Stimulating different types of eco-innovation in the UK: Government policies and firm motivations

Ecological Economics 70 (2011) 1546–1557

Contents lists available at ScienceDirect

Ecological Economics

j ourna l homepage: www.e lsev ie r.com/ locate /eco lecon

Analysis

Stimulating different types of eco-innovation in the UK: Government policies andfirm motivations

Pelin Demirel ⁎, Effie KesidouNottingham University Business School, UK

⁎ Corresponding author at: Nottingham University BuNottingham NG8 1BB, UK. Tel.: +44 115 8467420.

E-mail address: [email protected] (P.1 Environmental technologies cover a broad range o

tions aimed at alternative energy production or providiproblems (Cooke, 2008). Eco-innovation is defined as “thnew, or significantly improved, products (goods andmethods, organisational structures and institutional arrangintent – lead to environmental improvements compared t2009 p 19).

0921-8009/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.ecolecon.2011.03.019

a b s t r a c t
a r t i c l e i n f o
Article history:Received 5 December 2010Received in revised form 21 February 2011Accepted 24 March 2011

JEL classification:Q5 environmental economicsO3 technological change

Keywords:Cleaner productionEnvironmental regulationEnvironmental taxesEnvironmental Management SystemsEco-R&DISO14001

In this paper, we adopt a recent OECD framework and examine the role of external policy tools and internalfirm specific factors for stimulating three different types of eco-innovations that range on a spectrum of lowerto higher technological and environmental impacts: End-of-Pipeline Pollution Control Technologies, IntegratedCleaner Production Technologies and Environmental R&D. Using a novel firm-level dataset from a DEFRA survey,we estimate a Tobit model, which provides empirical evidence showing that these eco-innovations aremotivated by different external policy tools and internal firm specific factors. Our findings indicate that End ofPipeline Technologies and Integrated Cleaner Production Technologies are mainly driven by equipment upgrademotives with a view of improving efficiency while environmental regulations are effective in stimulating theEnd-of-Pipeline technologies and Environmental R&D. Interestingly, alongside government induced regulations,we find that market factors, mainly motivated by cost savings, are effective in driving Environmental R&D.Finally, ISO14001 certification is effective in strengthening the positive impact of environmental managementsystems on both End-of-Pipeline technologies and Environmental R&D while CSR policies have no significantimpact on motivating any of the eco-innovations.

siness School, Wollaton Road,

Demirel).f different technology applica-ng solutions to environmentale creation or implementation ofservices), processes, marketingements which –with or withouto relevant alternatives” (OECD,

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Due to growing concerns for the environmental impact of the indus-trial society, governments are carefully considering their strategies forsustainable development; an increasingly popular philosophy whichpromotes that the ‘environment should no longer be sacrificed toeconomic growth: rather, the two should be reconciled’ (Aggeri, 1999,706). In easing the unambiguous trade-offs between environmentalprotection and economic growth, eco-innovations have a central roleto play through improving environmental technologies that measure,detect and treat pollution, avoid it at the source, and ensure that the endproduct has a life span with minimal environmental impact.1

An expanding body of empirical and theoretical literature on eco-innovation aims to understand the circumstances which are moreconducive to environmental technology investments. The ‘ecological,economic and social’ dimensions of eco-innovations require an inter-

disciplinary approachwhich combines insights from environmental andinnovation economics and is aware of the different methodologicallenses of theneoclassical and evolutionary schools of thought (Rennings,2000, p. 322).

In this paper, by merging the valuable insights from thesedisciplines, we examine the determinants of investment into differenttypes of eco-innovations: End-of-Pipeline Pollution Control Technolo-gies, Integrated Cleaner Production Technologies and EnvironmentalR&D. Our findings, based on unique data from DEFRA's firm levelsurvey “Environmental Protection Expenditure by Industry”, provideimportant insights on the specific external policy tools and internalfirm factors that affect each type of eco-innovation by incorporatingall three types of eco-innovations in the same conceptual framework.

The paper is structured in the following way: Section 2 discusses thebackground literature and the proposed conceptual framework of thepaper. Section3presents thedata andmethodologyused for the empiricalanalysis. The results are discussed in Section 4. Finally, Section 5 includesthe conclusions and discusses the policy implications of the study.

2. Background Literature and Conceptual Framework

2.1. Types of Eco-innovation

A generalmodel of eco-innovation in linewith the OECD framework(2009) is presented in Fig. 1. We develop this model further to includea more detailed account of the characteristics specific to each type of

http://dx.doi.org/10.1016/j.ecolecon.2011.03.019

mailto:[email protected]

http://dx.doi.org/10.1016/j.ecolecon.2011.03.019

http://www.sciencedirect.com/science/journal/09218009

Fig. 1. Categorisation of eco-innovation.

3 Examples of integrated technologies include the following (Ashford, 1994):(a) improved housekeeping, which refers to improvements in management practices,monitoring, andmaintenance; (b) changes to process technologies, through optimization,which conserves raw materials and energy; (c) changes to products with the use of new

1547P. Demirel, E. Kesidou / Ecological Economics 70 (2011) 1546–1557

eco-innovation, namely, End-of-Pipeline pollution Control Technolo-gies, Integrated Cleaner Production Technologies and EnvironmentalR&D (Kemp, 1997; Frondel et al., 2007). As indicated in the figure, theseeco-innovations range on a spectrum from lower impact –more incre-mental innovations in pollution control technologies– to higher impactinnovations in environmental R&D.

2.1.1. End-of-Pipeline Pollution Control TechnologiesManufacturing firms apply end-of-pipeline solutions in order to

treat, handle,measure or dispose emissions andwastes fromproduction(DEFRA, 2006).2 As the name suggests, these technological solutionsare incorporated into existingmanufacturing processes at thefinal stageand are not essential parts of the production process. End-of-pipelinetechnologies leave theproductionprocessmostly unchanged; therefore,they are considered to be highly incremental innovations. Since end-of-

2 Examples of end-of-pipeline technologies include effluent treatment plant andexhaust air scrubbing systems (DEFRA, 2006).

pipeline solutions denote the implementation of non-essential tech-nologies, companies perceive them as costly investments that hampertheir competitiveness (OECD, 2009; Porter and van der Linde, 1995a,1995b).

