns attainable by taking exploitation actions in more competitive

markets, it was argued that intensifying competition would breed exploitation response by

incumbents up to a point where the returns to exploitation could only off-set the costs. At such

a point engaging in additional direct competitive measures would only risk accelerating

competition further, potentially diminishing returns even to earlier exploitation actions. For

instance, it is logical that firms refrain from competing on cost-efficiency by adding novel

equipment when there is an overhanging risk that rivals competitive responses will result in

lower prices, hence eroding profits from older less efficient production units. This proposed

pattern of exploitation actions as the level of competition increases (hypothesis 2) was strongly

corroborated by the results.

By the reversed logic firm exploration actions were argued to increase as function of

competitive intensity. The higher the intensity – the lower the returns from current activities

and further exploitation – the more attractive the exploration actions become. In turn, at low

levels of competition there is little pressure to be efficient and firms were thus argued to

engage in higher levels of exploration actions also at low competitive intensity. The findings

support only partial efficacy of the hypothesis. Increasing intensity did have an exponential

effect on exploration but theorized higher exploration in low competitive environments was

not found. The failure to find evidence of increasing exploration in low competitive

environments might be due to the absence of high growth periods in the sample. In hindsight,

arguments regarding increased exploration at low levels of competition, resting on the

argument that firms experiment with various ways to offer value to customers, might be

premature. It might only hold in the growth stage of industries where routines have not been

formalized in the population and low competition has not forced the emergence of dominant

designs (Brown and Eisenhardt, 1997).

Moreover, the observed increasing exploration actions did not outweigh the decrease in

exploitation actions, implying that, contrary to expectations (Hypothesis 1), the aggregate of

actions was not increasing for increasing competitive pressure over its entire spectrum. Instead,

all actions increased in a similar manner as exploitation, increasing at with growing intensity at

low levels of competition and decreasing with further increasing competition. The findings also

suggested differently pronounced effects for large and small firms to the extent that hypothesis

2 did not hold for the smaller firms.

256

Taken together the findings have significant implications for firms and managers – especially in

the pulp and paper industry setting. The highly significant economic impact of the competitive

intensity on all dependents suggests that the pressure exerted by competition plays a material

part in the dynamics governing firm behavior. Whereas prior research has mainly focused on

the pattern of actions – speed, variation, size (e.g. Smith et al., 1991; Ferrier, 2001) – largely

ignoring the nature of the actions taken; the results compliment research on competition by

suggesting a need to treat actions and responses, not as a homogenous, but to discern their

differential effects on competition and different expected returns. In doing so, this research

argued and found that the Schumpeterian action-response is not necessarily escalating for all

levels of competition (as explained above). Contrary to arguments from theory on

Hypercompetition (D’Aveni, 1994) and Red Queen competition (Barnett and Hansen, 1996) it is

likely that there comes a point were increased competition no longer generates more response.

De facto, the argument that beyond a threshold level, the propensity of firms to engage in

exploitation action will decrease is in line with classic arguments from theory on static

competition (Porter, 1980; Miller and Chen, 1994). Firms have been expected to refrain from

actions if the fear of retaliation is too great, in other words if subsequent responses will erode

the achieved benefits. While this mechanism has been used previously as an explanation to

why rivalry does not break out in industries or between pairs of close competitors (comparable

to single round games in game theory), this research shows that the same arguments hold for

sequential games on an industry level. Moreover, considering the observed decrease in

exploitation actions when competitive intensity increases and if firms can be considered

rational, findings put in question maxims such as: more active competitors’ fare better (e.g.

Smith et al., 1992). Rather than merely taking action, adding the dimension of the nature of

actions to competitive dynamics, introduces the need for taking the appropriate action.

The fact that these tendencies were observed to be more pronounced for larger firms in turn

corroborates arguments regarding the higher visibility and competitive impact of large firm’s

actions. Large firms, potentially accelerating competition by their actions, may be quicker to

refrain from making exploitation moves as such behavior in small firms was not found. The

observed dissimilar behavior of small firms is not entirely unexpected and ads to portrait of

small firms competing irregularly and by different means vis-à-vis their larger counterparts

(e.g. Chen and Hambrick, 1995).

Adding to evolutionary and organizational adaptation perspectives, the results indicate that

competition is a key force driving firm level exploration action, i.e. innovations and efforts to

257

innovate. As such the findings give evidence of an incremental view of evolution where firms

gradually increase their innovation efforts as the environment becomes more hostile, rather

than shifting through dichotomous states (see debate in Levin et al., 1999). Hence, the research

compliments insights on how industries evolve through patterns of intense competition and

innovation. Findings suggests that the threat of high competition in existing domains,

complimented by the diminished possibilities of earning satisfactory returns underpin the shift

towards more exploration in actions. The findings further echo arguments on the non-exclusive

effect of exploration on exploitation in loosely coupled units (Gupta et al., 2006), and

strengthens the already dominant view of firms paths – established routines and capabilities –

affecting the nature of firms future actions (Benner and Tushman, 2003; Nelson and Winter,

1982).

In conclusion, the results suggest that current research on competition would be

complimented largely by adding the dimension of action nature, and incorporating an

exploitation and exploration perspective. This study expands knowledge on how companies act

in relation to their environment. Findings not only enrich theory on competitive interaction,

but also give new insight to the debate on organizational adaption as a consequence of hostile

environments. Further, the study makes a methodological contribution as it advances the use

of content analysis as a creative approach to studying organizational actions (Uotila et al.,

2009), and demonstrates latest practice in analyzing overdispersed count data.

7.2 Managerial implications

The implication of this research for managers is manifold. First, understanding competitor’s

reactions is of paramount importance for managers (Coyne and Horne, 2009). The findings

forward the understanding of how competitors are likely to act given a competitive

environment and might as such help managers make more accurate scenarios predicting rival

activity. Knowing that the propensity of competitors to act within current domains (e.g. cutting

costs, improving processes and adding scale) increases when competition in the industry is less,

while the reverse holds for higher competition, should inform managers taking strategic and

competitive choices. For example, decision to engage in direct competitive behavior in

previously un-competitive environments should be made expecting increases in the amount of

direct exploitation actions of rivals to follow. Similarly, managers taking exploitation action in

already hostile settings should evaluate the outcomes expecting low increases in rival

retaliation. This might, depending on situation, significantly change anticipated outcomes,

258

which might otherwise be overly positive or negative if projections are based on either constant

or increasing likelihoods of rivalrous activity.

Second, if increased levels of competition foster more innovative behavior it suggests, in

similarity to arguments on the benefits of competition (Porter, 1991), that rivalry might increase

the competitiveness and renewal of firms. The implication for managers and policy makers

alike is that seeking or even inviting confrontation and competition, at least to a certain degree,

can be beneficial for overcoming barriers to innovation and change (e.g. short sightedness and

excessive emphasis on exploitation). The implication would thus be that instead of attempting

to create regulatory barriers to competition or implicit cooperation firms should attempt to

spur levels of competition. Over time, such induced exploratory activity has been argued to

accumulate to experience getting locked into routines, which as suggested by management

scholars should enhance exploratory behavior even further (Nelson and Winter, 1982).

Moreover, worries that intense competition focuses managerial attention exclusively on short

term competitiveness appears to be premature as findings suggest the opposite. Instead,

intense competition might force competitors to break patterns of exploiting current know-how

when their returns are eroded, and attempt to create new capabilities. From an industry

viewpoint then, increased competitive intensity may be needed for successful market renewal

in the long run and for sustaining viability in the face of competing substitute markets. Worries

that the industry is disappearing may thus be somewhat self-solving as the resulting higher

levels of competition will lead to higher explorations and hence higher chances of industry

rejuvenation. However, a caveat is in place; the findings imply that firms indeed increase their

exploration efforts but do not reveal anything on the success of taking such action.

7.3 Limitations

Several limitations applying to the conducted research merit discussion. First, the single

industry research design leaves doubts on the external validity of the findings. Without further

analysis caution should be applied in generalizing the results beyond similar settings. Indeed,

the paper industry entails several idiosyncrasies which might affect outcomes. First, the

structural contingencies – high fixed and exit costs, industry fragmentation, and commodity

nature of goods – might affect the rewards to and implications of exploitation or exploration

actions. The structural factors in the paper industry, implying a high interconnectedness

between firms, is fortuitous for research on competitive interaction, but also further suggests

that findings may perhaps not hold for alternate, less interconnected settings. These

259

weaknesses notwithstanding, the fundamental arguments developed should still be applicable

to other similar sectors.

Second, in spite of corrective measures the sample selection bias caused by the non-random

sample (Echambadi et al., 2006), may still skew findings towards large firms. The implication

for reliability should however be limited as the covariate of size was controlled in multiple ways

and its implications thoroughly debated. Nonetheless, the results indicate unpredicted

behavior in small firms.

Third, in the data gathering process a threat to validity arises from the construct measurement,

in this case the terms applied in identifying the actions (Echambadi et al., 2006). To minimize

this threat of poor construct measurement related bias, variables operationalization was rooted

in previous theoretical and empirical work, as described in the methodology section, and

applied direct measures (MacKenzie, 2003). Further, testing the categorization through having

experts assess and categorize the news items based on personal judgment, not coding schemes,

yielded highly similar results.

Fourth, there is a potential for systemic error in the recording of company actions in newsfeeds.

To limiting the threat the study gathered actions from a source combining in turn multiple

sources – including all the significant newspapers and journals covering the field. Following

previous competitive dynamics scholars (Ferrier, 1999; Miller and Chen, 1994), there is low

likelihood that actions go unnoticed by industry specialty press. Moreover, actions not noted

by press, which then go unobserved, have been argued to have (a) little significance and hence

no material implication for competition, and (b) be impossible to react to for competitors and

thus carry no implications for competition (Ferrier, 1999; Smith et al. 1992).

260

REFERENCES

ANAND, J., L. F. MESQUITA & R. S. VASSOLO. 2009. The dynamics of multimarket

competition in exploration and exploitation activities. Academy of Management Journal,

52, 802-821.

BAIN, S. J. 1956. Barriers to new competition. Cambridge, MA: Harvard University Press.

BANBURY, C. M. & W. MITCHELL. 1995. The effect of introducing important incremental

innovations on market share and business survival. Strategic Management Journal, 16,

161-182.

BARNETT, W. P. 1997. The dynamics of competitive intensity. Administrative Science

Quarterly, 42, :128-160.

BARNETT, W. P. & M. T. HANSEN. 1996. The Red Queen in organizational evolution. Strategic


BARNEY, J. 1991. Firm resources and sustained competitive advantage. Journal of Management,

17, 99-120.

BAUM, J. A. C. & H. J. KORN. 1999. Dynamics of dyadic competitive interaction. Strategic


BECKMAN, C. M. 2006. The influence of founding team company affiliations on firm behavior.

Academy of Management Journal, 49, 741-758.

BENNER, M. J. 2002. Process management and technological innovation: A longitudinal study

of the photography and paint industries. Administrative Science Quarterly, 47, 676-706.

BENNER, M. J. & M. L. TUSHMAN. 2003. Exploitation, exploration, and process management:

The productivity dilemma revisited. Academy of Management Review, 28, 238-256.

BOWMAN, E. H. 1982. Risk seeking by troubled firms. Sloan Management Review, 23, 33-42.

BROWN, S. L. & K. M. EISENHARDT. 1997. The art of continuous change: Linking complexity

theory and time-paced evolution in relentlessly shifting organizations. Administrative


CAMERON, K. S., M. U. KIM & D. A. WHETTEN. 1987. Organizational effects of decline and

turbulence. Administrative Science Quarterly, 32, 222-240.

CAMERON, C. A. & P. TRIVEDI. 2009. Microeconometrics using stata. College station, TX:

Stata Press

CASTROGIOVANNI, G. J. 1991. Environmental munificence - a theoretical assessment.

Academy of Management Review, 16, 542-565.

CAVES, R. E. & P. GHEMAWAT. 1992. Identifying mobility barriers. Strategic Management

Journal, 13, 1-12.

CHEN, M. J. 1996. Competitor analysis and interfirm rivalry: Toward a theoretical integration.

Academy of Management Review, 21, 100-134.

CHEN, M. J. & D. C. HAMBRICK. 1995. Speed, stealth, and selective attack - how small firms

differ from large firms in competitive behavior. Academy of Management Journal, 38,

453-482.

261

CHEN, M. J. & I. C. MACMILLAN. 1992. Nonresponse and delayed-response to competitive

moves - the roles of competitor dependence and action irreversibility. Academy of


CHEN, M. J., K. G. SMITH & C. M. GRIMM. 1992. Action characteristics as predictors of

competitive responses. Management Science, 38, 439-455.

CHEN, M. J., K. H. SU & W. P. TSAI. 2007. Competitive tension: The awareness-motivation-

capability perspective. Academy of Management Journal, 50, 101-118.

CHILD, J. 1972. Organizational structure, environment and performance - role of strategic

choice. Sociology-the Journal of the British Sociological Association, 6, 1-and.

COHEN, J. 1960. A coefficient of agreement for nominal scales. Educational and Psychological

Measurement, 20, 37-46.

COYNE, K. P. & J. HORN. 2009. Predicting Your Competitor's Reaction. Harvard Business

Review, 87, 90-+.

CYERT, R. & J. MARCH. 1963. A Behavioral Theory of the Firm. Englewood Cliffs, NJ: Prentice-

Hall.

D’AVENI, R. 1994. Hypercompetition: Managing the dynamics of strategic maneuvering. New

York, NY: Free Press.

DIESEN, M. 2007. Economics of pulp and paper industry. Paper making science and technology.

Jyväskylä, FI: Gummerus.

DIMAGGIO, P. J. & W. W. POWELL. 1983. The iron cage revisited - institutional isomorphism

and collective rationality in organizational fields. American Sociological Review, 48, 147-

160.

DOBREV, S. D. & T. Y. KIM. 2006. Positioning among organizations in a population: Moves

between market segments and the evolution of industry structure. Administrative


DUTTON, J. E. & S. E. JACKSON. 1987. Categorizing strategic issues - links to organizational

action. Academy of Management Review, 12, 76-90.

ECHAMBADI, R., B. CAMPBELL & R. AGARWAL. 2006. Encouraging best practice in

quantitative management research: An incomplete list of opportunities. Journal of

Management Studies, 43, 1801-1820.

FERRIER, W. J., K. G. SMITH & C. M. GRIMM. 1999. The role of competitive action in market

share erosion and industry dethronement: A study of industry leaders and challengers.


FERRIER, W. J. 2001. Navigating the competitive landscape: The drivers and consequences of

competitive aggressiveness. Academy of Management Journal, 44, 858-877.

FERRIER, W. J. & LEE, H. 2002. Strategic Aggressiveness, Variation, and Surprise: How the

Sequential Pattern of Competitive Rivalry Influences Stock Market Returns. Journal of

Managerial Issues, 14, 162-180

GHEMAWAT, P. & JERI COSTA. 1993. The organizational tension between static and dynamic

efficiency. Strategic Management Journal, 14, 59-73.

262

GHOSAL, V. & U. NAIR-REICHERT. 2009. Investments in modernization, innovation and gains


536-547.

GRESOV, C., H. A. HAVEMAN & T. A. OLIVA. 1993. Organizational design, inertia and the

dynamics of competitive response. Organization Science, 4, 181-208.

GREVE, H. R. 1998. Managerial cognition and the mimetic adoption of market positions: What

you see is what you do. Strategic Management Journal, 19, 967-988.

GREVE, H. R. 2007. Exploration and exploitation in product innovation. Industrial and

Corporate Change, 16, 945-975.

GRIFFIN, A. 1997. PDMA research on new product development practices: Updating trends and

benchmarking best practices. Journal of Product Innovation Management, 14, 429-458.

GUPTA, A. K., K. G. SMITH & C. E. SHALLEY. 2006. The interplay between exploration and

exploitation. Academy of Management Journal, 49, 693-706.

HAMBRICK, D. C. & R. A. D’AVENI. 1988. Large corporate failures as downward spirals.

Administrative Science Quarterly, 33, 1-23.

HANNAN, M. T. & J. FREEMAN. 1977. Population ecology of organizations. American Journal of

Sociology, 82, 929-964.

HANNAN, M. T. & J. FREEMAN. 1984. Structural inertia and organizational-change. American

Sociological Review, 49, 149-164.

HARRIGAN, K. R. & M. E. PORTER. 1983. End-game strategies for declining industries. Harvard

Business Review, 61, 111-120.

HAUSMAN, J., B. H. HALL & Z. GRILICHES. 1984. Econometric-models for count data with an

application to the patents r and d relationship. Econometrica, 52, 909-938.

HAVEMAN, H. A. 1993. Organizational size and change - diversification in the savings and loan

industry after deregulation. Administrative Science Quarterly, 38, 20-50.

HE, Z. L. & P. K. WONG. 2004. Exploration vs. exploitation: An empirical test of the

ambidexterity hypothesis. Organization Science, 15, 481-494.

HOETKER, G. 2007. The use of logit and probit models in strategic management research:

Critical issues. Strategic Management Journal, 28, 331-343

HOLMQVIST, M. 2004. Experiential learning processes of exploitation and exploration within

and between organizations: An empirical study of product development. Organization

Science, 15, 70-81.

HOPKINS, H. D. 2003. The response strategies of dominant US firms to Japanese challengers.

Journal of Management, 29, 5-25.