2.1.2. Integrated Cleaner Production TechnologiesIntegrated technologies refer to new or modified production

facilities, which are more efficient than previous technologies, andcontribute to pollution reduction by cutting down the amount of inputsused for production and/or by substituting the inputs with more envi-ronmentally friendly alternatives3 (OECD, 2009). Similar to the end-of-

technologies, which reduces the consumption of resources, waste and emissions; and(d) changes to inputs by substituting toxic materials with environmentally friendlyalternatives.

http://archive.defra.gov.uk/evidence/statistics/environment/envsurvey/expn2007/index.htm


1548 P. Demirel, E. Kesidou / Ecological Economics 70 (2011) 1546–1557

pipeline technologies, integrated cleaner production technologiesmostly represent environmental process innovations (Renning et al.,2006). However, integrated cleaner production technologies aredesigned to ensure that environmental protection is an integral partof manufacturing processes. In other words, unlike end-of-pipelinesolutions that attempt to control pollution by adopting an ‘after-event,react and retreat approach’, integrated solutions focus on preventingpollution by adopting a ‘forward-looking, anticipate and preventphilosophy’ (Ashford, 1994, p. 4). Compared to the continuousincreasing costs of end-of-pipeline technologies, integrated technolo-gies are less costly, since they have the potential to save costs byreducing the use of raw materials, energy and the costs of complyingwith regulations (Ashford, 1994).

2.1.3. Environmental R&DThe main aim of environmental R&D is to improve products and

processes byproviding solutions for cleanerproduction and consumption.Manufacturing firms that conduct environmental R&D on a systematicbasis attempt to increase the stock of knowledge in the field of envi-ronmental protection and use this knowledge to devise new appli-cations (DEFRA, 2006). Environmental R&D has a higher technologicalimpact compared to thepreviouslydiscussed categories of eco-innovationbecause (1) it enhances absorptive capacity as environmental R&Dbroadens the horizons of the company in environmental matters (Cohenand Levinthal, 1990) and (2) the scope of environmental R&D is not onlylimited to process innovations but also covers product innovations.Similar to the case of generic R&D, environmental R&D is subject to highrisks and high costs.

2.2. Drivers of Different Types of Eco-Innovations

Since the three types of eco-innovations have different character-istics, we expect that they may have different drivers. In this paper weconsider the role of two broad drivers: External Environmental PolicyInstruments and Internal Firm-Level Motivations.

2.2.1. External Environmental Policy InstrumentsVarious studies from environmental economics and eco-innova-

tion literatures attempt to decipher whether command and controlpolicies (such as environmental regulations) or market-based in-struments (such as environmental taxes) are more effective and costefficient in reducing pollution. Environmental economists see Regu-latory Push as themain driver of eco-innovation and often suggest thatmarket-based instruments are superior to command and controlpolicies in reducing pollution due to cost-efficiency and flexibilityadvantages4 (Milliman and Prince, 1989; Requate and Unold, 2003;Requate, 2005).

On the other hand, moving the emphasis from immediate pollutioncontrol to environmental innovations, the recent developments in eco-innovation studies highlight the value of command and control mecha-nisms for driving eco-innovations. Specifically, Porter (1991) and Porterand Van Der Linde (1995a, 1995b) have shown that environmentalregulations indeed create ‘win–win’ situations: Firms achieve highprofits and produce “green products” because environmental regula-tions boost R&D activities and thus, stimulate innovation and economicgrowth (Hart, 2004; Popp, 2005; Rhothfels, 2002). An alternativeexplanation of the “win–win” situation is offered by Rhothfels (2002)who shows that compliance with environmental regulations can drivefirms to become leaders in “green markets” and thus, become morecompetitive compared to their foreign peers.

4 Environmental economists use models, which treat pollution as a negativeexternality; the generators of these externalities are induced through a set ofPigouvian taxes and/or tradable permits to pay for the full range of social costs thattheir activities entail (Baumol and Oates, 1988).

Besides highlighting the value of environmental regulations, eco-innovation literature points out that the traditional economicapproach fails to understand the dynamics of different degrees ofincremental changes in environmental technologies and the drivingfactors behind these changes (Kemp, 1997; Nill and Kemp, 2009;Rammel and van den Bergh, 2003; van den Bergh et al., 2007). Inparticular, the traditional environmental economics approach focusesmainly on the impact of policy instruments upon the mostincremental types of eco-innovation, namely the end-of-pipelinesolutions and ignores higher impact innovations in integrated cleanerproduction technologies or environmental R&D (Kemp and Pontoglio,2008). This conceptual gap is addressed by a limited number ofstudies amongwhich, the study by Frondel et al. (2007) most stronglyconfirms the necessity to examine the impact of external policyinstruments on different types of environmental technologies. Theirresults indicate that command and control mechanisms play animportant role in stimulating end-of-pipeline solutions but a minimalrole for integrated cleaner production technologies. Market-basedinstruments, on the other hand, do not affect either type of eco-innovation. Similar findings regarding command and control mech-anisms are reported by the OECD survey (Johnstone, 2007) but theirresults on market-based instruments (i.e. taxes) differ, suggestingthat taxes stimulate changes in the production process. Cleff andRennings (2000) also find evidence in favour of market-based in-struments boosting eco-innovations.

In assessing the relationship between external policy instrumentsand environmental technologies, the complexity of environmentaltechnologies require new socio-technical paradigms which need tocombine market-based and command and control policies (Geels andSchot, 2007). Kemp (1997) argues that there is no single best policyinstrument to stimulate clean technology and different instrumentsmay play important roles depending on the context they operate inand the type of clean technology that needs to be stimulated. IndeedJaenicke et al. (2000) find that a combination of instruments performsbest in stimulating eco-innovations. Frondel et al. (2007) also agree thatthe type of instrument is less important but it is critical to ensure thatthese instruments sustain a stringent environmental policy.