JANSEN, J. J. P., F. A. J. VAN DEN BOSCH & H. W. VOLBERDA. 2006. Exploratory innovation,

exploitative innovation, and performance: Effects of organizational antecedents and

environmental moderators. Management Science, 52, 1661-1674.

KAHNEMAN, D. & A. TVERSKY. 1979. Prospect theory - analysis of decision under risk.

Econometrica, 47, 263-291.

263

KATILA, R. & G. AHUJA. 2002. Something old, something new: A longitudinal study of search

behavior and new product introduction. Academy of Management Journal, 45, 1183-1194.

KIM, H. E. & J. M. PENNINGS. 2009. Innovation and Strategic Renewal in Mature Markets: A

Study of the Tennis Racket Industry. Organization Science, 20, 368-383.

KIM, W. C. & R. MAUBORGNE. 1999. Creating new market space. Harvard Business Review, 77,

183-+.

KIM, W. & R. MAUBORGNE. 2005. Blue Ocean Strategy. Boston, MA: Harvard Business School

Press.

KING, G. & W. LOWE. 2003. An automated information extraction tool for international

conflict data with performance as good as human coders: A rare events evaluation

design. International Organization, 57, 617-+.

KIRZNER, I. 1973. Competition and entrepreneurship. Chicago, IL: University of-Chicago Press.

KLEPPER, S. 1996. Entry, exit, growth, and innovation over the product life cycle. American

Economic Review, 86, 562-583.

KOZA, M. P. & A. Y. LEWIN. 1998. The co-evolution of strategic alliances. Organization Science,

9, 255-264.

KRIPPENDORFF, K. 1980 and 2004 (2nd ed.). Content Analysis. An Introduction to its

Methodology. London, UK: The Sage Commtext Series, Sage Publications Ltd.

LAESTADIUS, S. 1998. The relevance of science and technology indicators: the case of pulp and

paper. Research Policy, 27, 385-395.

LAMBERG, J. A. 2005. Strategic actions and path dependence: Profiles and archetypes of

competitive behavior in a global industry. Helsinki, FI: Helsinki University of

Technology.

LAMBERG, J. A., & J. OJALA. 2006. Evolution of competitive strategies in global forestry

industries: Introduction. In Lamberg et al. (Eds.): The evolution of competitive strategies

in global forestry industries: Comparative Perspectives, 1–30. Dordrecht, NE: Springer.

LAMBERG, J. A., H. TIKKANEN, T. NOKELAINEN & H. SUUR-INKEROINEN. 2009.

competitive dynamics, strategic consistency, and organizational survival. Strategic


LANT, T. K. 1992. Aspiration level adaptation - an empirical exploration. Management Science,

38, 623-644.

LAVER, M., K. BENOIT & J. GARRY. 2003. Extracting policy positions from political texts using

words as data. American Political Science Review, 97, 311-331.

LEVINTHAL, D. A. 1991. Random-walks and organizational mortality. Administrative Science

Quarterly, 36, 397-420.

LEVINTHAL, D. A. & J. G. MARCH. 1993. The myopia of learning. Strategic Management

Journal, 14, 95-112.

LEVITT, B. & J. G. MARCH. 1988. Organizational learning. Annual Review of Sociology, 14:319-

340.

264

LEWIN, A. Y., C. P. LONG & T. N. CARROLL. 1999. The coevolution of new organizational

forms. Organization Science, 10, 535-550.

LITTLE, R. J. A. 1992. Regression with missing X’s – a review. Journal of the American Statistical

Association, 87, 1227-1237.

LUMPKIN, G. T. & G. G. DESS. 2001. Linking two dimensions of entrepreneurial orientation to

firm performance: The moderating role of environment and industry life cycle. Journal

of Business Venturing, 16, 429-+.

MACMILLAN, I., M. L. MCCAFFERY & G. VANWIJK. 1985. Competitors responses to easily

imitated new products - exploring commercial banking product introductions. Strategic


MARCH, J. G. 1991. Exploration and exploitation in organizational learning. Organization

Science, 2, 71–87.

MARCH, J. G. & Z. SHAPIRA. 1992. Variable risk preferences and the focus of attention.

Psychological Review, 99, 172-183.

MCGAHAN, A. M. & B. S. SILVERMAN. 2001. How does innovative activity change as industries

mature? International Journal of Industrial Organization, 19, 1141-1160.

MCGAHAN, A. M. 2004. How industries change. Harvard Business Review, 82, 86-+.

MILLER, D. & M. J. CHEN. 1994. Sources and consequences of competitive inertia - a study of

the united-states airline industry. Administrative Science Quarterly, 39, 1-23.

MILLER, D. & M. J. CHEN. 1996a. Nonconformity in competitive repertoires: A sociological

view of markets. Social Forces, 74, 1209-1234.

MILLER, D. & M. J. CHEN. 1996b. The simplicity of competitive repertoires: an empirical

analysis. Strategic Management Journal, 17, 419-439.

MILLER, D. & P. H. FRIESEN. 1982. Innovation in conservative and entrepreneurial firms - 2

models of strategic momentum. Strategic Management Journal, 3, 1-25.

NADKARNI, S. & P. S. BARR. 2008. Environmental context, managerial cognition, and strategic

action: an integrated view. Strategic Management Journal, 29, 1395-1427.

NELSON, R. R. & S. G. WINTER. 1982. The Schumpeterian tradeoff revisited. American

Economic Review, 72, 114-132.

NORTON, E. C., WANG, H. & AI, C. 2004. Computing interaction effects and standard errors in

logit and probit models. Stata Journal, 4, 154-167

OCASIO, W. 1995. The enactment of economic adversity - a reconciliation of theories of failure-

induced change and threat-rigidity. In Research in Organizational Behavior: an Annual

Series of Analytical Essays and Critical Reviews, 17.

OJALA, J., LAMBERG, J.-A., A. AHOLA & A. MELANDER. 2006. The Ephemera of success:

Strategy, structure and performance in the forest industries. In Lamberg et al. (Eds.):

The evolution of competitive strategies in global forestry industries: Comparative

Perspectives, 1–30. Dordrecht, NE: Springer.

PESENDORFER, M. 2003. Horizontal mergers in the paper industry. Rand Journal of

Economics, 34, 495-515.

265

PETERAF, M. A. 1993. The cornerstones of competitive advantage - a resource-based view.


PORTER, M. E. 1980. Competitive Strategy: Techniques for analyzing Industries and

Competitors. New York, NY: Free Press.

PORTER, M. E. 1985. Competitive advantage: Creating and sustaining superior performance.

New York, NY: Free Press.

PORTER, M. E. 1991. Towards a dynamic theory of strategy. Strategic Management Journal, 12,

95-117.

PRAHALAD, C. K. & G. HAMEL. 1990. The core competence of the organization. Harvard

Business Review, 68, 79-91.

RUMELT, R. P. 1974. Strategy, Structlure and Economic Performance. Boston, MA: Harvard

University Press.

SAJASALO, P. 2003. Strategies in transition. The Internationalization of the Finnish Forest

Industry Companies. Jyväskylä, FI: University of Jyväskylä.

SCHUMPETER, J. 1934. The theory of economic development. Cambridge, MA: Harvard

University Press.

SCHUMPETER, J. 1950. Capitalism, socialism, and democracy. New York, NY: Harper and Row.

SILVERMAN, B. S. & J. A. C. BAUM. 2002. Alliance-based competitive dynamics. Academy of


SINGH, J. V. 1986. Performance, slack, and risk-taking in organizational decision-making.


SMITH, K. G., C. M. GRIMM, M. J. GANNON & M. J. CHEN. 1991. organizational information-

processing, competitive responses, and performance in the united-states domestic

airline industry. Academy of Management Journal, 34, 60-85.

SMITH, K. G., C. M. GRIMM, S. WALLY & G. YOUNG. 1997. Strategic groups and rivalrous firm

behavior: Towards a reconciliation. Strategic Management Journal, 18, 149-157.

STAW, B. M., L. E. SANDELANDS & J. E. DUTTON. 1981. Threat-rigidity effects in

organizational-behavior - a multilevel analysis. Administrative Science Quarterly, 26,

501-524.

STUART, T. E. & J. M. PODOLNY. 1996. Local search and the evolution of technological

capabilities. Strategic Management Journal, 17, 21-38.

TEECE, D. J. 1986. Profiting from technological innovation - implications for integration,

collaboration, licensing and public-policy. Research Policy, 15, 285-305.

TEECE, D. J., G. PISANO & A. SHUEN. 1997. Dynamic capabilities and strategic management.


TRANFIELD, D., D. DENYER & P. SMART. 2003. Towards a methodology for developing

evidence-informed management knowledge by means of systematic review. Brittish

journal of management, 14, 207-22

266

UOTILA, J., M. MAULA, T. KEIL & S. A. ZAHRA. 2009. Exploration, exploitation, and financial

performance: analysis of s&p 500 corporations. Strategic Management Journal, 30, 221-

231.

UTTERBACK, J. M. & F. F. SUAREZ. 1993. Innovation, competition, and industry structure.

Research Policy, 22, 1-21.

VAN DE VEN, A. H. & P. E. JOHNSON. 2006. Knowledge for theory and practice. Academy of

Management Review, 31, 802-821.

VENKATRAMAN, N. 1989. Strategic orientation of business enterprises - the construct,

dimensionality, and measurement. Management Science, 35, 942-962.

VERMEULEN, F. & H. BARKEMA. 2001. Learning through acquisitions. Academy of


WALSH, J. P. 1995. Managerial and organizational cognition - notes from a trip down memory

lane. Organization Science, 6, 280-321.

WESTPHAL, J. D. & E. J. ZAJAC. 1998. The symbolic management of stockholders: Corporate

governance reforms and shareholder reactions. Administrative Science Quarterly, 43,

127-153.

WHETTEN, D. A. 1989. What constitutes a theoretical contribution. Academy of Management

Review, 14, 490-495.

WIGGINS, R. R. & T. W. RUEFLI. 2005. Schumpeter's ghost: Is hypercompetition making the

best of times shorter? Strategic Management Journal, 26, 887-911.

YOUNG, G., K. G. SMITH & C. M. GRIMM. 1996. "Austrian'' and industrial organization

perspectives on firm-level competitive activity and performance. Organization Science,

7, 243-254.

ZAMMUTO, R. & CAMERON, S. 1985. Environmental decline and organizational response. In

Research in Organizational Behavior: an Annual Series of Analytical Essays and Critical

Reviews, 7.

267

CHAPTER 8

SUCCESS STRATEGIES IN DECLINING INDUSTRIES: A CASE SURVEY

ANTTI SIHVONEN

Aalto University Helsinki School of Economics [email protected]

1 INTRODUCTION

The state of decline has become more prevalent in our western societies due to the present

global recession. The Oxford English Dictionary provides the following two related definitions

of the word decline:

verb: become smaller, weaker, or less in quality or quantity

noun: a gradual and continuous loss of strength, numbers, or value.

In short, decline, as a word, has two meanings. First, decline implies diminishing of an entity.

Second, decline implies gradualism and continuity. Therefore, decline in itself implies explicitly

weakening or loss, but it also implicitly implies preservation. Thus, the word itself has a

dualistic meaning. This idea of duality plays a central role in this research.

1.1 Research area and problem

Industries are not static but evolve over time and this evolution may lead to growth but also to

decline of an industry (Klepper 1996; Jonavick and McDonald 1994; Gort and Klepper 1982;

Ghemawat and Nalebuff 1985; 1990). As scholars have an implicit bias towards studying growth,

268

the decline of industries has received less attention, and our knowledge of the declining

industries is still imperfect (Cameron and Zammuto 1983; Whetten 1980). This opens up a

fertile ground for new research, on which this research focuses on.

This research concentrates on declining industries as the context of operations for

organizations. Declining industry, for the purposes of this research, can be defined as a

deteriorating environment that leads to diminished opportunities for organization operating in it

(Ghemawat and Nalebuff 1985; 1990;Whetten 1980; Zammuto and Cameron 1985, p. 226). The

definition of a declining industries explicitly states that the environment deteriorates, it also

implicitly states that it offers continuity and possibilities for organizations, as otherwise

industries would not decline but only seize to exist. Therefore, success in this research will be

defined in relation to environment, meaning that the organizations which are able to stay in

business while others exit are successful. This is congruent with the population ecology

perspective of success, where success can be defined as selection, where survival equals success

as the organization continues to exist while other seize to do so (Aldrich 2008, p. 29; Hannan

and Freeman 1977). Therefore, deriving from the notions, the research problem of this research

problem, the explanandum, can be defined as:

Why some organizations succeed in declining industries, while others do not?

After establishing the explanandum, the question to be explained, basis for the explanans, the

answer to the question, has to be established.

The success of organizations is explained in this research with the concept of strategy. The

reason for choosing strategy as the explanans is that on the level of organization as a whole, the

direction, and the execution of this direction is guided by strategy (Porter 1985, p. 1; Prahalad

and Hamel 1994). Strategy, for the purposes of this research, is defined as a pattern in a stream of

decisions made by an organization (Mintzberg 1978; Mintzberg and Waters 1982; 1985; Miles

and Snow 2007, p. 7). This means that organizations exhibit patterns of behavior that can be

labeled as the strategy of an organization, as they guide the organization to a direction.

Therefore, the reasons for success in declining industries will be explained by the patterns of

decisions, labeled strategy, that organizations use to steer their selves in this context.

269

Widely used way of classifying strategy of an organization is the Miles and Snow typology,

which is built around archetypes of strategies (Hambrick 1983; Miles et al. 1978; Miles and Snow

2003; Zahra and Pearce 1990). This research adopts this typology, as it has been identified to be

a comprehensive way of classifying strategy of an organization (Zahra and Pearce 1990; Segev

1989).

1.2 Research gap

This thesis seeks to contribute to two central theoretical discourses where research gaps exist;

these are the strategic choice discourse and the population ecology discourse (Aldrich 1979;

Hannan and Freeman 1977; Child 1972; 1997; Miles et al. 1978). By choosing these two

perspectives, it is possible to analyze success from both the level of single organization and the

population (Zammuto 1988). A closer description of the interaction of population ecology and

strategic choice perspective can be found from the theoretical synthesis.

Firstly, this research aims to produce an empirical investigation of success strategies in

declining industries that aggregates conclusions from a large case material. As organizations

experience different types of decline, a more fine-grained approach is needed to describe the

conditions of decline (Cameron 1983; Zammuto 1983). Therefore, this research utilizes a

typology of decline by Zammuto and Cameron (1985) to identify and describe different decline

conditions. This research aims at identifying what kind of strategies lead to success in each of

these decline conditions.

Secondly, this research extends the research on the Miles and Snow typology, which has been

studied in other parts of the industry life cycle but not the decline phase (Hambrick 1983; Zahra

and Pearce 1990; Zajac and Shortell 1989). These studies together verify that all the strategy

types are present in industries that are in other stages of the life cycle, and that performance

differences exist between the strategy types. Extending the typology into the last stage of the

industry life cycle can bring new insight of the success of these strategies, and how these

strategies behave in adverse environments, and in relation to other stages of the life cycle.

270

1.3 Research questions

The research problem is operationalized through the use of three research questions. These

questions are formulated so that they expand the research problem into three distinct

questions that can be answered within the scope of this research.

The first research question aims at opening the array of success strategies in declining

industries, establishing the strategy types that can succeed in declining industries. The

different strategy types are represented by the strategic archetypes of the Miles and Snow

typology (Hambrick 1983; Miles et al. 1978; Miles and Snow 2003; Zahra and Pearce 1990). This

logic leads to presenting the first research question:

1. What strategy types are successful in declining industries?

The second research question is built on the success strategies identified by the first question.

As decline types are diverse, the second question aims at defining what kind of strategies

succeed in which decline conditions. This would establish fit between strategy types and the

decline conditions (Prescott 1986; Venkatraman and Camillus 1984; Venkatraman and Prescott

1990; Zammuto and Cameron 1985). This logic leads to presenting the second research

question:

2. Which strategy types succeed in which types of decline?

The third research question aims extending the second question. After establishing the success

of different strategy types in different decline conditions, the third question aims at producing

higher order conclusion of the success strategies. This leads to presenting the third research

question:

3. How is the strategy typology, as a whole, aligned to the decline conditions?

This would yield findings at the level both typologies as whole and how these two typologies

are aligned to each other when success is used as a constant variable. As the function of a good

research is not only to record occurrences in the past but also to produce theories predicting

271

the future, the function of the third question is to open this path for the research (Eisenhardt

1989; Eisenhardt and Graebner 2007).

These three research question guide the research and its direction. By answering the three

research questions, an answer for the research problem can be generated, and the purpose of

this research can be reached.

1.4 Previous research

This subsection will briefly discuss the previous research concerning declining industries to

produce a concise overview of research in this field. Previous research in declining industries

has focused on three different streams of research that are strategic management, population

ecology and exit behavior of organizations from the declining industry (Ghemawat and

Nalebuff 1985;1990; Harrigan 1980a; 1980b; Harrigan and Porter 1985; Zammuto and Cameron

1985; )

Strategic management research on declining industries culminates around the works of

Harrigan. These articles focus on environmental attractiveness and structural factors and

relative competitive strength of the organization. Deriving from these works, Harrigan

developed strategic options for organization for organization operating in declining industries.

These included strategies such as harvest the industry, exit quickly, shrink selectively or focus

on a niche. Although being managerially relevant, these studies lack grounding on explicit

types of decline and suggestions on what kind of responses are effective in these contexts.

(Harrigan 1980a; 1980b; Harrigan and Porter 1985) These ideas initially sparked the interest for

choosing the approach for this research.

Population ecology research on declining industries has focused greatly around the work of

Zammuto and Cameron. Their research culminated around the generation of a typology of

different decline conditions. This typology was intended to depicting why organizations face

different types of decline and why there is variation in the prescriptions on how to respond to

decline. (Cameron 1983; Cameron and Zammuto 1983; Zammuto 1983; Zammuto and Cameron

1985) This typology was also adopted in this research to depict different types of decline.

272

Discourse on exit behavior crystallizes around the reasoning of why organization and which of

them exit from a declining industry when demand begins to shrink (Ghemawat and Nalebuff

1985;1990; Reynolds1988). The discourse focuses on unveiling the underlying reasoning for exit

behavior of organizations. As this study concentrates on success in declining industries, the

discourse of exit behavior is complimentary to the current research, but these two still have

different focus.

2 THEORETICAL APPROACH TO THE PHENOMENON

This section presents the main theoretical approaches used in this research and builds the

theoretical framework. First, industry life cycles are introduced. The decline stage of the

industry life cycle and different decline conditions are emphasized. The concept of strategy is

presented as well as the strategy typology of Miles and Snow (2003) is presented. In the

following theoretical discussion I will draw from these approaches to formulate a theoretical

framework.

2.1 Industry life cycles

Industry life cycles can be defined as the way in which evolution occurs in industries, particularly

those that are technologically progressive (Klepper 1996; 1997; Jonavick and MacDonald 1994;

Gort and Klepper 1982; Utterback and Abernathy 1975). The theory of Industry life cycles is

based on product life cycle concept (Abernathy and Clark 1985; Utterback and Abernathy 1975).

This concept was first introduced in the marketing literature but from where it has been

adapted to industry evolution literature as it has been shown to capture the way how industries

evolve (Klepper 1997).

Depending on the approach, industry life cycle has three or four stages. Authors such as

Williamson (1975, p. 215) recognize only three stages in the life cycle, and in so doing leave the

decline stage only as an afterthought. Instead this research adopts the four stage view of

industry life cycles separating the decline as a distinct stage in the life cycle. In the following

four paragraphs these stages will be described in detail.

273

In the introduction stage market volumes are low, uncertainty of the industry is high, product

designs are evolving and are possibly primitive, and production technology is still developing

(Agarval and Bayus 2004; Filson 2002; Klepper 1996; 1997; Jonavick and MacDonald 1994).

Many firms enter the industry in this stage and a variation of different versions of the product

is offered to the customers (Klepper 1996; 1997). Therefore, introduction is a stage where empty

space in the market is abundant and the actual product is not yet clearly defined.

In the growth stage output of the industry is high, the design of the offering starts to stabilize

and find its final form and specialized production equipment suited for the industry starts to

overtake the old equipment (Filson 2002; Klepper 1996; 1997; Jonavick and MacDonald 1994;

Utterback and Abernathy 1975). Therefore growth stage marks an initial stabilization of the

industry, in other words the product design and the method of producing are beginning to

become unified and standardized.

Maturity stage is the third stage in the life cycle as it marks the final stabilization of the

industry. Mature industry is marked by slow growth in output, further decline in the entry of

new organizations in the industry, stabilization of the market shares between the actors in the

market. This results in a shift in emphasis from innovation to increasing manufacturing

efficiency and enhancing marketing and management activities. (Klepper 1996; 1997; Jonavick

and MacDonald 1994; Utterback and Abernathy 1975)

The last stage of the life cycle is the decline stage, which is the focus of this research. The

decline stage of the life cycle has received little attention in the industry life cycle literature as

can be seen from the research made by Quinn and Cameron (1983). The definition of the

decline stage is built around the concept of niche. Niche can be defined as an environmental

location populated by a number of organizations (Zammuto 1983; Zammuto and Cameron 1985,

p. 226). The decline stage can be defined as a condition where the niche cannot support the

amount of activities, thus diminishing the carrying capacity of the niche (Cameron and

Zammuto 1983; Zammuto and Cameron 1985, p. 228). Obvious determinant of decline is

shrinking demand, resulting in capacity reductions among the organizations and pressure to

exit the industry (Ghemawat and Nalebuff 1985). Other possible determinants include changes

274

in demand, government regulations or changes in the availability of resources (Zammuto and

Cameron 1985, p. 226).

Figure 1: Product life cycle, adapted from Cox (1967)

A life cycle is usually portrayed as a bell shaped curve (figure 1 above), where the stages follow

each other in an orderly manner (Cox 1967; Cunningham 1969; Dean 1950). Despite the fact

that the bell shaped curve is the most commonly used life cycle curve, 12 different life cycle

trajectories can be identified (Rink and Swan 1979). Illustrations of different the life cycle

patterns are presented in appendix 3. This observation undermines the general belief that

declining industries always end in termination and are a place from which one should try to

exit immediately.

The undervaluation of the decline stage in the life cycle rises from an implicit conception that

growth is the only goal of an organization (Cameron and Zammuto 1983; Whetten 1980). This

results in a tendency to study growth rather than other conditions.

The decline stage, on the other hand, can also be seen as a waiting game where organizations

exit and just few end up staying (Ghemawat and Nalebuff 1990; 1985). Resulting from this, a

declining industry can be seen as a post red ocean where, when the dust settles, a small

275

population of organizations can survive and can be safe from competition. Therefore, this

exemplifies a post red ocean strategy achieved by hanging on in the industry so long that

competitor’s end up exiting the industry (Kim and Mauborne 2004). In this vein a declining

industry can be seen more as an opportunity than a threat.

The traditional bell shaped life cycle curve leads to termination. However, complete extinction

appears to be missing from the industry life cycle discourse. Central studies of the industry life

cycle, exemplify industries that do not end up in termination (Klepper 1996; 1997; Jonavick and

MacDonald 1994; Gort and Klepper 1982). The concept of death is absent from the discourse

and the term renewal is rather used to describe the development of an industry.

Although the industry life cycle concept has been widely accepted, the research on has

limitations. Bulk of the research has focused on manufacturing industries. Therefore, research

on non-manufacturing industries is scarce and undermines the reliability of the concept

outside the field of manufacturing. Second, industries with rapid technological development

and frequent discontinuities may not be applicable, as they develop too rapidly to fit into the

traditional life cycle model (Anderson and Tushman 1990). This implies that in some cases the

life cycle may not follow the standard bell shaped curve but differentiates from it. In addition,

as product life cycle can take many forms (see appendix 3), it would be logical that industry life

cycles can take many forms as well.

The next subsection will discuss the decline stage of the life cycle in depth. From here onwards

this stage will be referred with the term environmental decline. As the industry life cycle does

not always follow the bell shape curve, environmental decline is used to determine reaching of

a decline stage, whether it leads to termination or not.

2.2 Environmental decline

This research adopts a population ecology perspective to the decline advocated by Zammuto

and Cameron (1985). Population ecology is a field of research which studies the behavior of

populations of organizations, and especially, their relationship to the changing environment

(Aldrich 1979, p. 27; Hannan and Freeman 1977). Population ecology views environment as

276

deterministic, in which environmental pressures determine the successful organizations (Aldrich

1979, p. 27; Hannan and Freeman 1977; Zammuto 1988). Single organization in the population

is a semi passive entity that is subject to environmental selection and in some cases able for

extreme adaptation (Hannan and Freeman 1977).

The reason for choosing this perspective on decline is twofold. First, this perspective aims at

explaining why differences exist in decline conditions, thus expanding the concept of decline

into a typology rather than keeping it as a single phenomenon. This enables a more fine-

grained approach to the phenomenon of environmental decline. Second, through adopting this

view, it enables a tighter definition of the group under scrutiny. As the term industry is vague,

this research adopts the ecological niche as the focus of study. To clarify this definition, niche

can be defined as an environmental location that is populated by a number of organizations

(Cameron and Zammuto 1983; Zammuto and Cameron 1985, p. 226). Niche is, therefore, a

habitat of a population of organizations within an industry (Taggart 1995). This definition gives

leeway as many of the cases regard global industries in which the area of interest is a local

niche.

Environmental decline in this research is defined as a change in the ecological niche that

diminishes the carrying capacity of the niche (Cameron and Zammuto 1983; Zammuto and

Cameron 1985, p. 228). Therefore, environmental decline views decline from a population

perspective excluding the decline of single organization. The decline of a single organization

belongs to the rubric of organizational decline that should not to be mixed up with

environmental decline (Cameron, Kim and Whetten 1987).

Niche is defined by a set of condition such as physical, biological and social conditions that

constrain the performance of the population of organization in the niche (Cameron and

Zammuto 1983; Zammuto and Cameron 1985, p. 226). Every niche has a distinct carrying

capacity that defines the level of population which a niche can support at a point in time

(Zammuto and Cameron 1985, p. 226). Changes in the carrying capacity of the niche affect the

potency of success and survival of organizations in a given niche. Carrying capacity of a niche

can diminish in two major ways resulting in environmental decline. Firstly, these could be

changes that occur for reasons that are outside the control of the population, for example,

277

changes in the amount of customers, or, secondly, changes that arise from purposeful actions

inside the population, for example technological advancements (Zammuto and Cameron 1985,

p. 226, 227).

In the ecological niche, each organization inhabits an organizational domain that is the part of

the niche in which an organization operates (Zammuto and Cameron 1985, p. 227). This domain

is defined by the clients the organization serves, the technology the organization uses and the

product or service the organization produces (ibid.). In conditions of environmental decline, it

is the organizational domain that the organization adjusts in order to respond to the decline.

2.2.1 Determinants of the type of environmental decline

Zammuto and Cameron (1985, p. 228-231) define two types of changes in the configuration of an

ecological niche that can result in environmental decline. These two types regard the way how

the environment changes and act as one dimension of the change. The second dimension of

change regards the continuity of change in the environment. The first dimension of change

regards how the environment changes

Change in niche size refers to the diminishment of activities that the niche can support (Cameron

and Zammuto 1983; Zammuto and Cameron 1985, p. 228). This can occur due to a variety of

reasons such as shrinkage of available resources, decline in demand of goods or services

produced by the occupants of the niche or increased constrains posed to the niche such as

government regulations. Therefore, resulting in either reduced ability to produce

goods/services or diminishing demand of the output. Accordingly, the carrying capacity of the

niche diminishes and carrying capacity is lost.

In contrast, change in the niche shape means that the type of organizational activities supported

by the niche is changed (Cameron and Zammuto 1983; Zammuto and Cameron, p. 229). Then,

the change is a result of changes such as transformation of the production technology or a

change in demand. Change in niche shape therefore results in a transformation of the carrying

capacity of the niche to generate a new niche or modify the existing niche. Therefore, carrying

capacity is not lost but it has been transformed to support other types of activities.

278

In addition to the way how an environmental niche changes, the continuity of a change is of

essence. Change can occur in a two types that are continuous or discontinuous (Zammuto and

Cameron 1985, p. 230; Tushman and Andersson 1986). In this context continuous change is a

change that is consistent with past changes representing a long term trend whereas

discontinuous change represents change that deviates from the past and hence is unpredicted

(ibid., p. 231). The pattern of change in the environment affects the way in which organizations

react and perceive change in their environment.

2.2.2 A typology of environmental decline

Based on the type of change in the configuration of the niche and the continuity of decline,

Zammuto and Cameron (1985, p. 232) have formulated a typology of environmental decline that

incorporates the determinants of the type of change in the environment (figure 2). In the

typology, Y axis represents the type of change in the niche configuration, that is the change in

niche size or niche shape, and the X axis represents the continuity of environmental change.

Figure 2: Typology of environmental change (Zammuto and Cameron 1985, p. 232)

279

The typology introduces four different types of environmental changes. These types of

environmental change are erosion, contraction, dissolution and collapse. These conditions of

environmental change are discussed in detail in the following paragraphs.

Erosion is a condition of decline in which the niche size gradually decreases decreasing the

carrying capacity of the niche (Cameron and Zammuto 1983; Zammuto and Cameron 1985, p.

231). In this condition, the carrying capacity of the niche steadily decreases hindering the ability

of organizations in the niche to survive. In practice, erosion is a slow, steady and predictable

decline of the environment of a population of organizations. It does not cause an immediate

threat for the survival of organizations and gives time to consider alternatives (Cameron and

Zammuto 1983).

Contraction is a decline condition in which the size of the niche suddenly decreases decreasing the

carrying capacity of the niche (Cameron and Zammuto 1983; Zammuto and Cameron 1985, p.

231). This type of decline suddenly decreases the carrying capacity of the niche that cannot be

predicted by the organization before the change in the environment actually occurs. This

generates a rapid shock in the niche placing the survival of an organization under jeopardy

(Cameron and Zammuto 1983).

Dissolution represents a decline condition where a niche gradually transforms into another as the

shape of the niche changes and the carrying capacity transforms (Cameron and Zammuto 1983;

Zammuto and Cameron 1985, p. 231). This type of change represents a type of decline where the

niche evolves into another due to changes such as technological change or change in demand.

This makes the old way of operating progressively less acceptable in the environment

(Cameron and Zammuto 1983). Therefore, the niche goes through a steady evolution where the

demand, resources or other factor transforms the niche, not necessarily resulting in lost

carrying capacity.

Collapse is a decline condition in which the change in niche shape is so rapid that the existing

niche is more or less wiped out and replaced by a new niche that is able to respond to the changes

in the carrying capacity of the niche (Cameron and Zammuto 1983; Zammuto and Cameron

1985, p. 232). This represents a dramatic and rapid abolishment of the old niche that is replaced

280

by a new niche. In practice, collapse is a sudden and dramatic change in the type of activities

the niche supports which can result from changes such as rapid technological change or change

in legislation. Despite this, the carrying capacity is not lost, but transformed rapidly due to

changes in demand, resources, legislation or any other similar reason.

The value of the typology lies in its ability to explain why there are different conditions of

decline between populations, why different types of organizations in a population do well when

others do not and what types of strategies are likely to be productive in different conditions of

decline (Zammuto and Cameron 1985, p. 232, 233). For the purposes of this research, the

typology enables the condition of decline to be broken down into distinct decline conditions.

This enables a more fine-grained analysis of the success strategies in declining industries.

2.3 Strategy as a concept and its relationship to environment

In this section, strategy as a concept will be discussed. This section will begin by defining

strategy for the purposes of this research. Strategic choice perspective is introduced where the

strategy, structure process view of Miles and Snow (2003) is chosen for closer examination.

Last, the concept of strategy environment coalignment is discussed.

Strategy is a term used far and wide and it has multitude of different definitions. For this

research, strategy is defined as a pattern in a stream of decisions made by and organization

(Mitzberg 1978; Mintzberg and Waters 1982; 1985; Miles and Snow 2003, p. 7). This perspective

enables strategy to be inferred from the behavior of the organization, where one can associate

strategy with intent and structure with action (Miles and Snow 2003, p. 7). Therefore, the

strategy of the organization manifests from the stream of choices it makes, in this research in

particularly regarding adjustments to environmental changes.

2.3.1 Strategic choice perspective

Strategic choice perspective was originally introduced as a corrective view to counter the view

that organizations are designed and structured by their operational contingencies, moreover it

states that leading groups in organizations have an active role and power to influence the

281

structures of the organization (Child 1972; 1997). The strategic choice perspective views leading

groups within organizations as empowered to steer the organization to different directions and

in this vein organization is not a victim of its environment but an actor that that is able to

make choices at least partly regardless of the environment (ibid.). In this sense, strategic

choices made by the organization are reflections of its top management and their perceptions

(Hambrick and Mason 1984; Miles et al. 1978).

Key issue in the strategic choice perspective is the relationship between organization and the

environment (Child 1997). On the contrary to environmental deterministic views such as

population ecology, strategic choice perspective draws attention to the various possibilities

enabling choice on part of the organizational actors and arguing that organizations act to create

their environment (ibid., Miles and Snow 2003, p. 5). Thus, the relation with environment is not

a constraining but rather an interactive (Child 1997). This is due to the notion that

organizational decision-makers have to respond to environmental feedback, which result in

new action choices through learning and, hence, making the role of environmental both a

constraint and also enabler of choice (ibid.). Although, population ecology and strategic choice

perspectives can be regarded as competing perspectives, their relationship will be discussed in

the theoretical synthesis.

Strategic choice is especially visible in the study of U.S. firms and hospitals by Miles and Snow.

In this research they developed a typology of policies organizations adopt to adapt to changes

in their environment (Miles et al. 1978; Miles and Snow 2003; Child 1997). The next two

subsections will present the Miles and Snow strategy typology and the underlying logic.

2.3.2 Organizational strategy, structure and process

Organizational adaptation to environmental change is a highly complex phenomenon. Miles,

Snow, Meyer and Coleman (1978) separate three aspects that are of essence in adapting to

changes in environment. These are strategy, that is the way an organization defines its product-

market domain, and construct mechanisms consisting of organizational structures and

processes (Miles et al. 1978). Through modifying the mechanisms organizations pursue these

strategies. Miles and Snow (2003, p. 21) argue that organizations adapt to their environment

282

through adaptive cycles where managers have to give attention and decide on three types of

problems that define the way how the organization adapts. These problems are the

entrepreneurial problem, the engineering problem and the administrative problem (ibid.; Miles

et al. 1978). In the following paragraphs these problems will be described in detail.

Entrepreneurial problem refers to deciding the product-market domain in which the organization

operates. The manner in which organizations solve this problem manifests when organizations

commit resources to achieving objectives that are set to the product-market domain.