In the empirical part of this paper, we investigate the role playedby the environmental regulations and environmental taxes imple-mented in the UK for boosting the different types of eco-innovationsdiscussed in Section 2.1.

2.2.2. Internal Firm-Level MotivationsIn this section, we focus on three firm-level motivations that are

potential drivers of eco-innovations; namelyOrganisational Capabilities,Efficiency, and Corporate Image.

(a) Organisational Capabilities: EnvironmentalManagement Systems(EMS) engender important organisational capabilities in the areaof environmental protection. EMS are voluntary organisationalframeworks that detail the procedures to manage the impacts ofthe organisation on natural environment (Darnall, 2006). EMSare aimed at the continuous improvement of corporate environ-mental performance with an attempt to get ahead of the existinggovernment regulations to reduce emissions and waste disposal(Kollman and Prakash, 2002). European Union's EnvironmentalManagement and Audit Scheme (EMAS) and ISO14001 consti-tute themostdiffused formsof formalisedEMSandboth schemesrequire third party certification and investigation. Bansal andHunter (2003) argue that these two schemes reinforce legitimacywhich cannot be claimed through in-house EMS.There is little consensus on the impact of EMS upon environ-mental performance or eco-innovation. Russo and Harison's(2005) findings show that environmental performance does notrespond to EMS implementation. Similarly, Boiral (2007) findsthat formal EMS fail to improve the environmental performance,



yet introduce many cumbersome and bureaucratic procedures.Rondinelli and Vastag (2000) argue that the implementation ofEMS cannot ensure that the company will attain environmentalsustainability.On the other hand, some studies find the EMS to be an importantdeterminant of environmental innovations (Hamschmidt andDyllick, 2006). Van Dijken et al. (1999) and Biondi et al. (2002)find that the implementation of EMS is an important determinantin the case of SMEs. Other studies pin point the importance ofdifferent characteristics of EMS that affect the quality of theirimplementation within the firm and in turn, the organisationalchanges that occur and subsequently affect environmentalperformance (Rehfeld et al., 2007; Anton et al., 2004; Arimuraet al., 2008). Renning et al. (2006) confirm the importance ofEMAS certification for environmental innovations among certi-fied facilities in Germany. In particular, their findings indicate therole played by different aspects of EMAS (i.e. maturity andstrategic importance of EMAS and the learning processes) forstimulating eco-innovation. They confirm that most features ofEMAS (as noted above) affect environmental process innovationsbut not product innovations because the implementation of EMSis aimed at improving the environmental quality of processes.Anton et al. (2004) also find that a more comprehensive imple-mentation of EMS (as opposed to a limited implementation)improves environmental performance. Finally, Wagner (2008)offers empirical evidence, which shows that EMS have a positiveeffect upon process innovations but no effect upon productinnovations.EMS are also expected tohave an indirect effect onorganisationalcapabilities by boosting environmental awareness and organisa-tional learning (Melnyk et al., 2003). Renning et al. (2006)highlight the significant role played by learning processes thatoccur during the implementation of EMS.

(b) Efficiency: We consider two dimensions of efficiency; (1) costsavings arising from expenditures in environmental improve-ments and (2) equipment upgrades undertaken with thepurpose of environmental protection.Eco-innovations that can lead to more efficient means ofproduction may give rise to cost savings that can in returnmotivate further investments into eco-innovations (Hitchenset al., 2003). Even though, many firms report environmentalprotection activities to be costly for them, Ashford (1994) arguesthat this is only the case for end-of-pipeline pollution controltechnologies and that, companies can save costs by investing incleaner production technologies that reduce pollution and wasteat its source. “When cleaner production and pollution controloptions that solve the same environmental problems are properlyevaluated against one another, the cleaner production optionswill usually be less costly to implement, operate and maintain…

because cleaner production technologies reduce cost of rawmaterials, energy, pollution control, waste treatment and clean-up, and regulatory compliance” (Ashford, 1994, p. 7). In line withthis statement, a recent OECD study using data from seven OECDcountries indicates that there are larger cost savings from invest-ments in cleaner production technologies (Johnstone, 2007).Frondel et al. (2007) provide evidence that cost savings are animportant factor that drives cleaner production technologies.On the other hand, Palmer et al. (1995) argue that cost savingsdue to environmental innovations are as low as 2% of the envi-ronmental compliance costs and are unlikely to provide enoughstimuli to drive environmental innovations. They suggest thatcost-offsets due to environmental innovations should be ratherhigh in order for cost-savings to be a driver for eco-innovations.Equipment upgrade activities can similarly lead to more efficientenergy use and further cost savings. For example, YokogawaElectric, a Japanesemanufacturer, developed a technology, which

controls the pumping pressure of air conditioning systemsresulting in large energy savings. Several businesses such asequipment factories, hotels, and supermarkets that upgradedtheir existing air conditioning systems toYokogawa's newsystemwere able to reduce their energy consumption significantly(OECD, 2009).

(c) Corporate Image: Corporate Social Responsibility (CSR) which,we consider to be an important element of corporate image, is arecent and controversial concept that embraces environmentalissues as one of its three pillars— the other two being theemployment/labour practices, and human/social rights. TheEarth Summit in Rio (1992) highlighted the importance ofenvironmental issues for CSR and coined the term ‘Environ-mental Social Responsibility’ (Hart, 2004; Russo and Fouts,1997).The effectiveness of environmental actions motivated by CSRhas come under question due to the ‘voluntary’ basis forcompliance under CSR. In other words, are ‘voluntary’processes such as CSR effective enough to stimulate eco-innovation? Aggeri (1999) argues that although voluntaryagreements provide weaker incentives compared to externalpolicy instruments, they are well adapted to manage uncer-tainty and coordination issues arising when dealing withsustainable development problems. In particular, innovation-oriented voluntary agreements include a stronger coordinationscheme, which is necessary in order to achieve sustainabledevelopment.On the other hand, Williamson et al. (2006) suggest that thevoluntary nature of CSR cannot satisfy the need for sustainableproduction. They argue that the use of regulatory structuresthat provide the minimum standards for many activitiescovered by CSR is the only effective way to encourage greenactivities in companies. Recent empirical evidence on theinfluence of voluntary programmes compared to formalregulation on emissions in the metal-finishing industry in-dicates that the effectiveness of voluntary programmes yieldedlittle, if any, reductions in emissions, while the regulator threatreduced emissions significantly (Brouhle et al., 2009). Hence,the impact and effectiveness of environmental CSR policies oneco-innovation remains unresolved begging for further re-search.Overall, the review of the literature suggests that it is highlyunlikely that all types of eco-innovations are stimulated by thesame drivers. Our conceptual framework in Fig. 2 attempts toexplore the role played by various factors in motivating thedifferent types of eco-innovation by disentangling the effec-tiveness of external and internal factors. This framework isimportant to ecological economics for analytical purposes inguiding empirical research and also for informing policymakers and businesses regarding the expected environmental,technological, and economic impact of various external policyinstruments and internal business activities. Given the abun-dance of mixed evidence on the role that these internal andexternal factors play for different types of eco-innovations, weapply an empirically oriented strategy and leave it for the datato reveal the direction and significance of the impact of thesefactors in the specific case of the UK.