This means that organizations project an image that defines the market in which the

organization aims at operating in. In addition, it also defines the organizations orientation

towards the market which can be for example orientation towards innovation or efficiency. The

entrepreneurial problem is most visible in organizations that are new or rapidly growing. (Miles

et al. 1978; Miles and Snow 2003, p. 21)

Engineering problem refers to the management’s solution and operationalization of the

entrepreneurial problem. Solving this problem involves selecting appropriate technology for

producing the chosen product or service while ensuring simultaneously the proper functioning

of the production system. Solving the engineering problem formulates a manifestation of the

solution to the entrepreneurial problem identified earlier as it is guided by it. (Miles et al. 1978

Miles and Snow 2003, p. 22)

Administrative problem refers to reduction of uncertainty in the organizational systems. This

involves a balancing act between rationalization and stability of the organizational structures

versus uncertainty and instability generated by evolution of the organization through

innovation. Therefore, the way how administrative problem is solved, largely contributes to the

stance and organization takes on structural stability and innovativeness. (Miles et al. 1978;

Miles and Snow 2003, p. 23)

Despite the complexity of organizational adaptation, organizations exhibit patterns in the way

they move through the adaptive cycle and solve the three interrelated problems (Miles et al.

1978). These patterns of adaptation in solving their entrepreneurial, engineering and

administrative problems represent the strategies organizations employ. Hence, the way an

283

organization moves through the adaptive cycle defines the organizations strategy. These

patterns are reflections of the strategic choices the organizations top management makes

(Hambrick and Mason 1984; Miles et al. 1978).

By adopting this view, strategy of an organization can be inferred from the structures, processes

and decisions of an organization. This enables one to define an organizations strategy from the

way it adjusts to its environment by examining its pattern of adjustment. Therefore, strategies

can be categorized on the basis of how the organization adjusts to environmental changes.

2.3.3 Miles and Snow strategy typology

Four types of strategies can be identified on the basis of the way an organization moves

through the adaptive cycle. These strategy types are named defender, prospector, analyzer and

reactor (Miles et al. 1978; Miles and Snow 2003, p. 29). They will be defined more closely in the

following paragraphs.

Defenders are strategy types that strive to generate a protected domain in the market that they

can defend and hold (Miles et al. 1978; Miles and Snow 2003, p. 29; Snow and Hrebiniak 1980).

Defenders actively prevent competition from entering their territory by for example using

competitive pricing and superior quality products. Top managers of these organizations

concentrate on the domain the organization dominates and seldom search for new

opportunities outside their own domain (Miles and Snow 2003, p. 29; Snow and Hrebiniak

1980). Thus they focus on a small organizational domain that seldom goes through changes. As

a conclusion, defenders aim at defending their domain in all possible ways.

Defenders define their entrepreneurial problems in terms of sealing off a portion of the market

to create a stable organizational domain. This problem can be solved in a number of ways, such

as occupying a narrow and a stable domain, aggressively defending the domain with for

example competitive prices and excellent service, ignoring developments of the market and

emphasis on cautious and incremental growth. (Miles et al. 1978) All these exemplify a way of

reacting which is focused around current operations and efficiency.

284

Defenders define their engineering problem as how to produce and distribute goods as

effectively as possible to dominate the domain in which they operate. Defenders tend to solve

this problem by focusing on cost efficient technologies, single technological core and

continuous improvement in technology to remain as efficient as possible. (Miles et al. 1978)

Furthermore, this represents a coherent solution to the entrepreneurial problem identified

earlier as these choices lead to focus and efficiency.

Defenders define their administrative problem in finding ways to maintain a control over the

organization to ensure efficiency of the operations. Defenders solve this problem by having

financial and production experts as the leaders of the dominant coalition whose tenure is long,

intensive cost focused planning, functional structure with a high degree of formalization,

centralized control of the organization. (Miles et al. 1978) These represent coherent ways of

responding to the administrative problem as they enhance stability and efficiency of the

organization.

Solutions to all these three problems have costs as well as benefits. First, it is difficult to

displace a defender from its position in the niche, but if a major change in the market would

occur a defenders’ survival could be threatened. As defenders solve their entrepreneurial

problem through technological efficiency they are not able to respond to substantial

technological changes. In addition, their administrative system is well suited for maintaining

stability whereas it lacks a capability to locate new market opportunities and respond to new

products introduced to the market. (Miles et al. 1978) While at the same time the

characteristics of this type have benefits, they also have downsides that need to be dealt with.

Prospectors are almost polar opposites of the defenders as they strive to find and exploit new

product and market opportunities (Miles et al. 1978; Miles and Snow 2003, p. 29; Snow and

Hrebiniak 1980). Prospectors actively seek for new markets and opportunities while keeping

low levels of attachment to their current technological core, and therefore aim at generating

competitive advantage by being first to exploit new opportunities. Top managers of these

organizations emphasize effectiveness as they focus on product research and development,

market research and basic engineering (Snow and Hrebiniak 1980). Prospectors focus on large,

285

constantly developing domains where they can utilize their capabilities to follow and create

change in a dynamic environment (Miles et al. 1978).

Prospectors define their entrepreneurial problem as locating new markets and exploit new

product opportunities. They tend to solve the problem by occupying a broad and continuously

developing domain, constantly monitoring the market for new opportunities, creating change

in the industry where they operate and growing through product and market developments

that can occur in spurts. (Miles et al. 1978) These represent a vivid organization that constantly

defines itself through the opportunities it manages to seize while keeping eyes open for new

opportunities to be seized.

Prospectors define their engineering problem as how to avoid long-term attachment to a single

technology that would constrain it. They solve this problem by focusing on flexible,

prototypical technologies, utilizing multiple technologies simultaneously and having low levels

of routinization. (Miles et al. 1978) This enables prospectors to be minimally attached to a

single technology and hence able to move quickly to adapt new technologies that present

opportunities to be on the cutting edge.

As prospectors focus on seeking new opportunities, they define their administrative problem as

how to facilitate and coordinate the diverse operations that they run simultaneously. They tend

to solve this problem by having marketing as well as research and development personnel as

the leading members of the dominant coalition that is large, diverse and transitory. They tend

to have comprehensive planning that is problem oriented, which is complemented by a product

structure with low degree of formalization and decentralized control. In order to accommodate

this, they have complex coordination mechanisms and performance measurement against

closest competitors. (Miles et al. 1978) All these choices steer the organization towards

proactive stance to its environment and new opportunity generation.

All these problems together inflict some benefits to prospectors but also incur some costs. The

organization is quite well protected against changing environments due to its capabilities while

it runs on a risk of low profitability as it is not able to focus on single technology that it could

exploit for sustained amounts of time. Technologically, the organization can respond very

286

quickly when the domain changes, yet cannot gain maximum efficiency. The administrative

system is also configured to accommodate rapid change and therefore runs on a constant risk

of misutilization of resources due to these diverse activities. (Miles et al. 1978) While these

characteristics give substantial benefits, they also incur large costs that need to be dealt with.

Analyzers are a combination of defender and prospector strategy types (Miles et al. 1978; Miles

and Snow 2003, p. 29; Snow and Hrebiniak 1980). They are organizations that attempt to

minimize risk while simultaneously maximizing potential profits by balancing efficiency and

power in the current market and at the same time seeking for new lucrative opportunities

(Miles et al. 1978; Miles and Snow 2003, p. 29). Top managers of these organizations emphasize

stability and efficiency in the stable areas of operations, whereas at the same time in the more

turbulent areas of operations they closely watch the developments of the market and adopt

new approaches as soon as they have been established effective (Miles and Snow 2003, p. 29;

Snow and Hrebiniak 1980). Analyzers thus focus simultaneously on a hybrid domain which is

partly stable and partly dynamic where they can utilize their dual technological core (Miles et

al. 1978).

Analyzers define their entrepreneurial problem as how to simultaneously locate and exploit

new product and market opportunities while keeping a firm grip of the existing products and

customers in the stable domain. They tend to solve this problem by occupying a hybrid domain

that is both stable and a changing one; they both occupy a stable part of domain and scan the

market to generate growth by market penetration and product development. (Miles et al. 1978)

The engineering problem of analyzers rises from the conflicting demands for technological

flexibility and stability. Analyzers tend to solve this problem by utilizing a dual technological

core, where part of it is stable and part of it is flexible. (Miles et al. 1978) This enables an

analyzer to take a full advantage of its dual focus by both exploiting current technologies and

exploring new ones to maximize the benefits it can receive.

Administrative problem of analyzers is also similar as it can be crystallized around how to

divide the organizations structure and processes to hold both stable and dynamic operations.

Analyzers can have a number of solutions for this problem such as that marketing and

287

engineering form the most influential members of the dominant coalition. In addition,

intensive planning occurs between marketing and production regarding the stable portion of

the domain, whereas the new markets are planned among marketing, engineering and product

managers. These organizations can also have loose matrix structures that have moderate

centralized control and complex coordination mechanisms to hold together the different areas

of operations. To complement these, performance appraisal can be based both on efficiency

and effectiveness. (Miles et al. 1978) These all exemplify how analyzers strive to keep their dual

operations running simultaneously and as efficiently as possible.

These characteristics give analyzers benefits but incur also costs. While the analyzers are able

to incur low levels of investment in research and development and have an ability to imitate

successful products that minimizes their risk; they require a well balanced domain to ensure

flexibility and stability simultaneously. The dual core that analyzers leverage to serve stable and

changing operations can never assume full efficiency or effectiveness as it constantly drifts

between these two. The administrative system of analyzers is ideally suited for the dual

operations required by the technical and entrepreneurial problems bit it runs on a constant risk

of losing the balance that may be difficult to restore. (Miles et al. 1978) Therefore, while

analyzer can combine the two extreme strategy types, they run on a constant risk of falling out

of balance and losing efficiency and effectiveness. Still, this strategy type exemplified an

organization that both lives in the present by exploiting current opportunities but also explores

new opportunities can be used in the future.

Reactors are organizations that exhibit inconsistent and unstable strategies (Miles et al. 1978;

Miles and Snow 2003, p. 29; Snow and Hrebiniak 1980). Reactors lack the consistent response

mechanism and proactive stance towards their environment that the other three hold. This

result from their adaptive cycles, which is inconsistent and inappropriate. Top managers of

these organizations are generally unable or unwilling to develop the competencies required to

assume a stable strategic form. Due to their reactive relationship to environment and inability

to respond accordingly, they tend to perform worse than all the other three strategy types that

exhibit a consistent strategy (Miles et al. 1978). Due to these reasons organizations cannot

continue to behave like reactor indefinitely and will either move to a consistent strategy or

seize to exist.

288

Figure 3: Strategic emphasis of each of the strategy types, adapted from Miles and Snow (1978)

The strategy types can be portrayed in a continuum from a pure focus to current market and

efficiency, to a constant search for new markets to expand and move into. Figure 3 illustrates

the continuum of the type of emphasis each strategy type gives. Note, that reactors are not

presented in the figure, as they do not exhibit consistently any strategy type and therefore

cannot be illustrated in such a continuum, as they cannot be placed constantly in any part of

the continuum.

Thus far the Miles and Snow typology (2003) has been studied in multiple industries, such as

hospitals, retailing, tobacco production, banking and electronics production (Meyer 1982;

Chaganti and Sambharay 1987; McDaniel and Kolari 1987; Smith et al. 1989; Moore 2005). In

addition, the typology has also been studied with research spanning multiple industries

(Hambrick 1981; 1983; Snow and Hrebiniak 1980). At the time of writing this study (February

2010), the seminal article by Miles, Snow, Meyer and Coleman (1978) was cited over 4500 times

in the Google Scholar system. This implies that the Miles and Snow strategy typology is highly

cited and studied theory of organizational adaptation. Exhaustive review of the studies

involving the typology would be an independent research in itself.

In studies of the strategy typology, defenders and prospectors are the most frequently included

strategy types in studies, and analyzers follow close behind in the frequency of appearance

(Zahra and Pearce 1990). Reactors, on the other hand, are not that often included in studies of

the strategy typology despite they are able to succeed in certain environments (Snow and

Hrebiniak 1980). This research includes the reactors, although they are predicted not to

succeed in conditions of change.

289

Although being widely accepted, the typology has received critique. First, self-typing, as the

method of group extraction, is widely used to identify different strategy types (Snow and

Hambrick 1980; Snow and Hrebiniak 1980; Zahra and Pearce 1990). This means that the

managers of organizations are used to characterize the strategy of their organization (ibid.).

This method has received wide critiques as the managers exhibit a tendency to intentionally

avoid identifying their organization as a reactor, and view their organization as unique and

therefore not directly fitting to the typology (Conant et al. 1990; Snow and Hambrick 1980;

Zahra and Pearce 1990). Despite this, this approach has also its proponents (James and Hatten

1995; Shortell and Zajac 1990). This research uses classification by investigator as the method of

group extraction and therefore avoids being drawn into this debate.

Second, research on the typology suffers from overemphasis of between-group differences

(Zahra and Pearce 1990). This implies that within group differences have been largely ignored

(ibid.). Therefore, the unique differences within strategy types could give valuable insight into

why performance differences exist within the strategy types. This research deals with the

matter by analyzing within and between group differences to gain deeper understanding of the

performance of each of the strategy types in decline conditions.

By adopting the strategy typology, strategy types of organization can be extracted from their

actions. Therefore, an organization can be assigned a strategy type based on how it adapts to

environmental changes. Second, the typology introduces broad strategy types under which the

research data can be distributed. This allows the generation of groups of data that have similar

features.

To conclude, each of the strategy types, except reactors, exhibit strengths and weaknesses that

can lead them to succeed in an environment. The way how each of the strategy types succeeds

appears to be largely related to the environment in which the organization operates, as each

type portrays characteristics suitable to different environments.

290

2.4 Strategy and environment coalignment

Strategic choices do not occur in a vacuum. Rather, organizations adapt to their environments

to achieve coalignment. This alignment of external environment and organizations structures

and processes is of great importance

Correct alignment of the strategy of organization and the environment has a positive impact on

the performance of an organization (Prescott 1986; Venkatraman and Camillus 1984;

Venkatraman and Prescott 1990). Therefore, the correct alignment of strategy to the declining

environment is of essence as competition intensifies and the carrying capacity of the niche

starts to reduce. Despite the fact that organizational ecology and strategic choice perspectives

have been identified as competing perspectives they can be aligned so that they are not

mutually excluding and can be operationalized simultaneously (Hrebiniak and Joyce 1985;

Zammuto 1988)

This research takes a reductionist view on the strategy and environment coalignment. “The

reductionist perspective of a coalignment is based on a central assumption that the

coalignment between two constructs (such as environment and strategy) can be understood in

terms of pairwise coalignment among the individual dimensions that represent the two

constructs” (Venkatraman and Prescott 1990). The coalignment of environment and strategy is

therefore viewed as matching the strategic choices to environmental categories that result in

success.

As the research produces snapshots of organization strategies in the decline conditions,

dynamic interaction of the strategy and environment is left outside of the scope of this

research. This is certainly a limitation of this research but it is congruent with the method of

matching strategic choices to decline conditions.

The research framework presented in the next subsection is built on the ideal profiles of

strategy types that match the environmental condition where success as an outcome is used as

a constant variable. Theoretical postulations are built upon the descriptions of environment

where each of the Miles and Snow strategy type is perceived to be the most efficient (Miles et

291

al. 1978). This notion is supported by Zammuto (1988), who argues that some strategies are

more successful in different decline conditions than others.

2.5 Theoretical synthesis and research framework

In this chapter, two different theoretical discourses have been introduced. The aim of this

discussion has been to introduce declining industries as a context of operations and strategy as

the way how organizations react to changes in their environment.

First, industry life cycles were presented and emphasis was given to an environmental decline

and the typology of decline conditions (Cameron and Zammuto 1983; Zammuto and Cameron

1985). This typology of decline conditions forms the context of this research.

Second, strategy as concept was introduced and strategic choice perspective as a strand of

strategy research was presented. From the strategic choice perspective, the classic strategy

typology of Miles and Snow was chosen to represent different strategy types that organizations

can employ (Miles and Snow 1978; 2003).

Strategy and environment coalignment was discussed to bridge the concept of strategy with

environment. The concept of strategy-environment coalignment was brought up as a

moderator of success of organizations (Prescott 1986; Venkatraman and Camillus 1984;

Venkatraman and Prescott 1990; Zammuto 1988). As mentioned earlier, for the purposes of this

research, a reductionist perspective is adapted to portray the strategy-environment interaction.

Furthermore, this is also congruent with the methodology chosen for this research.

Organizational ecology and strategic choice perspectives have been identified as competing

perspectives. This debate culminates on the question, “is organizational life determined by

intractable environmental constrains or is it actively created through strategic managerial

choices” (Astley and Van De Ven 1983). Despite this, these two perspectives can operate

simultaneously (Hrebiniak and Joyce 1985; Zammuto 1988). This rises from the level of analysis

that is used (Astley and Van De Ven 1983). Population ecology examines populations of

organization from the level of niche that defines the bounds of a population. Individual

organizations generally choose only to inhabit only a part of the niche, establishing a sub space

292

in the niche labeled organizational domain (Hannan and Freeman 1977; Zammuto 1988). The

concept of domain is accepted in both of the discourses (Hannan and Freeman 1977; Miles and

Snow 1978). It is the domain of activities that a single organization can alter to align itself to its

niche. Therefore, a single organization has the capability to alter its domain within the niche,

while simultaneously the population of the niche is subject to environmental selection (Astley

and Van De Ven 1983; Hannan and Freeman 1977; Zammuto 1988). Despite this, the

organizations are constrained by the perceptions of their management regarding possible

choices (Miles et al. 1978; Hambrick and Mason 1978). Correct strategy-environment

coalignment leads to selection and retention (Aldrich 1979, p. 27; Venkatraman and Camillus

1984; Venkatraman and Prescott 199). In this vein, some strategic choices are more successful in

different environmental conditions than others, while organizations still are able to choose

their response to environmental changes (Zammuto 1988). Although, the choices of

organizations are limited to those which the managers believe will allow effective operation,

meaning that choice is constrained by the backgrounds of the managers (Hambrick and Mason

1984; Miles et al. 1978).

Building on aforementioned arguments, this research utilizes the typology of decline

conditions identified by Zammuto and Cameron (1985) as the context, where organizations

make different strategic choices represented by the typology of strategies identified by Miles

and Snow (2003). As the goal of this research is to recognize successful strategies, the objective

of the empirical part is to find different strategies that result in success in different conditions

of environmental decline and result in strategy-environment coalignment (Prescott 1986;

Venkatraman and Camillus 1984; Venkatraman and Prescott 1990).

The Miles and Snow strategy types have quite explicit definitions of which environments they

fit the best and drawing from there, how they would fit to the different decline conditions

(Miles et al. 1978; Zammuto 1988). Therefore, it is possible to postulate that each of the strategy

types would succeed in a certain environment. Each strategy type is assigned to only one

decline condition, as the emphasis of each of the types best fits to only one decline condition

based on theory. The figure 4 below portrays the theoretical framework built upon the ideal

characteristics of different strategy types and conditions of environmental decline. The following

293

paragraphs identify how the Miles and Snow strategy types (Miles et al. 1978) would fit into the

decline conditions presented by Zammuto and Cameron (1985) and the underlying reasoning.

Figure 4: Strategy-environment coalignment framework

Defenders are organizations that succeed in relatively stable industries where turbulence is

minimal as they are unable to respond to quick major shifts in the market (Miles et al. 1978;

Snow and Hrebiniak 1980). This would rule out the possibility that a defender would succeed in

a niche facing discontinuous change. The main advantage of a defender is its efficiency in a

small domain in the market, which it is able to defend with this efficiency. If the niche shape

were to change, the defender could easily end up in a situation where its domain would rapidly

deteriorate due to the change. These notions conclude that defenders have the best strategy-

294

environment coalignment with the decline condition of erosion as they would be able leverage

their capability for efficiency in a steadily dimishing environment (Zammuto 1988).

Prospectors are organizations that succeed in dynamic environments as they are able to

respond quickly to the changes in the surrounding environment and also act as the initiator of

change in their environment (Miles et al. 1978; Snow and Hrebiniak 1980). The main advantage

of a prospector is its ability to find and exploit new product and market opportunities before

the competitors can do so (Miles et al. 1978). This would imply that prospectors succeed in

environments where the niche shape changes. On the other hand, prospectors are unable to

succeed in more stable environments as their orientation towards innovation undermines their

efficiency (ibid.). This, on the other hand, would rule out the possibility to succeed in

continuously changing environments. These notions conclude that prospector have the best

strategy-environment coalignment with the decline condition of collapse as they would be able to

leverage their capability to create change and respond to it in a rapidly changing niche (Zammuto

1988). The rapid and discontinuous change would also shield the prospectors from analyzers

that would be able to overpower prospectors with their ability to leverage both the ability to

move to new developing markets and be efficient.

Analyzers are a combination of defender and prospector strategies, they are organizations that

try to minimize risk but at the same time maximize opportunity for profit (Miles et al. 1978;

Snow and Hrebiniak 1980). The main advantage of an analyzer is its ability to keep firm grip of

the current market but at the same time follow developments in the market to be able to be an

early entrant into new markets (Miles et al. 1978).This would imply that analyzers would

succeed in niches where the niche shape changes, as they would be able to follow the first

movers into the new developing niche. As analyzers’ adaptive approach is a balanced one, they

do not fit into rapidly changing markets, but rather into markets that evolve at a moderate pace

in which they can follow the prospectors and enter new markets without incurring the

prospector’s extensive research expenses (ibid.). This would imply that analyzers fit best to

conditions of continuous changes as they would be able to capitalize on their ability for

efficient but also the ability to be an early entrant to new niches. These notions conclude that

analyzers have the best fit with the decline condition of dissolution as they would be able to

295

leverage their ability to exploit current niche opportunities, while simultaneously following the

development of new niches in the industry.

Reactors are organizations that do not exhibit proactive approach to the changes in the

environment (Miles et al. 1978; Miles and Hrebiniak 1980). Their reactions to the changes in the

environment are inconsistent and unstable. As Miles et al. (1978) state, when faced with a

change a reactor will either assume a consistent strategy or perish, they should not be present

in declining industries.

These descriptions of how the strategy types would align with environment according to theory

formulate the theoretical framework of this research. Following from this, the empirical part of

this research aims at uncovering the behavior of the strategy types in declining industries and

relates these findings to the research framework by enhancing it. Furthermore, these findings

are also compares to previous studies of the Miles and Snow typology to reveal similarities and

differences.

3 METHODOLOGY

This section presents the method use in conducting the empirical research of this thesis. This

section presents the case survey method and its position in the category case study methods. It

also uncovers the method used to generate data and the methods used to analyze it. In

addition, the reliability and validity of the data are discussed.

3.1 Case study research

A case study as a research strategy can be defined as an empirical inquiry that investigates a

phenomenon within its context, where the boundaries between the phenomena and the context

are not clearly evident (Yin 2003, p. 13). Case studies have been proven to be an excellent

method of generating and testing theories in the field of strategy (Gibbert et al.2008). When

considering this research in the light of case as a research strategy, the goal is to investigate a

phenomenon that cannot be separated from its context as it is the context that largely defines

296

the research. Therefore, this research complies with the guiding idea behind case study as a

research strategy.

Case studies can have multiple different aims. Case studies can aim at illuminating a decision, a

set of decisions, individuals, organizations, processes, programs, neighborhoods, institutions or

events and the underlying reasoning behind them (Yin 2003, p. 12). The aim of this research

complies with aims of a case study method as the aim is to illustrate choices by organizations

which lead to success in the context of declining industries.

Case study research in the field of strategy has shifter to rely substantially on positivistic research

tradition as the criteria used to assess the rigor of field research are drawn from this tradition

(Eisenhardt 1989; Gibbert et al. 2007; Yin 2003). These criteria are construct validity, internal

validity, external validity and reliability and they will be discussed in the detail in the reliability

and validity subsection. It appears that despite being declared dead by many contemporary

philosophers of science, the positivistic tradition still holds strong as it is deeply rooted in our

western thinking (Kincheloe and Tobin 2009). This research also has a positivistic undertow, as

the research first builds a theoretical framework which is then empirically tested and corrected;

this resembles the use of scientific method (Behling 1980; Kincheloe and Tobin 2009). Secondly

the research uses a reductionistic perspective of the studied phenomena that also rises from the

positivistic tradition (ibid.).

3.1.1 Case survey method

Case survey method was first used and introduced as a research approach by Yin and Yates in

1974 in their research on decentralization and urban services. From there onwards the method

has been refined by authors such as Larsson (1993) and Lucas (1974). As such, the research

method has never broken into mainstream and has remained only as a small strand in the case

study methodology.

The essence of a case survey methodology is the use of existing case studies as data from which

tendencies are aggregated through a survey (Lucas 1974; Larsson 1993; Yin and Heald 1975). The

method aims at balancing qualitative and quantitative research approaches, as the data consists

297

of case studies exploring phenomena in their real-life context which represents the qualitative

approach, of which tendencies are drawn with the use of a survey which represents the

quantitative approach. This research approach therefore aims at combining these two

approaches to get the best out of both research strategies.

Using case studies as the data stems well with the research setting as case studies seek to study

phenomena in their context, in this research declining industries, and the research approach

emphasize this importance. Case studies hence give a good access to the reasoning behind

success in declining industries.

Producing generalizations out of a case study is difficult due to the nature of case study as a

research strategy (Stake 1995, p. 7). The fundamental idea behind a case study is to produce in-

depth insight of an entity or a limited number of entities that at best yield tendencies but not

generalizable patterns (Stake 1995, p. 7). Instead, case survey can produce patterns between

case studies that give rough indications of the studied phenomenon.

As the amount of cases diminished substantially from the initial sample to the final sample, the

case survey method was applied in a way that it would accommodate both qualitative and

quantitative analysis.

3.2 Constructing the survey

The survey form was generated on the basis of the decline conditions of Zammuto and

Cameron (1985) and the strategy types of Miles and Snow (2003). The goal of the survey was to

enable the plotting of different strategy types into the different decline conditions.

As this research takes a reductionist perspective to the strategy-environment coalignment, the

survey was broken down into three parts. The first part regarded background information and

had the case number, so that the answers could be tracked back to specific cases and also

identification of the organizations product-market i.e. product or service market and served

market i.e. B2B market or B2C market.

298

The second part of the survey regarded environmental decline types identified by Zammuto

and Cameron (1985). As the decline types in essence form a matrix, the identification of

decline type was broken down into two questions, one regarding the type of decline i.e. change

in niche size or change in niche shape and the second question regarding the continuity of

decline from continuous to discontinuous. These formulate the two dimensions of the matrix

where type of decline forms the Y axis and continuity the X axis.

Position in the matrix was measured with a 7 point semantic differential scale where end points

were associated with bipolar labels (Malhotra and Birks 2007, p. 350; Collis and Hussey 2003, p

184). These labels were change in niche shape and change in niche size for Y axis, and

continuous and discontinuous for X axis. This allows obtaining numerical values from

qualitative data (Collis and Hussey 2003, p. 184). Therefore, the decline condition of the case

could be defined.

The third part of the survey regarded the strategy type of organization in the Miles and Snow

(2003) strategy typology. This was measured with a nominal scale as the use of nominal scale

enables the responses to be classified to categories which are not comparable with each other

but rather equal options. This part of the survey was adapted from Snow and Hrebiniak (1980)

as they had already conducted survey research with a nominal scale of the strategy types.

The number of questions used in the survey was relatively modest. The reason for this is that

the person who created and answered the survey was the same i.e. the author. Hence, no

additional insight should be possible to be generated with additional questions as consistency

should exist between answers.

A complete survey form can be found from the appendix 2. This form has all the three parts of

the survey and is an exact duplicate of the form used to conduct the research.

3.3 Collecting the research data

The data collection and analysis follows loosely the case survey research process advocated by

Larsson (1993) and Bullock (1986). As the research has been done by only one person, many of

299

the steps in the process were left out. The following paragraphs describe the applicable steps in

the case survey research process that were conducted.

The data collection was initiated after the initial research questions were developed and the

basis case selection criteria were generated. The initial criteria for case selection rose from the

research questions. These criteria were that the context of the case had to be declining industry

and that the case had to report success. The cases were collected from all the available major

journal article portals available to HSE students and personnel and all the portals were

searched until data saturation was reached. The available portals were at that time Google

Scholar, EBSCO, Emerald, Jstor, SpringerLink, Proquest, Sage and Science Direct. The

collection of data was conducted from the mid November of 2009 until the end of January 2010.

Table 1 describes the keywords used to search the cases, the portals used to search for the cases

and the distribution of cases among the sources. The amount of cases is biased towards Google

Scholar, EBSCO and Jstor as they were the first portals to be searched. The initial sample

amounted to 81 cases that were selected by reading the abstracts of the articles and comparing

them to selection criteria mentioned earlier.

Table 1: Distribution of cases among the keywords and search databases

At this point, the survey form was generated to convert the cases into variables; a closer

description of the survey can be found from the previous subsection. After the generation of

the survey, the cases were coded by using the survey and a closer reselection criterion on the

cases were made to drop cases that did not have adequate information. The reselection criteria

300

included that the case had to be a case study as many journal articles initially chosen were not

applicable as case studies. Second, the case study had to have a clear description of the

declining industry so that the type of decline could be assessed. Third, the case study had to

include adequate information to indicate which strategy type performs in the decline

condition, and also the case had to have a description of the performance outcome, as many

cases only dealt with strategy but not performance. In addition, at this point, the non-academic

case studies were dropped to increase the reliability of the case studies.

The reselection and coding shrunk the amount of cases to a third of the original case mass leaving

27 cases with adequate description to generate a completed survey form (see appendix 1 for closer

identification of cases). The tight selection criteria were used to ensure quality of the cases

accepted for the research. This is exemplified by the fact that the only journals from which

multiple studies were accepted were Strategic Management Journal (6 cases) and Long Range

Planning (3 cases).

3.4 Methods of analysis

The research data was analyzed on three levels. First, the groups of case studies that employed

a certain strategy and a decline condition were analyzed to find similarities and differences

between case studies. This was used to draw syntheses of the behavior of strategy types in

different decline conditions. Second, each strategy type was analyzed by comparing the groups

of cases in each of the decline conditions to draw conclusion on the level of a strategy type.

Therefore, synthesis of the behavior of each of the strategy types could be established on the

level of strategy type. Third, the strategy types were compared together to formulate

conclusions of the behavior of the strategy typology in the decline conditions. This resulted in

the higher order outcomes of this research.

The analysis of the data was done in three stages combining multiple methods of analysis. The

data was analyzed by conducting pattern matching and cross case synthesis (Yin 2003, p. 116,

133). These two analysis strategies were used because of two main reasons. First, pattern

matching functions well in this research because the empirical research aims at testing the

research framework presented in the theoretical synthesis, resulting in pattern matching.

301

Second, as the data consists of multiple case studies cross case synthesis can be used to analyze

similarities and differences between cases. The next paragraphs describe in more detail the

process of analyzing the case studies.

After the initial data set was gathered, the first phase of analysis was conducted. This consisted

of reading through all the case studies. By doing so, the cases that did not fit the reselection

criteria were excluded from the survey. After reselecting the case studies that were included in

the final data set, the case studies were coded by using the survey. By doing so, each cases was

assigned a strategy type the organization employed and a decline condition in which the

organization operated.

After the case studies were coded using the survey, the second phase of analysis was conducted.

The case studies were divided into three groups according to the strategy type employed by the

organization. Each of the three groups was analyzed separately. In each of the groups, the

analysis started by plotting the case studies into the decline conditions. By doing so, the

pattern of distribution of the case studies could be compared to the theoretical framework.

After doing so, case studies in each of the decline condition were analyzed to form a cross case

synthesis of the characteristics of each strategy type in each of the decline conditions. These

findings of within strategy type differences were then analyzed to achieve an overall

understanding of the strategy type in declining conditions and the reasons for success.

After each of the strategy types were individually analyzed, the last phase of analysis was

conducted to compare strategy types with each other. This enables comparing each of the

strategy types with each other and define how each of the types function in the decline

conditions. Therefore, higher level conclusions can be drawn of the success strategies in

different decline conditions.

The goal of utilizing such analysis strategy is for two reasons. First, the behavior of the strategy

typology as a whole in different decline conditions case be established. This leads into

answering the research questions. Second, in order to explain deviation within a strategy type

and increase validity, within type analysis is conducted. This enables a more fine-grained

analysis of the strategy types.

302

3.5 Reliability and validity

Yin (2003, p. 34) describes four tests to evaluate the quality of any empirical social research.

These four design tests are construct validity, internal validity, external validity and reliability.

This subsection aims at applying all of these to this study to give a clear picture of the reliability

and validity of this research.

The purpose of construct validity is to “establish correct operational measures for the concepts

being studied” (Yin 2003, p. 34). Two distinct concepts were chosen to act as the foci of this

research and the phenomenon was researched through these lenses. To increase construct

validity the survey was built around the decline conditions identified by Zammuto and

Cameron (1985) and the way of identifying strategy types were adapted from a survey

conducted by Snow and Hrebiniak (1980) of which the other is an original author of the

typology. The within group analyses were conducted on the basis of the strategy employed and

which decline condition the organization inhabited.

This research utilizes existing case studies as the research data of which all are published in

academic journals. A chain of evidence was established as each of the cases can be followed

from the combination of words used to find the case and the source, through the analysis and

into the findings. This process has been meticulously documented. The only aspect that

undermines the construct validity is the key informants were not used to review the results.

This is a procedure also suggested by Larsson (1993) and hence is left as a shortcoming when

regarding construct validity.

Internal validity established a causal relationship where certain conditions are shown to lead to

another (Gibbert et al 2007; Yin 2003, p. 34). This research uses cross case synthesis and pattern

matching to enhance internal validity by aggregating findings across cases. This is possible

because the data consists of multiple different case studies. The cases are analyzed both

quantitatively and qualitatively.

External validity refers to the generalizability of the findings beyond the immediate case study

(Gibbert et al. 2007; Yin 2003, p. 34). As Eisenhardt (1989) argues, cross-case analysis of four to

303

ten cases can act as a basis for analytical generalization and as this research has in total 27 cases

the amount of data is adequate to produce analytical generalization.

Reliability of the research refers to possibility of repeating the case study with the same results

(Yin 2003, p. 34). The process of conducting the case survey followed loosely the procedure of

conducting case survey research identified by Larsson (1993) and the application of these steps

have been explained in the methodology section. Despite this fact, the reliability of this

research has been compromised in a sense that the research was conducted by a single person.