3. Data and Methodology

3.1. Data and Descriptive Statistics

We use a dataset of 289 UK firms that responded to the ‘DEFRAGovernment Survey of Environmental Protection Expenditure byIndustry’ in years 2005 and 2006. The Department for Environment,Food and Rural Affairs (DEFRA) conducts the survey with the aim of

Source: Authors.

ECO-R&D

ECO-INNOVATIONS

End-of-Pipeline Solutions

Integrated Solutions

EXTERNAL POLICY INSTRUMENTS

-Command & control

-Market-based

INTERNAL FIRM-LEVEL MOTIVATIONS

-Organisational Capabilities

-Efficiency

-CSR strategy

Higher Impact

Lower Impact

Fig. 2. Conceptual framework.


estimating howmuch the UKmanufacturing sector spends (annually)to protect the environment.5 The survey represents a unique source ofinformation on the UK environmental spending with a very high levelof coverage across all manufacturing industries.6 With the exceptionof Kesidou and Demirel (2010), the data has not been used foracademic research purposes.

The survey questions give us a valuable opportunity to explore thedeterminants of different eco-innovations by providing informationon firms' investments into (1) End-of-pipeline pollution controltechnologies (EOP), (2) Integrated cleaner production technologies(INT) and (3) Environmental R&D (ECORD). As the statistics in Table 1reveal, the majority of UK investments into environmental protectiongoes into the integrated cleaner production technologies, followed byend-of-pipeline pollution control technologies and environmentalR&D. Frondel et al. (2007) and Lanoie et al. (2007) report similarlevels of investment for 7 OECD countries based on the OECD survey of3100 establishments.

Additionally, the DEFRA survey provides information on firms'Cost Savings from environmental activities (CS), their EquipmentUpgrade (Eq_Upgrade), and Corporate Social Responsibility (CSR)motivations, whether or not they subscribe to EnvironmentalManagement Systems (EMS), the external validation status of theirEMS (i.e. ISO14001) as well as how they perceive EnvironmentalRegulations (ENV_REG) and Environmental Taxes (ENV_TAX). Table 1reports a brief definition of all variables used in this study and theassociated descriptive statistics.

3.2. Methodology

Among the 289 firms in the dataset, only 70 firms invested in end-of-pipeline technologies (EOP), 66 invested in integrated cleanertechnologies (INT) and 73 invested in environmental R&D (ECORD) in

5 A stratified random sample of companies (based on SIC code and firm size) fromthe Inter Departmental Business Register (IDBR) held by the Office for NationalStatistics (ONS) was targeted by the survey in years 2005 and 2006. Note that firmswith less than 10 employees were not covered in the sampling process.

6 In particular, the response rates are respectively 18.7% and 20.4% in 2005 and2006. 1466 firms responded to the survey in 2005 and 1599 firms responded to thesurvey in 2006. 289 firms that are included in the following analysis responded in bothyears.

2006. A further breakdown of investment into each eco-innovationvariable is presented in Fig. 3. As indicated, 54.8% of firms in thedataset did not invest into any type of eco-innovation while 45.2% offirms invested in at least one type of eco-innovation. The share of eco-innovators in this study (45.2%) is strikingly similar to that of genericinnovators (45%) reported in the UK Community Innovation Survey(Community Innovation Survey, 2010).

We examine the determinants of different types of eco-innovationby using an econometric model where each of the three eco-innovation variables EOP, INT and ECORD (normalised by the totalcapital of the firm) in year 2006 are independently regressed on a setof lagged internal and external determinants from year 2005 asdiscussed in detail in Section 2.2.7 Since the dependent eco-innovation variables (ECOINN) are censored from below at zero (i.e.not all firms invest into eco-innovation), the appropriate estimationtechnique is a Tobit model8 (Greene, 2003):

ECOINN�i;t = x′i;t−1β + ε ð1Þ

And ECOINNi;t =ECOINN�

i;t if ECOINN�i;t N 0:

0 if ECOINN�i;t ≤ 0:

(ð2Þ

where ECOINN variable is greater than zero only when the latentECOINN* variable exceeds zero which represents cases where firms'willingness to invest into the eco-innovation activities is positive(Sigelman and Zeng, 1999). The independent variables, included inthe vector x correspond to the external and internal determinants ofeco-innovation as discussed below.

3.2.1. External Policy Determinants of Eco-innovationsPrior work by environmental economists has shown that environ-

mental regulations and environmental taxes are important determi-nants of eco-innovations (Hart, 2004; Popp, 2005; Rhothfels, 2002).

7 Our rational for using lagged independent variables mainly rests on the fact thatthere would be an expected lag between implementation of CSR, EMS, environmentalregulations, taxes etc. and their impact on eco-innovations. While using laggedobservations results in a loss of data, we consider this to be an appropriate sacrificegiven the yearly variation in the data and potential lag effects.

8 The Tobit model is especially helpful in this case because it inherently controls forselection bias (Wooldridge, 2002).

http://www.bis.gov.uk/assets/biscore/science/docs/u/10-p107a-uk-innovation-survey-2009-science-and-innovation-analysis.pdf

Table 1Description of variables and descriptive statistics.

Variables 2005 2006

Mean St. Dev Min Max Mean St. Dev Min Max

Eco-innovation variablesEOP: End-of-Pipeline Pollution Control Technologies (£) 201,333 1,790,586 0 2.41e+07 85,879 634,398 0 1.00e+07INT: Integrated Cleaner Production Technologies (£) 410,167 3,652,364 0 5.83e+07 753,337 1.01e+07 0 1.72e+08ECORD: Environmental research and development (£). 11,698 55,332 0 530,910 33,873 333,699 0 5,448,494

External determinantsENV_REG:=1 if the firm invested in environmentalprotection due to environmental regulation compliance.

0.238 0.427 0 1 0.262 0.440 0 1

ENV_TAX: =1 if the firm invested in environmentalprotection because of environmental taxes.

0.041 0.198 0 1 0.044 0.206 0 1

Internal determinantsCS: Total cost savings resulting from environmentalimprovements (£)

31,056 118,859 0 1,244,716 58,363 286,969 0 3,761,611

Equ_Upgrade:=1 if the firm invested in environmentalprotection because of equipment upgrade.

0.194 0.396 0 1 0.153 0.360 0 1

EMS: =1 if the firm has implemented environmentalmanagement systems.

0.384 0.487 0 1 0.446 0.498 0 1

ISO14001:=1 if the firm has and ISO14001 certifiedenvironmental management system.

0.296 0.457 0 1 0.282 0.451 0 1

CSR: =1 if the firm invested in environmental protectionbecause of parent company or owner policy/CSR.

0.075 0.264 0 1 0.061 0.240 0 1

EMP: Number of employees 425 1009 0 12,287 434 1065 0 13,427TURNOVER (£) 2.06E+08 1.05E+09 0 1.10E+10 2.10e+08 1.08e+09 0 1.28e+10PRODUCTIVITY 915728.2 1.03E+07 0 1.73E+08 403607.9 2,264,999 12363.64 3.49E+07CAP: Total Capital (£) 1.12E+07 6.82E+07 684.334 9.27E+08 1.46E+07 8.90e+07 1.395973 1.05E+09


We use two proxies for the external determinants: (1) ENV_REG and(2) ENV_TAX. Both are binary variables that respectively assume thevalue 1 if a firm indicates that environmental regulations orenvironmental taxes have been effective in their decisions to investinto environmental protection in 2005.

3.2.2. Internal Determinants of Eco-innovationsPrevious empirical studies on innovation economics and environ-

mental management have underlined the internal determinants ofeco-innovation (Renning et al., 2006; Frondel et al., 2007; Kollmanand Prakash, 2002; Darnall, 2006). Among these firm level factors, we

EOP INT

ECORD

EOP Only

8.8%

INT Only

9.2%

ECORD Only

8.2%

EOP&INT

4.8%

EOP&INT&ECORD

7.1%

EOP&ECORD

4.4%

INT&ECORD

2.7%

No Environmental Protection 54.8%

Fig. 3. Breakdown of firms according to eco-innovation activities.

measure material and energy efficiency gains by considering the CostSavings due to environmental improvements (CS) and by taking intoaccount the Equipment Upgrade motivations (Equ_Upgrade) of firmsas reported in 2005.

The environmental organisational capabilities of firms are mea-sured by a binary variable that takes the value 1 if the firm has anEnvironmental Management Systems (EMS) in place in 2005. Wefurther investigate whether the external certification of EMS is animportant determinant of eco-innovation using another dummyvariable that takes the value 1 if the firm owns an approvedISO14001 certification. Finally, we explore the role of voluntaryactivities of companies to protect the environment by including anindependent dummy variable, CSR that indicates whether corporatesocial responsibility policies have played an important role in thedecision to invest into environmental protection in 2005.

As an additional control, we account for the size of the firms —

measured by the number of employees. We also control for labourproductivity – measured by an output/labour ratio – since betterperforming companies are more likely to pursue environmental goals(Nakamura et al., 2001). We also include dummies for LOW-emission,MEDIUM-emission and HIGH-emission industries.9 Table 2 presentsthe emission intensity for each sector while Table 3 is a full correlationmatrix for the dependent and independent variables used in theeconometric model.

9 The set of industries covered in the sample correspond to the UK SIC 2003 codes10, 11, 14, 15, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36,40 and 41. We have classified these industries in three groups with respect to theirpollution intensity: LOW, MEDIUM and HIGH Emission. This classification is based onDEFRA's greenhouse gas (GHG) emissions data for 93 UK Economic Sectors between1990 and 2008. We have taken an average for the GHG emissions for each sector in theUK across the years and ranked these values in ascending order. In what follows, theindustries classified as LOW emission are those that are at the lowest quartile of the UKpollution intensity, those classified as MEDIUM emission are those that are within the2nd and 3rd quartiles of the range and finally, the industries classified as HIGHemission are those in the top quartile of the UK pollution intensity.

Table 2Classification of Industries with respect to pollution intensity.

SIC Pollutionintensity

40 Electricity, Gas, Steam and Hot Water Supply HIGH27 Manufacture of Basic Metals HIGH11 Extraction of Crude Petroleum and Natural Gas; Service

Activities Incidental to Oil and Gas ExtractionHIGH

41 Collection, Purification and Distribution of Water HIGH23 Manufacture of Coke, Refined Petroleum Products and Nuclear

FuelHIGH

24 Manufacture of Chemicals and Chemical Products HIGH26 Manufacture of Other Non-Metallic and Mineral Products HIGH15 Manufacture of Food Products and Beverages HIGH10 Mining of Coal and Lignite; Extraction of Peat HIGH21 Manufacture of Pulp, Paper and Paper Products MEDIUM25 Manufacture of Rubber and Plastic Products MEDIUM20 Manufacture ofWood And Products of Wood And Cork, Except