This is against the suggestions of Larsson (1993) as he notes that case survey research should be

done by a research team to eliminate possible bias and that people coding the cases and

building the theory should be different to minimize bias. This unfortunately was not possible

when conducting this research and hence remains a big shortcoming of this research.

4 FINDINGS

This section gives a general description of the data generated for this research. After this, a

closer description of each strategy type and their fit to different decline conditions are made.

The analysis of each strategy type is divided into part describing their distribution in the

decline environments and a part describing the similarities and differences among the strategy

type.

4.1 General description of the data

The survey yielded a total of 27 cases. In two occasions two cases were used to construct one

case entity. This leaves a total of individual 25 cases for this research. The case studies were

equally distributed among journals. The only academic journal from which multiple case

studies were accepted were Strategic Management Journal, with 6 cases, and Long Range

Planning, with 3 cases. A closer identification of the cases included in the sample of this

research can be found from the appendix 1.

The distribution of cases according to the product-market is heavily skewed towards products

as 19 of the cases regarded a product market when compared to only 6 cases from the service

304

market. This may be a consequence of a number of reasons, the major one being the fact that

products are tangible and, thus the development or death of a product industry is more

concrete as service companies are less tied to production equipment. The distinction between

product and service markets were left outside the scope of this research due to the small size of

the sample. In addition, evidence of the difference between product and service are

inconclusive as both views have proponents (see for example Fein 1998, and Goldfarb et al.

2007).

The sample shows a bias towards B2B markets, as 15 of the cases regard B2B markets and 10

cases regard the B2C market. Both of these dimensions however were left outside the scope of

analysis due to the small sample size.

Figure 5 depicts the distribution of cases among the different decline conditions and different

strategy types. In the figure, different strategy types have been distributed among the decline

conditions based on the survey results.

Figure 5: Distribution of strategy types in different decline environments

305

Decline in the environment was broken down into two components, the type of change and the

continuity of change. The type of change was skewed towards change in the niche size, as 16

out of the 25 cases had a decline in the size of the niche. This can be, at least to some extent,

explained by the fact that declining industries are perceived as those in which the niche size

decreases, rather than those where the niche shape changes. This might be because changes in

niche shape are perceived more often as evolution or development of the industry. As a result,

the way how the data was collected may not have been supportive towards cases where the

change occurs in the niche shape rather than the niche size.

The continuity of change was quite evenly distributed in the sample as continuous decline was

experience in 11 cases and discontinuous decline in 14 cases. This indicates that the sample

represented quite evenly the different types of continuity in a niche.

All the four decline conditions were represented in the sample. Among these, erosion and

collapse were well represented as they represented 16 out of 25 cases. Dissolution and

contraction, on the other hand, were not as solidly represented as cases in which change

occurred due to changes in niche shape were only 9 cases.

Three of the four strategy types were represented in this sample. Out of these three, defenders

and analyzer were foremost present, as the sample included 12 analyzers and 10 defenders. The

sample indicated only three prospectors and not a single reactor. In all the case studies, the

successful case company held a clear direction and therefore no reactors were found from the

data set.

To conclude the general description of the data, it is evident that the amount of data may not

be applicable for sole statistical analysis. Although, for the purposes of doing meta-analysis,

which involves quantitative as well as qualitative analysis, the sample size is adequate.

4.2 Strategy types in different decline conditions

The 25 cases included in this study are discussed here separately according to the strategy type

employed. This analysis is split into an analysis of the strategy types and the conditions where

they appear in the sample. This is followed by a within group analysis of differences in order to

306

formulate a better understanding of the groups and their success in different decline

conditions.

4.2.1 Defenders in declining industries

Defenders are organizations which gain competitive advantage by occupying a niche in the

market that they protect by being efficient to prevent competitors from entering the niche

(Miles et al. 1978). This is, of course, a portrait of an extreme case as the defenders in the

sample demonstrated a variety of ways of employing this strategy.

The sample had in total ten cases in which defender strategy was employed to succeed in

declining industry. Figure 6 depicts the distribution of these cases between the different

decline conditions.

Figure 6: Distribution of defender strategy type in the decline conditions

As can be observed from the figure 6, in nine out of the ten cases, the decline was due to the

changes in the niche size. This can be accounted as strong evidence that defenders are able to

307

succeed in conditions in which change is due to the change in the niche size. Change in niche

size results in an intensified competition and thus the exit of some of the organizations

(Ghemawat and Nalebuff 1985, 1990). As defenders are organizations that define their

entrepreneurial, engineering and administrative problems around securing a domain in the

market by defending it with efficiency, they fit well in conditions where total carrying capacity

of the niche decreases (Miles et al. 1978). In this sense, the distribution of cases appears to favor

the decline in size of the niche rather than change in niche shape. This is congruent with the

notion of Miles et al. (1978) that defenders exhibit tendency to succeed in stable environments.

The figure also depicts that seven of the cases exemplify continuous decline, while three

experience discontinuous decline. This emphasis of cases on the side of continuous decline is

also posited by Miles et al. (1978) as defenders have a tendency to succeed better in stable

environments and, furthermore, that the primary risk of defenders is that they are unable to

respond to major changes in the market environment. In this sense the distribution of cases

regarding the continuity of change is congruent with the notion of Miles et al. (1978) that

defenders exhibit tendency to succeed in stable environments.

As the total number of cases is low, within group comparisons give additional depth to the

analysis. For these purposes, the cases will be divided into three distinct groups based on the

decline condition in which the strategy is employed.

The seven cases that are positioned in the erosion quadrant share similar features. All of the

cases, except one case, depict a rather successful adaptation to the decline of the niche. The

only case, in which success is not as straightforward, is a case of Jesuits in the USA and the

declining church membership. This case describes how the Jesuits of U.S. provinces were able

to sustain their operations by removing overlapping operations and, thus, increasing efficiency

despite losing members. Among the rest of the cases, success had been attained through

economies of scale, high quality offering, cost cutting and evading diversification. These all

represent the choices defenders have. When looking at the cases, they portray a pattern in

which diversification is either not possible or it would lead to decreased performance.

Therefore, it appears that in these cases, focus and efficiency has been the most lucrative

option.

308

The three cases that fall outside the erosion quadrant exemplify cases where success has not

been as explicit but rather success in relation to environment from which others have or are

exiting. The two cases that portray contraction of the industry, which is discontinuous change

in niche size, are quite similar. First of the cases portrays how British steel casting industry

changed through a series of declines in orders and government subsidies for exit. Organizations

that stayed in the industry were the ones that were not diversified. They were able to survive

and succeed as population density decreased when diversified organizations exited. The second

case concentrates on the Israeli electronics industry where decline was caused by local

recession diminishing the local demand, while at the same time the global demand was also

experiencing decline. The successful Israeli organizations defended their domain and wait that

more space would be generated to the niche. In both of the cases the only option left for the

companies in the niche was to defend their own domain and hope that space would be

generated to the niche so that the companies could continue their operations more

successfully.

The one case in the collapse environment in which niche shape changes discontinuously the

market has traditionally been volatile. The case regards international fur industry

concentrating on the Greek niche of Kastoria and Siatista. In this case, the collapse of the niche

results from changing preferences from fur to other kind of materials, as well as shifting

carrying capacity of the niche to low cost countries in Asia. For the moment, the Greek niche

has been able to survive by increasing efficiency. Despite this, the future of the niche is left

open as efficiency might not lead to success in the future due to low cost competition from

Asian countries. This could lead the Greek niche to be driven out of the market.

In contrasting the three cases outside the erosion environment to the cases in the erosion

quadrant, a significant difference seems to emerge. It appears that defenders are more fit to the

environment of erosion than other decline conditions, as they appear to perform better in

conditions of erosion. This statement is supported by two details that arise from the data. First,

more than two thirds of the cases are in one quadrant.

Second, when the cases were compared to each other, cases in the erosion environment tended

to portray more of a success when compared to the cases in other decline conditions.

309

Organizations in the erosion environment were able to continue or increase their operations by

increasing efficiency and focus on their domain, whereas organization in other environments

did not have other choices available than to defend the domain that they occupy and wait for

more space to be generated into the niche.

Together these finding support the notion of Miles et al. (1978) that defenders succeed in

environments where change is moderate, enabling defenders to fully utilize their capability to

be efficient. Therefore, it is evident that defenders can be successful in declining industries, and in

particular have a tendency to succeed if the decline is gradual so that they have the opportunity to

leverage their efficiency to generate success when competition intensifies (Zammuto 1988).

4.2.2 Prospectors in declining industries

Prospectors are practically polar opposites of defenders as they gain their competitive

advantage by finding and exploiting new markets (Miles et al. 1978). This is of course an ideal

profile of a prospector but generally they can be defined as companies that are innovators in

product and market development.

The sample had three prospectors. For this reason, nothing conclusive of the strategy type

cannot be said, as their number falls under the minimum amount of cases suggested by

Eisenhardt (1989) for drawing analytic generalizations. Despite this, the three cases will be

examined to achieve an understanding of how the prospectors behave in declining

environments. Figure 7 depicts the distribution of the prospectors in the decline conditions.

310

Figure 7: Distribution of prospector strategy type in the decline conditions

The lack of cases supports the portrait of a prospector as they actively try to alter and create

markets and, thus, it would be unlikely that they would end up remaining in a declining

industry. Rather, prospectors would be the ones responsible for the decline.

On all of the cases the change occurs in the niche shape. This is congruent with the

environment where prospectors strive. If the change would occur in the niche size, the

prospectors would be cornered by more efficient organizations in the niche and driven out of

the market instead, when the change occurs in the niche shape, prospectors are able create or

follow this change. As prospectors define their entrepreneurial, engineering and administrative

problem solving around mobility of the organization and innovation; environment where niche

shape changes tend to be more suitable for them (Miles et al. 1978).

Al three of the cases also exhibit discontinuous change. This also follows from the with the

environmental requirements of prospectors, as Miles et al. (1978) posit that prospectors strive

in more dynamic conditions than all other strategy types in the typology. If the change would

311

be continuous, prospectors would not be able to succeed because they would continuously be

driven out of market by, for example, analyzers that are more effective. As prospectors define

their entrepreneurial, engineering and administrative around mobility of the organization and

innovation, discontinuous environment change tends to be more suitable for them (Miles et al.

1978).

Based on the distribution of the cases regarding the type of change in the niche and continuity,

the environment of collapse appears to fit this strategy type best. All three of the cases that

exhibit a prospector also portray them inhabiting this decline condition. The discontinuous

change in niche shape enables the prospector to exploit its capabilities to change form and use

its capabilities to locate and exploit new market opportunities.

A closer examination of the three cases was done to gain more in-depth insight of the cases. A

first similarity between the cases concerns as all the cases focusing on technology intensive

industries. Two of the cases portray a case study of a high-tech organization and the third

portrays a case of medical-surgical hospitals and their transformation through changes in

technology.

Two of the three cases emphasized corporate entrepreneurship as a key to reinventing the

organization and the domain. In the case of medical-surgical hospitals the niche constitutes

four counties in the San Francisco bay area and entrepreneurial orientation was described as a

key in reinventing the organization and its domain. The other case described coping with

decline in dynamic environment with the use of corporate entrepreneurship and recombinative

organizational form. Both of these exhibit a case in which the adaptation to the evolution of the

environment is absorbed by proactive entrepreneurial orientation.

The third case describes the migration of prior radio producers to TV producers in the

television production in the U.S. consumer electronics industry. The case portrays how

organizations prior in the radio production took over the television production as they were

innovation oriented by nature. When comparing them to other entrants to the television

production niche, these firms exhibited greater levels of innovation as they already had

functioning R&D departments that were geared towards market generation.

312

Even though all these case studies of prospectors fall under the decline condition of collapse,

all the case organizations seemed to do financially well as the market did not disappear but was

rather replaced by a new one. Two of the cases portray exceptional financial performance as

one of them focuses on top performing Fortune 500 high-tech organization and the other case

regarded prior radio producers that had greater market shares and longer survival rates than

other organizations in the industry.

The three prospector cases exemplify two types of prospectors. The prior radio producers

emphasize product innovation while the medical-surgical hospitals as well as the case

regarding recombinative origination form gain competitive advantage through reconfiguring

the form of the organization.

Although generalizations from such a small number of cases cannot be made, the prospector

cases have similarities that are congruent with the archetype of a prospector described by Miles

et al. (1978). All the cases reside in the collapse environment which is congruent with the notions

that prospector inhabit dynamic environments in which they can utilize their ability to locate and

exploit new market opportunities (Miles and Snow 2003; Zammuto 1988). One reason for the

lack of prospectors in the sample is that in the extreme case prospectors should not even be

found from declining industries as they should be the ones generating the change through

inventing new niches.

4.2.3 Analyzers in declining industries

Analyzers are organizations that are combinations of defenders and prospectors. These

organizations aim at maximizing opportunities for profit while at the same time minimizing

risks (Miles et al. 1978). The sample had in total 12 analyzers. The figure 8 depicts the

distribution of these cases between the different decline conditions.

313

Figure 8: Distribution of analyzer strategy type in the decline conditions

When looking the distribution of analyzers regarding the type of change in the niche, the cases

are distributed quite evenly in the matrix. For this reason, it is viable to postulate that since

analyzers are a combination of defenders and prospectors, they can strive in both types of

environments as they are able to draw competitive advantage from efficiency but also from the

ability to follow changes in the market (Miles et al. 1978).

When regarding the distribution of analyzers based on the continuity of change, the number of

cases is skewed towards discontinuous change as four of the cases reside in the side of

continuous change and eight on the side of discontinuous change. It is here viable to postulate

that as analyzers are a combination of defenders and prospectors they can strive in changing

environments better if they can leverage their capabilities to be both efficient and innovative.

The distribution of analyzers in all of the quadrants implies that they are able to draw

competitive advantage from multiple sources and, thus strive in different kinds of

314

environments. Although it appears that they exhibit a tendency to operate in conditions of

discontinuous change.

A closer examination of the cases was made where the cases are distributed into groups based

on the quadrant where the case resides. The within group analysis can shed light into the

reason behind the success of analyzers in all of the decline conditions and how the competitive

advantage is created in each of the decline conditions.

The two cases in the erosion quadrant portray success in steadily declining industries. The first

case regards the decline of MBA programs in North America concentrating on Canada and

specifically the domain of University of Toronto. In this case success is attained by leveraging

existing knowledge in similar niches such as in executive MBA programs and non-degree

executive programs to stabilize the financial situation of the organization. In doing so, the

university is at the same time able to maintain their core operations of MBA programs by

supplementing them with other programs to enhance cash flow. The second case regards the

US tobacco industry from 1950-1979 concentrating on the most successful organizations. The

industry itself is in steady decline but the successful organizations defend their core operations

by forming lobbying groups to protect their core business, expanding their product lines to

cover multiple niches and to penetrate foreign niches as well. These operations are

supplemented by diversifications to other segments of consumer goods where they can leverage

their existing capabilities.

These two cases share similar features in two ways. First, the organizations diversify to related

niches in order to stabilize their operations and spread risk. Second, in both of the cases the

organizations leverage their core competencies to related niches where they seek stability. It

appears that when facing erosion, analyzers have a tendency to diversify to related niches to

stabilize their operations and to generate an exit plan in case one should be needed in the

future.

The five cases in the contraction quadrant portray success in niches where the niche size

decreases discontinuously. First of all, three of the cases are positioned in or in close proximity

of the defense industry. The two other cases represent the Japanese aluminum industry and

315

U.S. collectibles industry, concentrating on baseball card niche. All the cases share common

traits such as they all portray how the organizations spread risk by diversifying. All the cases in

or in proximity to defense industry share a pattern of leveraging core competencies to diversify

either within the industry or to industries in close proximity. In the case of baseball cards, risk

is spread by leveraging core competencies by issuing new brands to cover more niches and

spreading the release of products throughout the year. The last case concerns Japanese

aluminum smelting niche in the aluminum industry where risk is shared by group formation

and loss sharing among the organization occupying multiple niches. All the organizations have

activities related to the aluminum smelting niche which indicates that despite the aluminum

smelting niche is declining the corporations wish to control large parts of the value chain.

The five cases are similar in two ways. First, all the organizations diversify to decrease risk,

either within the industry or beyond the industry. Secondly, in four of the five cases the

organizations diversify by leveraging their core competencies. It appears that when facing

contraction, analyzers have a tendency to diversify to related fields to stabilize their operations.

Two cases among the analyzers inhabit the dissolution quadrant in which the niche shape

changes continuously. The first case is a historical analysis of the British cotton production

where foreign competition resulted in a steady change of the niche, as foreign competitors

changed the type of activities the niche sustained. In this niche, the organizations that

performed best were the ones that were vertically integrated, had diverse product range and

multiple plants. These organizations were successful due to their ability to modernize

equipment, increase product differentiation and do mergers and acquisitions within the

industry. The second case regard the niche of single sex boarding schools in which demand of

customers is transforming from boarding schools to day schools and hence the niche shape

changes; this is only case where the type of change was explicitly stated according the typology

of Zammuto and Cameron (1985). In this case the focal organization diversified to day

schooling and hence followed the change in niche shape while keeping its core business which

attracted foreign students.

Both of these cases exemplify organizations that are able to follow changes in the environment

by diversifying according to the needs of the new developing niche. Also, in both of the cases,

316

the organizations diversify by leveraging their core competencies. It appears that when facing

dissolution, analyzers have a tendency to diversify to the newly emerging niche to continue

their operations successfully.