Furniture; Manufacture of Articles of Straw and PlaitingMaterials

MEDIUM

28 Manufacture of Fabricated Metal Products, Except Machineryand Equipment

MEDIUM

17 Manufacture of Textiles MEDIUM36 Manufacture of Other Products MEDIUM29 Manufacture of Machinery and Equipment MEDIUM34 Manufacture of Motor Vehicles, Trailers and Semi-trailers MEDIUM22 Publishing and Printing MEDIUM14 Other Mining and Quarrying MEDIUM35 Manufacture of Other Transport Equipment MEDIUM31 Manufacture of Electrical Machinery and Apparatus MEDIUM33 Manufacture of Medical, Precision and Optical Instruments,

Watches and ClocksMEDIUM

32 Manufacture of Radio, Television and CommunicationEquipment and Apparatus

LOW

18 Manufacture of Wearing Apparel; Dressing and Dyeing of Fur LOW30 Manufacture of Office Machinery and Computers LOW19 Tanning and Dressing of Leather; Manufacture of Handbags,

Saddlery, Harness And FootwearLOW

16 Manufacture of Tobacco Products LOW


3.2.2.1. Decomposing the Marginal Effects. To have a more compre-hensive understanding of the impact of these internal and externalfactors upon eco-innovations, we decompose the marginal effectsattained through the estimation of the Tobit model in Eq. (1) usingMcDonald and Moffitt's (1980) suggestion of decomposing the slopevector β into:

∂E ECOINNi jxi½ �∂xi

= Prob ECOINN N 0∂E ECOINNi jxi; ECOINNi≻0½ �

∂xi

+ E ECOINNi jxi; ECOINNi≻0½ � Prob ECOINN N 0½ �∂xi

ð3Þ

such that a change in xi has two effects: (1) the effect on the probabilitythat the observation will fall in the positive part of the distribution(the impact upon convincing firms to undertake investments into eco-

Table 3Correlation matrix.

1 2 3 4 5

1. EOP 1.00002. INT 0.1383* 1.00003. ECORD 0.2017* 0.4564* 1.00004. Ln(emp)(t−1) 0.0702 −0.0526 −0.0757 1.00005. Prod (t−1) 0.0247 −0.0210 0.0635 −0.0518 1.00006. Regulations (t−1) 0.0719 −0.0077 0.0611 0.1203 −0.03707. Taxes (t−1) 0.0477 0.0531 0.0954 0.0302 −0.01488. ln(CS)(t−1) 0.1717* 0.1353 0.1087 −0.0725 −0.01279. Equipment Upg (t−1) 0.1319 0.2449* −0.0312 0.0264 −0.036810. EMS (t−1) 0.1536* 0.0433 −0.0161 0.3951* 0.066511. ISO14001(t−1) 0.1317 0.0840 −0.0017 0.4028* 0.085312. CSR(t−1) 0.0928 0.0573 −0.0279 0.0869 −0.0199

innovations) and (2) the effect on the conditional mean of ECOINN* inthepositive part of the distribution (i.e. the impact upon the level of eco-innovations for those firms with positive spending) (Greene, 2003,p. 766). This decomposition is especially important for understandingthe factors that can motivate 54.8% of the firms that do not undertakeany kind of eco-innovations. The decomposed marginal effects as inpart (1) and (2) in Eq. (3) are reported in all regressions across Tables 4,5 and 6.

4. Results and Discussion

As initially anticipated, the results provided in Tables 4, 5 and 6confirm that different types of eco-innovation indeed have differentdeterminants. Moreover, these determinants have a different impacton the level of eco-innovations and the probability of eco-innovating.These findings have important implications for policy makers whomay wish to stimulate a certain kind of eco-innovation by focusing onits specific determinants. Below we discuss the role of internal andexternal determinants for motivating EOP, INT and ECORD.

4.1. Impact of External Factors

With regard to the external policy instruments, our results indicatethat environmental regulations are effective in driving certain types ofeco-innovation, while environmental taxes fail to motivate any of thethree eco-innovations considered.

4.1.1. Environmental RegulationsBeyond the findings of the existing literature, incorporating all

three types of eco-innovation in the same framework allows us to seethe presence of a U-type relationship between regulations and eco-innovations: regulations are able to impact the least and most sig-nificant eco-innovations; namely end-of-pipeline technologies (EOP)and environmental R&D (ECORD). Their impact is less than clear for theintegrated cleaner technologies (INT), which stand in between EOP andECORD in terms of environmental and technological significance.

Moreover, decomposing themarginal effects of regulations indicatesthat their impact is especially large in getting companies to invest intoEOP and ECORD while they have a smaller, yet significant impact uponthe levels of EOP and ECORD undertaken by eco-innovators. This sug-gests that regulations are being effective in converting firms to eco-innovators either at the lower or the higher end of the eco-innovationspectrum.

These findings are in line with those of Frondel et al. (2007) whoshow that regulations are important for the end-of-pipeline technol-ogies, but not for integrated cleaner technologies. According toFrondel et al. (2007, p. 573), because regulations are very prescriptive,and usually ‘impose technology standards that can only be metthrough End-of-Pipeline abatement measures’, they stimulate in-vestments in process innovations; specifically in end-of-pipelinetechnologies, while they have no impact upon integrated cleaner

6 7 8 9 10 11 12

1.00000.0461 1.00000.0390 0.2376* 1.0000

−0.1731* 0.0293 0.1100 1.00000.2642* 0.0137 0.0985 0.0016 1.00000.2149* −0.0208 0.0695 0.0025 0.8205* 1.0000

−0.0679 0.0067 | −0.0016 −0.0087 0.1473 0.1555* 1.0000

Table 4Determinants of End of Pipeline (EOP) Pollution Control Technologies.

Dependent variable EOP Marginal effectsfor the decisionto conduct EOP

Marginaleffects forEOPN0

EOP Marginal effectsfor the decisionto conduct EOP


EOP Marginal effectsfor the decisionto conduct EOP


In (Employees)(t−1) 0.042** 0.043** 0.009*** .044*** .044*** .010*** .055*** .055*** .012***0.017 0.012 0.004 .017 .012 .004 .019 .012 .004

Productivity (t−1) .000*** .000*** .000*** .000*** .000*** .000*** .000*** .000*** .000***.000 .000 .000 .000 .000 .000 .000 .000 .000

External determinantsENV_REG (t−1) .133*** .134*** .029*** .141*** .143*** .031*** .174*** .176*** .038***

.043 .042 .009 .046 .045 .010 .048 .039 .011ENV_TAX(t−1) − .130 − .130 − .029 − .127 − .127 − .028 − .180* − .181** − .040*

.097 .092 .021 .092 .088 .020 .104 .091 .023

Internal determinantsln (CS)(t−1) .174 .175 .038 .180 .181 .040 .207* .209* .046*

.122 .122 .027 .118 .119 .026 .110 .109 .025Equ_upgrade(t−1) .156** .157*** .035** .156** .157*** .034** .160** .162*** .035**

.061 .048 .014 .061 .048 .013 .063 .049 .014EMS(t−1) .115** .116*** .026**

.050 .038 .011ISO14001(t−1) .103** .103** .023**

.052 .041 .011CSR(t−1) .147 .148* .032

.091 .077 .020MEDIUM-Pollution .143 .143 .032 .151 .153 .034 .163 .165 .036

.133 .126 .029 .134 .127 .029 .139 .130 .030HIGH-Pollution .173 .174 .038 .182 .183 .040 .180 .182 .040

.134 .126 .029 .136 .127 .030 .138 .130 .030Constant − .702*** − .705*** −.588***

.228 .228 .157F 2.23** 2.20** 2.18**Log likelihood −64.053 − 67.839 −65.142Pseudo R2 0.306 0.298 0.294Left-censored observations 73 73 73Uncensored observations 216 216 216Number of obs 289 289 289

Significance *0.1, **0.5, ***0.01. Robust standard errors in italics.


technologies. Hence, our findings confirm the existing literature in thecontext of the UK.

Simultaneously, our results show that regulations stimulate in-vestments in ECORD, which gives rise to both process and productenvironmental innovations. This finding is in-line with the ‘PorterHypothesis’, which states that regulations boost environmental R&Dactivities and thus, stimulate eco-innovation.According to the literature,companies which invest in environmental R&D gain ‘strategic advan-tage’ from innovation and become leaders in ‘green markets’ (Carraro,2000; Montero, 2002; Hart, 2004; Popp, 2005; Rhothfels, 2002).

4.1.2. Environmental TaxesIn contrast to regulations, environmental taxesdonot appear to have

a significant impact on any of the three types of eco-innovation. Thisfinding is, again, similar to Frondel et al.'s (2007) results which confirmthat market-based instruments do not affect either end-of-pipeline orintegrated technologies in Germany. In the specific case of the UK,environmental taxes have not been frequently used as a means ofregulating pollution levels since environmental laws have historicallybeen the preferred policy tool in this field (Ashford, 1994; Jordan andWurzel, 2003). Moreover, environmental taxes are commonly set at alow level and the innovation effects are, therefore, low or insignificant(Kemp, 1997).

4.2. Impact of Internal Factors

As far as internal factors are concerned, our results indicate thatefficiency (equipment upgrade motives and cost savings) and EMSfactors have a varied impact on the three types of eco-innovationswhile CSR is not a significant driver for any type of eco-innovation.

4.2.1. EfficiencyMachinery and equipment upgrades are important means of

increasing efficiency for companies and our results indicate that EOPand INT are driven by firms' willingness to upgrade their equipment.This suggests that firms consider the most energy efficient and envi-ronmentally friendly technologies when they are renewing existingfacilities.

Another indicator of efficiency, cost savings, appear to be animportant driver for only the most advanced type of eco-innovations,ECORD, while it has no significant impact on either EOP or INT. Thisresult is understandable for the case of end-of pipeline technologieswhich are considered to be costly rather than cost-saving investments(Ashford, 1994). The lack of impact in the case of integrated cleanertechnologies, on the other hand, gives support to the findings ofPalmer et al. (1995) who suggest that cost savings might not be largeenough to drive eco-innovations.

Finally, the results suggest that environmental R&D is not onlystimulated by regulation but it is also market driven, mainlymotivated by the cost saving potential of the outcomes that arisefrom environmental R&D. The decomposition of marginal effects onthe cost savings variable confirms thatmany firms aremotivated to doenvironmental R&D due to the cost saving possibilities while theimpact of cost savings are smaller for those firms that are alreadyinvesting into environmental R&D.

4.2.2. Environmental Management Systems (EMS)The impact of EMS, on the other hand, is similar to that of envi-

ronmental regulations where the most (ECORD) and least (EOP)significant eco-innovations respond to the presence of an EMS in thecompany. This effect is most clearly visible especially when the impact

Table 5Determinants of Integrated (INT) Cleaner Production Technologies.

Dependent variable INT Marginal effectsfor the decisionto conduct INT

Marginaleffects forINTN0

INT Marginal effectsfor the decisionto conduct INT


INT Marginal effectsfor the decisionto conduct INT


In (Employees)(t−1) .043* .027* .010* .039* .025* .009* .054** .034** .012**.023 .014 .005 .023 .015 .005 .023 .014 .005

Productivity (t−1) .000 .000 .000 .000 .000 .000 .000 .000 .000.000 .000 .000 .000 .000 .000 .000 .000 .000

External determinantsENV_REG(t−1) .093 .059 .021 .096 .061 .022 .133* .084* .030*

.071 .046 .016 .070 .045 .016 .069 .044 .016ENV_TAX(t−1) .153 .096 .035 .160 .102 .037 .121 .076 .028

.169 .107 .098 .165 .106 .038 .167 .106 .038

Internal determinantsln (CS)(t−1) .213 0.134 .049 .214 .136 .049 .241* .152* .055*

.137 .085 .031 .139 .119 .032 .131 .080 .030Equ_upgrade(t−1) .293*** .185*** .067*** .291*** .186*** .067*** .297*** .187*** .068***