Three cases among the analyzers inhabit the collapse quadrant in which the niche shape

changes discontinuously. The first case regards U.S. hardwood lumber and component

manufacturers’ niche in the U.S. lumber industry and, in particular, how Chinese imports have

displaced U.S. furniture manufacturers. Despite the collapse of their niche in the furniture

industry, these companies have been able to shift their focus to manufacturing flooring,

kitchen cabinets and exporting the goods they used to sell to domestic manufacturers

generating new niches. This case study illustrates ways in which the companies were able to

steer their operations to diversify to related niches when their primary niche collapsed.

The second case regards U.S. railroads and how, in particular, formerly regulated industry was

deregulated and as a result new niches in the industry were generated. In this case those

organizations that managed to change their strategy so that it was aligned with the new

environment outperformed others. In this case the emphasis on innovation was the most

profitable strategy.

The third case handles the change of New Zealand footwear manufacturing from a protected

niche into a competitive one that was invaded by foreign organizations. The successful

companies entered retailing in order to diversify their operations and gain vertical integration,

but at the same time production capabilities were enhanced and organizations started to

export their goods to penetrate foreign niches.

All the three case exemplify organizations that are able to follow changes in the environment

by diversifying according to the needs of the new developing market. These three cases

exemplify organizations that are able to align their strategy to the rapid change of the

environment and survive by changing form.

The analyzers in different decline conditions exemplify a few similarities and differences when

within group characteristics are analyzed. First, a difference exists in how diversification is used

between the two different types of change. In all of the cases in which niche shape changed, the

317

companies used diversification to minimize risk. In cases demonstrating niche shape changes,

diversification was used to move to new niche that was developing. This indicates that

depending on the type of change in the environment, analyzers can leverage their capability to

diversify in two different ways.

Second, in nine of the twelve cases diversification was carried out within the changing industry.

This indicated that while analyzers are able to diversify their operations, they seldom diversify

outside the industry where they operate. In addition, in two of the remaining cases the

organizations diversified by leveraging their current capabilities in a similar industry. This

indicates that while analyzers are able to diversify, they are constrained by the choices they

have made in the past. This is consistent with the argument of Miles et al. (1978) that

organizations are limited to the choices that top management view effective, which in this case

hold substantial lock in to the declining environment as the organization able to succeed in it.

To conclude the analysis, it appears that analyzers in declining industries are able to inhabit all

the decline conditions by leveraging their ability to both focus and diversify at the same time.

These cases exemplify the strategy type of analyzer as the organizations in the sample try to

minimize risk while maximizing opportunity for profit. Although, in this sample, this stance is

dualistic as the organizations which niche size decreases use diversification to minimize risk

while organizations whose niche shape is under transformation use diversification to maximize

the opportunity for profit. In addition, these cases exemplify how analyzers solve their primary

risk of not being able to move quickly due to their dualistic focus in both exploration and

exploitation. In the sample, the organizations have solved this problem by diversifying within

the same industry or leveraging their current capabilities in a similar industry.

4.2.4 Reactors in declining industries

Reactors are organizations that do not exhibit a proactive stance towards their environment

and do not have a clear strategic direction (Miles et al. 1978). Reactors can exist only in

protected environments where no change occurs. When faced with changing environments

reactors either have to move towards a consistent strategy or perish (Miles et al. 1978).

318

The sample features no reactors. Although this is in accordance with suggestions of Miles and

Snow (2003) as declining industries are not very welcoming, although it would be a non

sequitur argument to conclude anything on the basis that it is not proven directly by the data.

Therefore, although the lack of reactors does not prove their inability to succeed in declining

industries, is this argument supported by the typology.

5 DISCUSSION

This aim of this section is to discuss the findings presented earlier. Aim of this section is also to

draw together the findings that can be contrasted against the theoretical framework presented

earlier. In addition, these findings are contrasted to existing studies of the strategy typology to

uncover similarities and differences between the functioning of the typology in declining

industries and other contexts.

5.1 Miles and Snow strategy types in declining environments

In total, 25 cases were used to analyze the strategies organizations use in order to succeed in

declining industries. The figure 5 depicted the distribution of cases among these decline

conditions. This figure supplemented with the within group analyses, opens the field successful

strategic choices in different decline environments to an analysis of the fit of different strategy

types to different decline conditions. Through this analysis, these findings can then be

contrasted to the indications that were presented in the theoretical framework of the fit of

different strategies to different decline conditions.

Defenders were represented in three of the four decline conditions identified by Zammuto and

Cameron (1985), although seven of these cases resided in the condition of erosion that

suggested their fit to this type of decline. The within group analysis revealed, defenders in the

condition of erosion were the only ones that exemplified true success. This indicates that

defenders demonstrate a tendency to fit with the decline condition of erosion, as in the decline

condition where defenders were most prominent to succeed according to the data.

319

Therefore these findings are congruent with the postulations made in the theoretical

framework. As it was postulated that because defenders require an environment where major

shifts do not occur, the decline condition of erosion would fit them best.

Prospectors were represented in only one of the decline conditions identified by Zammuto and

Cameron (1985). All of the cases resided in the condition of collapse which indicated the fit

between this strategy type and the environment. The within group analysis revealed that

organizations in this decline condition exemplify success. This indicates that prospectors

demonstrate a tendency to fit with decline condition of collapse, as that is the only decline

condition where prospectors were found in the data. Despite all the cases were in the condition

of collapse, the small number of cases undermines these findings.

These findings are congruent with the postulations made in the theoretical framework. As it

was therefore assumed that because prospectors’ main advantage lies in their capability to

create change as well as rapidly respond to change, they would require an environment where

change is rapid.

Analyzers were represented in all of the decline conditions identified by Zammuto and

Cameron (1985). The within group analysis uncovered the reasoning behind this. It appears

that they are able to leverage diversification for two purposes. When niche size changes they

diversify to stabilize the operations of the organizations and when the niche shape changes

they diversify to expand to the new niche that is taking shape. This indicates that analyzers

demonstrate a tendency to align their selves to all the decline conditions by being able to

leverage different aspects of their hybrid domain in different environments.

These findings contradict the postulations made in the theoretical framework as it was

assumed there that analyzers are able to succeed only under the condition of dissolution in

which they can leverage their hybrid domain. Instead, according to the data, they are able to

inhabit all the decline conditions as they are able to shift balance according to the

environment.

320

Reactors were not present in the data. This is congruent with what was postulated in the

theoretical synthesis. It would be tempting to conclude that the data supports the theory but

the argument would be a non sequitur and hence no such argument will be made.

Figure 9: Revised version of the strategy-environment coalignment framework based on findings

Together, these findings enable adjustment of the theoretical framework to the empirical

findings of this research (figure 9). The findings of this study support the theoretical framework

regarding all strategy types except the analyzers. Analyzers appear to be able to leverage

different parts of their hybrid domain in different decline conditions enabling them to succeed

in all of the decline conditions identified by Zammuto and Cameron (1985). Therefore, whereas

defenders and prospectors succeed in distinct decline condition, the strategy type of analyzers can

be augmented to cover all the decline conditions. Based on this, figure 9 depicts a revised version

of the research framework that has been adjusted according to the empirical findings of this

research.

As the cases regard only successful organizations it can be implied that defenders exhibit a

tendency to succeed in erosion and prospectors exhibit a tendency to succeed in collapse.

Analyzers, as a combination of these two forms, have a tendency to succeed in all the decline

conditions as they are able to modify their hybrid domain to fit the current environments.

321

It must be emphasized, that the strategy types of defenders and analyzers were emphasized as

they together represented 22 out of the 25 cases. This indicates that successful organizations

operating in declining industries exhibit ties to the niche that cannot be easily broken even

when the niche declines. This is natural for defenders as they aim to seal and protect a domain

in the market and with their largest risk of inability to respond to major changes in the market,

which is also well exemplified by the data (Miles et al. 1978). Analyzers, on the other hand, have

the ability to diversify as they are a combination form of defenders and prospectors (Miles et al.

1978). Despite this, the analyzers in the sample were not generally able to diversify beyond the

industries they operate in. This indicates that both of the strategy types exhibit path

dependency, which at least to some extent can be explained by the notion that all the cases

were successful so dramatic changes, were not necessary, as following the old path maintained

their success.

On the other hand, the amount of defenders and analyzers may be overemphasized because of

the nature of the two other strategy types. Prospectors, according to their ideal definition,

should not even be present in the sample as they strive to create and exploit new market

opportunities. This is maintained by the data, as the three cases that did not exemplify strategy

type of an analyzer or defender were labeled as prospectors as they inhabited the environment

type of collapse. In addition, reactors were not present in the sample as reactors either perish

or assume a consistent strategy when they are faced with a changing environment (Miles et al.

(1978).

The emphasis of analyzers and defenders can also be due to way how data was collected. As

prospectors inhabit environments where change is rapid, the change in niche can rather be

called evolution than decline which helps to explain the low number of prospectors in the

sample.

5.2 Comparing existing literature to the empirical finding of this study

Snow and Hrebiniak (1980) in their research of strategy, distinctive competences and

organizational performance argue that as different types of organizational strategies exist

simultaneously in the same industrial environment, the effects of natural selection is hindered.

322

This research contradicts these findings as distinct types of strategies appear to be successful in

different decline conditions. This implies that although variation exists as Snow and Hambrick

(1980) state, selection and retention follow a distinct path (Aldrick 1979, p. 27).

This research, as well as the one done by Snow and Hrebiniak (1980), share an important

common feature. Analyzers in both of the studies exemplify various sources of competitive

advantage depending on the context. This research, therefore, both contradicts and supports

the suggestions of Snow and Hrebiniak (1980).

In Hambrick’s 1987 research of the Miles and Snow typology, he studied defenders and

prospectors in innovative and non-innovative growing and mature industries. In his study, the

prospectors outperformed defenders only in innovative industries. This is supported by current

findings, as prospectors were present only in the condition which was rapidly changing.

Hambrick (1983) also suggested that in general the superior strategy was neither of the extreme

types but rather the middle type of analyzer. This is supported in the light of this research as

the analyzers appear to be able to leverage their hybrid domain differently in different decline

conditions, enabling them to succeed in all of the four different decline conditions.

In the context of this research, contradictory evidence has been presented of whether

organizational adaptation is managerially or environmentally derived. This is congruent with

Hrebiniak and Joyce (1985) arguing that classifying change solely to either of these categories is

misleading. In this research, a clear distinction exists among the two types of change.

Successful organizations in niches in which the niche size changes do not try to alter the

trajectory of decline but rather take it as granted and either hold their position in the niche or

diversify to related industries, therefore enhancing the pattern of decline. On the other hand,

successful organizations that operate in niches where the niche shape changes, aim at both

changing the industry and following such change, enhancing the pattern of niche shape

change. In light of these findings, choice is both a cause and a consequence of the

environmental condition in which the organizations operate as successful organizations drive

the niches to certain directions. This is consistent with the suggestions by Hrebiniak and Joyce

323

(1985), who postulate that choice is both a cause and a consequence of environmental

influences that organizations face.

Originally, a central premise of the typology was that the choice of defender, analyzer and

prospector strategy type can lead to effective performance, when the types are properly

implemented (Zahra and Pearce 1990; Miles et al. 1978). This has received both supportive and

contradictory findings (Hambrick 1983; Zajac and Shortell 1989; Smith et al. 1989; Zahra and

Pearce 1990). In the light of this research, analyzers appear to be able to succeed in all decline

conditions when the strategy type is properly implemented, supporting this premise. Defenders

and prospectors do not appear in all of the decline conditions and therefore contradict this

premise. Therefore, this study contradicts the aforementioned premise, as two of the three

strategy types were not able to succeed in all of the decline conditions.

It appears that while some of the findings are unique, the findings exhibit a tendency to be

similar to those presented in earlier studies regarding the strategy typology. This implies that

while declining industries differentiate from other industries, as their carrying capacity

decreases, correct alignment of strategy and environment leads organization to success, despite

the hostile nature of the environment.

6 CONCLUSIONS

This research started out with a broad research question of the kinds of strategies that lead

organization to success in declining industries. This problem was broken down into three

distinct research questions which will be examined here one by one to formulate a conclusion

for this research.

The first question regarded the array of strategies that are successful in declining industries.

The Miles and Snow (2003) strategy typology was used to exemplify different strategy types that

organizations can assume.

It appears that when an organization has a proactive stance towards its environment it holds a

change of succeeding in a declining industry. This is exemplified by the fact that all strategy

types except reactors were able to succeed in declining industries. This is also supported by

324

Miles et al. (1978) as they note that reactors are the only strategy type that is reactive towards

its environment and that when it faces changing environments it will either perish or assume

consistent form of strategy.

Now, it has been established that three strategy types can succeed in declining industries, it is

possible to make conclusions of the second research question. The second research question

regards the positioning of success strategies in declining industries. To portray different decline

conditions the typology of decline conditions by Zammuto and Cameron (1985) was used. It

appears that different strategy types have different kinds of tendencies to succeed in different

decline conditions. Defenders have a tendency to succeed in decline conditions where the

decline is continuous and where the niche size diminishes. This is also supported by Miles et al.

(1978) who postulate similar behavior for defenders.

Prospectors exemplify a tendency to succeed in decline conditions where the niche shape

changes discontinuously as they are able to follow changes in the niche and also create such

change. This is consistent with the suggestions of Miles et al. (1978) as they state that the prime

capability of prospectors is finding and exploiting new market opportunities.

Analyzers exemplify a tendency to succeed in all of the decline conditions, as they are able to

leverage their hybrid domain differently in different decline conditions. In decline conditions

where the niche size decreases, they diversify to enhance stability of their business. When faced

with situations in which the niche shape changes they tend to diversify to the new emerging

niche to accomplish congruence with the new emerging niche. These results are congruent

with Snow and Hrebiniak (1980) as their research implied that analyzers exemplify various

sources of competitive advantage. It is also congruent with the notions of Hambrick (1983) who

reached an outcome that analyzers in general are the most successful for of strategy.

As the three successful strategy types have been positioned to the field of decline conditions, it

is possible to proceed into making conclusions regarding the third and final research question.

The third question regarded how the strategy types as a whole are aligned to the decline

condition.

325

A pattern emerges among the three successful strategy types. The extreme strategy types inhabit

decline conditions that are at the two opposite ends of the spectrum. Defenders succeed in

decline conditions in which the decline occurs steadily and the niche size decreases.

Prospectors, on the other hand, succeed in the opposite decline conditions as they succeed

when niche shape changes discontinuously. Behaviors of both of these types are congruent

with the suggestions of Miles et al. (1978) regarding the environments where the different

strategy types succeed. Analyzers, strategy type that is a combination of the two extreme types,

can succeed in all the different decline conditions as they are able to leverage their hybrid

domain differently in different decline conditions. Therefore, depending on the decline

condition, analyzer can succeed in different ways.

To conclude, the research was able to address all the research questions and yield outcomes

that signify the different sources of competitive advantage in different decline conditions and

among the different strategy types. In addition, patterns emerged among the different strategy

types in different decline conditions. As such, it can now be concluded that research yielded

successful process with distinct outcomes.

6.1 Theoretical implications

This research has extended the application of Miles and Snow typology to declining industries

that were classified by the typology of decline conditions by Zammuto and Cameron (1985).

Theoretically the research has produced a few implications to the study of the Miles and Snow

(2003) typology.

First, the research has yielded similar outcomes as those suggested by Snow and Hrebiniak

(1980) and Hambrick (1983), as analyzers appear to be a superior type of strategy as they can

leverage their hybrid domain differently according to the decline condition. This research is

also congruent with Hambrick (1983) as the research implies that innovators prosper only in

rapidly evolving industries. On the basis of these outcomes, the behavior of different strategy

types in declining industries are similar to other industries. Therefore, while the context of

declining industries differs from other context due to its hostile nature, the successful

strategies do not deviate from other contexts.

326

Second, a central premise of the typology is that defender, analyzer and prospector strategy

types can lead to effective performance, when the types are properly implemented (Zahra and

Pearce 1990; Miles et al. 1978). This has received both supportive and contradictory evidence

(Hambrick 1983; Zajac and Shortell 1989; Smith et al. 1989; Zahra and Pearce 1990). This study

partially contradicts this premise, as only analyzers were able to succeed in all the decline

conditions, whereas defenders and prospectors were able to succeed only in certain decline

conditions.

This research also strengthens the argument of Hrebiniak and Joyce (1985), who argue that

choice is both a cause and a consequence of environmental influence. This research suggests

that the type of decline is both caused by the successful strategic choices and a consequence of

these choices. Therefore, successful strategies drive the niche to either a direction of

formulation of a new niche or the death of the current niche.

6.2 Managerial implications

The first and most relieving managerial implication is that it is possible to succeed in declining

industries. For most of managers even the thought of operating in such environment is scary.

Hence, hopefully this research can raise hope and illuminate the possibilities for success in

declining industries. As Ghemawatt and Nalebuff (1990) have stated, more than 10% of the

United States manufacturing output in 1977 was generated by a declining industry. Hence

declining industries should rather be seen as an opportunity where one can succeed if the

organizations strategy is configured in a correct manner.

What is managerially important in this research is that different strategy types succeed in

different kinds of decline conditions. This implies that managers leading organizations

operating in declining industries should do two kinds of things. First, they should try to

establish the kind of decline the organization is facing, and, secondly, reconfiguring strategy to

comply with the requirements of the environment to achieve strategic fit. This however is not

as easy as it may sound, as Miles et al. (1978) have also stated that breaking the pattern of

behavior is very hard as managers tend to respond similarly to changes.

327

Despite the positive outcomes of this research, this research reported only success and hence

does not tell anything about the failure rate in such industries. Therefore the success in these

kinds of environments is in no way guaranteed. Success, even with a correct strategy, cannot be

guaranteed in any way.

To conclude, although this research has presented possibility of success in declining industries,

these findings should be taken with a precaution as declining industries are environments

where by definition the market is diminishing. Despite this, the declining industries should not

be feared as they represent a substantial portion of for example manufacturing industries.

Hence, when faced with such conditions the organizations should ensure fit between the

strategy of the organization and the declining industry and continue operations accordingly.

6.3 Limitations and suggestions for further research

This research has three prime limitations that have to be stated explicitly. First, the

methodology used to conduct this research is primarily aimed to be done by a research team.

This is due to two reasons. First, to increase reliability of coding the persons coding the cases

and those designing the theory should be separated to minimize coding bias. Secondly, the

coding should be done by multiple raters to increase reliability of the coding. These problems

however were not possible to be solved in the context of this research and are major limitations

of this study. Therefore it would be beneficial to redo the research with a research team to

minimize coding bias and increase the reliability of the results.

Second, the sample size in this research was low as research in declining industries is not case

intensive area. In addition, many cases did not have adequate information to complete the

survey. This undermines the reliability of the findings of this research. Increasing the size of the

sample would increase the reliability of the results. Prospectors were few among the sample

and by increasing the sample size; a deeper understanding of their behavior in declining

industries could be gained. This might also be partly due to the fact that conditions of changing

niche shapes might not be considered conditions of decline but of evolution and change.

328

Size of the organization relative to the niche was left outside of the scope of this research as it

was not possible to accommodate it into the research. This leaves a gap in the research as size

affects the power of organizations to steer the niche in which they operate in. By filling this gap

a consensus could be reached of how much the choice of strategy affects success and how much

the size of the organization affects this.

As this research has regarded only success strategies in declining industries, it would be

interesting to contrast the success and failure rates of different strategy types in declining

industries. This would yield insight into what kind of risk is involved in different strategic

choices. By doing so, one could calculate the success/failure rates of different strategy types and

indicate whether certain strategic types have more risk than others.

The research was designed to produce only a snapshot. Therefore, it would be essential to gain

more knowledge of the ability of organizations to change their strategy type when faced with

decline and therefore succeed in the environment. This would provide crucial information of

how the dynamics of strategic choice and environmental selection function in the declining

industries.

329

REFERENCES

Abernathy, William J. and Kim B. Clark (1975), “Innovation: Mapping the winds of creative

destruction.” Research Policy, 14, pp. 3-22.

Agarval, Rajhsree and Barry L. Bayus (2004), “Creating and surviving in new industries.”

Advances in strategic management, 21, pp. 107-130.

Aldrich, Howard (1979), Organizations and Environments. Stanford, Stanford University Press

Anderson, Philip and Michael L. Tushman (1990), “Technological Discontinuities and

Dominant Designs: A Cyclical Model of Technological Change.” Administrative Science

Quarterly, 35, pp. 604-633.

Astley, W. Graham and Andrew H. Van De Ven (1983), “Central Perspectives and Debates in

Organization Theory.” Administrative Science Quarterly, 28 (2), pp. 245-273.

Behling, Orlando (1980), “The Case for the Natural Science Model For Research in

Organizational Behavior And Organization Theory.” Academy of Management Review, 5

(4), pp. 483-490.

Bullock, R. J. (1986), “A Meta-Analysis Method for OD Case Studies.” Group & Organization

Studies, 11 (1-2), pp. 33-48.

Cameron, Kim S. (1983), “Strategic responses to conditions of decline: Higher education and the

private sector.” Journal of Higher Education, 54, pp. 359-380.

Cameron, Kim S., Myung U Kim and David A. Whetten (1987),”Organizational Effects of

Decline and Turbulence.” Administrative Science Quarterly, 32, pp. 222-240.

Cameron, Kim S. and Raymond Zammuto (1983), “Matching Managerial Strategies to

Conditions of Decline.” Human Resource Management, 22 (4), pp. 359-375.

Chaganti, Rajeswararao and Rakesh Sambharay (1987), “Strategic Orientation and

Characteristics of Upper Management.” Strategic Management Journal, 8 (4), pp. 393-

401.

Child, John (1972), “Organizational structure, environment and performance: The role of

strategic choice” Sociology, 6, pp. 1-22.

Child, John (1997), “Strategic Choice in the Analysis of Action, Structure, Organizations and

Environment: Retrospect and Prospect.” Organization Studies, 18 (1), pp. 43-76.

Collis, Jill and Roger Hussey (2003), Business Research: A practical guide for undergraduate and

postgraduate students. New York, Palgrave MacMillan.

Conant, Jeffrey S., Michael P. Mokwa and P. Rajan Varadarajan (1990), “Strategic types,

Distinctive Marketing Competencies and Organizational Performance: A Multiple

Measures-Based Study” Strategic Management Journal, 11 (5), pp. 365-383.

330

Cox, William E. Jr. (1967), “Product Life Cycles as Marketing Models.” The Journal of Business,

40 (4), pp. 375-384.

Cunningham, M. T. (1969), “The application of product life cycles to corporate strategy: some

research findings.” European Journal of Marketing, 3 (1), pp. 32-44.

Dean, Joel (1950), “Pricing Policies for New Products.” Harvard Business Review, 28 (6), pp. 45-

53.

Eisenhardt, Kathleen M. (1989), “Building Theories from Case Study Research.” Academy of

Management Review, 14 (4), pp. 532-550.

Eisenhardt, Kathleen M. and Melissa E. Graebner (2007),”Theory building from cases:

Opportunities and challenged.” Academy of Management Journal, 50 (1), pp. 25-32.

Fein, Adam J. (1998), “Understanding evolutionary processes in non-manufacturing industries:

Empirical insight from the shakeout in pharmaceutical wholesaling.” Journal of

Evolutionary Economics, 8 (3), pp. 231-270.

Filson, Darren (2002), “Product and process innovations in the life cycle of an industry.” Journal

of Economic Behavior & Organization, 49, pp. 97-112.

Ghemawat, Pankaj and Barry Nalebuff (1985),”Exit.” Rand Journal of Economics, 16 (2), pp. 184-

194.

Ghemawat, Pankaj and Barry Nalebuff (1990),”The Devolution of Declining Industries.” The

Quarterly Journal of Economics, 105 (1), 167-186.

Gibbert, Michael, Winfried Ruigrok and Barbara Wicki (2008), “What passes as a rigorous case

study?” Strategic Management Journal¸ 29 (13), pp. 1465-1474.

Goldfarb, Brent, David Kirsch and David A. Miller (2007), “Was there too little entry during the

Dot Com Era?” Journal of Financial Economics, 86, pp. 100-144.

Gort, Michael and Steven Klepper (1982), “Time path in the Diffusion of Product Innovations.”

The Economic Journal, 92 (367), pp. 630-653.

Hambrick, Donald C. (1983), “Some Tests of the Effectiveness and Functional Attributes of

Miles and Snow’s Strategic Types.” Academy of Management Journal, 26 (1), pp. 5-26.

Hambrick, Donald C. and Phyllis A. Mason (1984), “Upper Echelons: The Organization as a

Reflection of Its Top Managers.” The Academy of Management Review, 9 (2), pp. 193-206.

Hannan, Michael T. and John Freeman (1977), “The Population Ecology of Organizations.”

American Journal of Sociology, 82 (5), 929-964.

Harrigan, Kathryn Rudie (1980a), “Strategy Formulation in Declining Industries.” Academy of

Manageent Review, 5 (4), pp. 599-604.

Harrigan, Kathryn Rudie (1980b), “Strategies for declining industries.” Journal of Business

Strategy, 1 (2), pp. 20-34.

331

Harrigan, Kathryn Rudie and Michael Porter (1983), “End-game strategies for declining

industries.” Harvard Business Review, 61 (4), pp. 111-120.

Hrebiniak, Lawrence G. and William F. Joyce (1985), “Organizational Adaptation: Strategic

Choice and Environmental Determinism.” Administrative Science Quarterly, 30 (3), pp.

336-349.

James, William L. and Kenneth J. Hatten (1995), “Further Evidence on the Validity of the Self

Typing Paragraph Approach: Miles and Snow Strategic Archetypes in Banking.”

Strategic Management Journal, 16 (2), pp. 161-168.

Jonavic, Boyan and Glenn M. MacDonald (1994), “The Life Cycle of a Competitive Industry.”

Journal of Political Economy, 102 (21), pp. 322-347.

Kim, W. Chan and Renée Mauborgne (2005),”Blue ocean strategy.” Harvard Business Review, 82

(10), pp. 76-84.

Kincheloe, Joel L. and Kenneth Tobin (2009), “The much exaggerated death of positivism.”

Cultural Studies of Science Education, 4 (5), pp. 513-528.

Klepper, Steven (1996),”Entry, Exit, Growt, and Innovation over the Product Life Cycle.” The

American Economic Review, 86 (3), pp.562-583.

Klepper, Steven (1997),”Industry Life Cycles.” Industrial and Corporate Change, 6 (1), pp. 145-181.

Larsson, Rikard (1993), “Case Survey Methodology: Quantitative Analysis of Patterns across

Case Studies.” Academy of Management Journal, 36 (6), pp. 1515-1546.

Lucas, Willian A. (1974), “The case survey method: aggregating case experience.” Santa Monica,

The Rand Corporation

Malhotra, Naresh K. and David F. Birks (2007), Marketing Resaerch: An Applied Approach,

Essex: Pearson Education Limited.

McDaniel, Stephen W. and James W. Kolari (1987), “Marketing Strategy Implications of the

Miles and Snow Strategic Typology.” Journal of Marketing, 51 (4), pp. 19-30.

Meyer, Alan D. (1982), “Adapting to environmental jolts” Administrative Science Quarterly, 27

(4), pp. 515-537.

Miles, Raymond E., Charles C. Snow, Alan D. Meyer and Henry J. Coleman Jr. (1978),

“Organizational Strategy, Structure and Process.” Academy of Management Review, 3

(3), pp. 546-562.

Miles, Raymond E. and Charles C. Snow (2003), Organizational, Strategy, Structure and Process,

Stanford: Stanford University Press.

Mintzberg, Henry (1978),”Patterns in strategy formulation.” Management Science, 24 (9), pp.

934-948.

332

Mintzberg, Henry and James A. Waters (1982), “Tracking Strategy in an Entrepreneurial Firm.”

Academy of Management Journal, 25 (3), pp. 465-499.

Mintzberg, Henry and James A. Waters (1982), “Of Strategies, Deliberate and Emergent.”

Strategic Management Journal, 6 (3), 257-272.

Moore, Marguerita (2005), “Towards a confirmatory model of retail strategy types: an empirical

test of miles and snow” Journal of Business Research, 58, pp.696-704.

Prahalad, C. K. and Gary Hamel (1994), “Strategy as a Field of Study: Why Search for a New

Paradigm?” Strategic Management Journal, 15 (Summer), 5-16.

Prescott, John E. (1986), “Environments as moderators of relationship between strategy and

performance.” Academy of Management Journal, 29 (2), pp. 329-346.

Porter, Michael E. (1985), Competitive Advantage: Creating and Sustaining Superior

Performance. New York: The Free Press.

Quinn, Robert E. and Kim Cameron (1983), “Organizational Life Cycles and Shifting Criteria of

Effectiveness: Some Preliminary Evidence.” Management Science, 29 (1), pp. 33-51.

Reynold, Stanley S. (1988), “Plant Closing and Exit Behaviour in Declining Industries”

Economica, 55 (220), pp. 493-503.

Segev, Eli (1989), “A Systematic Comparative Analysis and Synthesis of Two Business-Level

Strategy Typologies.” Strategic Management Journal, 10 (5), pp. 487-505.

Shortell, Stephen M. and Edward J. Zajac (1990), “Perceptual and Archival Measures of Miles

and Snow’s Strategic Types: A Comprehensive Assessment of Reliability and Validity.”

The Academy of Management Journal, 33 (4), pp. 817-832.

Smith, Ken G., James P. Guthrie and Ming-Jer Chen (1989), “Strategy, Size and Performance”

Organization Studies, 10 (1), pp. 63-81.

Snow, Charles C. and Donald C. Hambrick (1980), “Measuring Organizational Strategies: Some

Theoretical and Methodological problems.” The Academy of Management Review, 5 (4),

pp. 527-538.

Snow, Charles C. and Lawrence G. Hrebiniak (1980), “Strategy, Discinctive Competence, and

Organizational Performance.” Administrative Science Quarterly, 25 (2), pp. 317-336.

Soanes, Catherine and Sara Hawker (2005), Compact Oxford English Dictionary of Current

English, Oxford: Oxford University Press.

Stake, Robert E. (1995), The Art of Case Study Research. Thousand Oaks: Sage Publications.

Taggart, J. H. (1995), “Strategy Formulation in Declining Industries: A Biology Paradigm.”

Journal of Marketing Management, 11 (4), pp. 295-314.

Tushman, Michael L. and Philip Anderson (1986), “Technological Discontinuities and

Organizational Environments.“ Administrative Science Quarterly, 31 (3), pp. 439-465.

333

Utterback, James M. and William J. Abernathy (1975), “A Dynamic Model of Process and

Product Innovation.” The International Journal of Management Science, 3 (6), pp. 639-

656.

Venkatraman, N. and John C. Camillus (1984), “Exploring the Concept of “Fit” in Strategic

Management.” Academy of Management Review, 9 (3), pp. 513-525.

Venkatraman, N. and John E. Prescott (1990), “Environment-Strategy Coalignment: An

Empirical Test of Its Performance Implications.” Strategic Management Journal, 11 (1),

pp. 1-23.

Whetten, David A. (1980), “Organizational Decline: A Neglected Topic in Organizational

Science.” The Academy of Managemet Review, 5 (4), 577-588.

Williamson, Oliver E. (1975), Markets and hierarchies: Analysis and Antitrust Implications. New

York: The Free Press.

Yin, Robert K. (2003), Case study research: Design and Methods. Thousand Oaks: Sage

Publications.

Yin, Robert K. and Karen A. Heald (1975), “Using Case Survey Method To Analyze Policy

Studies.” Administrative Science Quarterly, 20 (3), pp. 371-381.

Yin, Robert K. and Douglas Yates (1974), “Street-Level Governments: Assessing

Decentralization and Urban services (An Evaluation of Policy Related Research).”

report, National Science Foundation, Washington.

Zahra, Shaker A. and John A. Pearce (1990), “Research Evidence On The Miles-Snow Typology.”

Journal of Management, 16 (4), pp. 751-768.

Zajac, Edward J. and Stephen M. Shortell (1989), “Changing Generic Strategies: Likehood,

Direction, and Performance Implications” Strategic Management Journal, 10 (5), pp. 413-

430.

Zammuto, Raymond F. (1983), ”Growth, stability, and decline in American college and

university enrollment.” Educational Administration Quarterly, 19, pp. 83-99.

Zammuto, Raymond (1988), “Organizational adaptation: Some implications of organizational

ecology for strategic choice.” Journal of Management Studies, 25 (2), pp. 105-120.

Zammuto, Raymond D., Kim S. Cameron (1985), “Environmental decline and organizational

response” in Research in Organizational Behavior, Vol. 7, L. L. Cummings and Barry M.

Staw, eds. Greenwich, JAI PRESS INC, 223-262.

334

APPENDICES

Appendix 1: Cases in the sample

335

336

Appendix 2: Survey form used to generate the data

Environmental decline and success strategy type

Survey questionnaire

case number:

Product-market: Product Service

Served market: B2B B2C

Environmental decline (Zammuto and Cameron 1985)Type of change

Change in in the niche results from:

Change in niche shape Change in niche size

1 2 3 4 5 6 7

Continuity of change

Change in the niche size is:

Continuous Discontinuous

1 2 3 4 5 6 7

Strategy type ( adapted from Snow and Hrebiniak 1980)

Which one of the following descriptions most closely fits the strategy of the organization?

Type 1 This type of organization attempts to locate and maintain a secure niche in the product market. The

organization tends to offer a more limited range of products or services than its competitors, and it tries to

protect its domain by offering higher quality, superior service, lower prices, and so forth. Often this type of

organization is not at the forefront of developments in the industry - it tends to ignore industry changes that

have no direct influence on current areas of operation and concentrates instead on doing the best job possible

in a limited area.

Type 2 The organization values being "first in" in new product and market areas even if not all of these efforts

prove to be highly profitable. The organization responds rapidly to early signals concerning areas of

opportunity, and these responses often lead to a new round of competitive actions. However, this type of

organization may not maintain market strength in all of the areas it enters.

Type 3 This type of organization attempts to maintain a stable, limited line of products or services, while at the

same time moving out quickly to follow a carefully selected set of the more promising new developments in

the industry. The organization is seldom "first in" with new products or services. However, by carefully

monitoring the actions of major competitors in areas compatible with its stable product-market base, the

organization can frequently be "second in" with a more cost-efficient product or service.

Type 4 This type of organization does not appear to have a consistent product-market orientation. The

organization is usually not as aggressive in maintaining established products and markets as some of its

competitors, nor is it willing to take as many risks as other competitors. Rather, the organization responds in

those areas where it is forced to by environmental pressures.

337

Appendix 3: Types of product life cycle patterns

(Adapted from Rink and Swan 1979)

Documents

capabilities and competitive advantage in the forest industry