.099 .054 .023 .098 .054 .023 .101 .055 .024EMS(t−1) .111 .070 .025

.075 .045 .017ISO14001(t−1) .136 .087* .031*

.079 .048 .018CSR(t−1) .189 .119* .043

.115 .071 .026MEDIUM-Pollution − .042 − .026 − .010 − .037 − .024 − .008 − .015 − .009 − .003

.177 .112 .041 .177 .112 .040 .180 .113 .041HIGH-Pollution .019 .012 .004 .027 .017 .006 .026 .017 .006

.171 .108 .040 .170 .109 .039 .173 .109 .040Constant − .704*** − .704*** − .753***

.144 .141 .219F 2.70*** 2.68*** 2.78***Log likelihood −119.360 −118.91 −119.076Pseudo R2 0.104 0.107 0.106Left-censored observations 70 70 70Uncensored observations 219 219 219Number of obs 289 289 289



of ISO14001 certification is considered. EOP and ECORD respondpositively to adopting ISO14001 certification while INT is notstimulated by either maintaining an EMS or subscribing toISO14001. A plausible explanation of this finding is related to theinnovative heterogeneity of firms where the least innovative firmsbenefit from having an organisational environmental structure tosupport them with the minimum compliance requirements throughEOP while the most innovative firms use EMS as an innovationplatform to build upon for ECORD. The decomposed marginal effectssuggest that EMS is especially effective in motivating firms to startinvesting in EOP and has a significant but smaller effect on increasingthe EOP investments of those firms that already invest in EOP. In thecase of ECORD, EMS is only effective for persuading firms to invest inECORD but cannot motivate increased ECORD investments for firmswith existing ECORD activities.

4.2.3. Corporate Social Responsibility (CSR)Finally, CSR fails to be a significant driver of any of the three eco-

innovations. This indeed poses questions on howmuchwe can rely onthe corporate goodwill and voluntary compliance in environmentalmatters. While environmental awareness and protection is animportant foundation for CSR, the costly nature of environmentalprotection and the externalities associated with these expendituresappear to get in the way of CSR as a powerful driver for environmentalprotection.

Finally, we find firm size to be a significantly important driver inthe case of EOP while its impact is not significant in the case of INT orECORD. Labour productivity, on the other hand, has a small butsignificant impact in driving EOP and ECORD investments. We do not

find any significant industry effects measured by the industrydummies constructed based on pollution intensity.

5. Conclusions

This paper looks into the determinants of different types of eco-innovation, namely, end-of-pipeline pollution control technologies,integrated cleaner production technologies and environmental R&D.These three types of eco-innovation differ with respect to theirtechnological significance, costs to companies and their benefits to theenvironment. By integrating these three different eco-innovationactivities in a single framework, we are able to analyze whether andhow certain factors stimulate each of these eco-innovations.

Our findings indicate that environmental regulations affect end-of-pipeline pollution control technologies and environmental R&D whilethey do not influence integrated cleaner technologies. By setting stricttechnology standards, regulations stimulate investments in end-of-pipeline technologies, which have the lowest environmental andtechnological impact while, at the same time, encourage investmentsin environmental R&D, which has the highest environmental andtechnological impact. Consequently, the results of this study suggestthat regulations can play an important role in combating pollution notonly in the short run, by stimulating investments in processinnovations such as end-of-pipeline technologies, but also in thelong run by driving investments in process and product innovationsthrough environmental R&D. The latter provides support for PorterHypothesis and has important implications for the environmentalinnovation policy. As markets for low-carbon products are estimatedto be worth at least $500bn per year by 2050 (Stern Review, 2006),regulation may play a key role in motivating firms to invest in

Table 6Determinants of Environmental Research and Development (ECORD).

Dependent variable ECORD Marginal effectsfor the decisionto conduct ECORD

Marginaleffects forECORDN0

ECORD Marginal effectsfor the decisionto conduct ECORD


ECORD Marginal effectsfor the decisionto conduct ECORD


In (Employees)(t−1) .016 .015 .003 .013 .012 .003 .025* .023* .005*.014 .013 .003 .014 .014 .003 .013 .012 .003

Productivity (t−1) .000*** .000*** .000*** .000*** .000*** .000*** .000*** .000*** .000***.000 .000 .000 .000 .000 .000 .000 .000 .000

External determinantsENV_REG(t−1) .108** .101** .023** .110** .103*** .024** .129** .123*** .029**

.048 .040 .010 .048 .040 .010 .051 .039 .011ENV_TAX(t−1) .107 .100 .023 .113 .106 .024 .087 .083 .019

.106 .099 .010 .105 .097 .023 .102 .097 .022

Internal determinantsln (CS)(t−1) .151*** .142*** .032*** .148*** .138*** .032*** .171*** .162*** .037***

.054 .051 .011 .053 .050 .011 .050 .045 .011Equ_upgrade(t−1) − .005 − .005 − .001 − .005 − .004 − .001 − .005 − .005 − .001

.049 .046 .011 .049 .045 .010 .049 .046 .011EMS(t−1) .081* .076** .017*

.049 .038 .010ISO14001(t−1) .102* .095** .022*

.054 .039 .011CSR(t−1) .046 .043 .010

.062 .059 .013MEDIUM-Pollution − .145 − .136 − .031 − .143 − .133 − .031 − .131 − .125 − .028

.149 .100 .030 .139 .099 .030 .136 .101 .029HIGH-Pollution − .114 − .106 − .024 − .110 − .102 − .024 − .106 − .101 − .022

.131 .100 .028 .130 .099 .028 .128 .100 .028Constant − .284*** − .274*** − .311***

.088 .088 .086F 2.24** 2.35** 2.74***Log likelihood −80.47 −79.78 −81.59Pseudo R2 0.119 0.126 0.106Left-censored observations 66 66 66Uncensored observations 223 223 223Number of obs 289 289 289



environmental R&D for the generation of eco-friendly products andservices that have significant market potential.

Our results further suggest that the impact of regulations isespecially large in motivating companies to adopt eco-innovationswhile they have a smaller, yet significant impact, upon the intensity ofinvestments on eco-innovations. These results are in-line with theliterature on the diffusion of generic innovations. In particular, at earlystages of diffusion the inter-firm adoption rate dominates the intra-firm intensity of use (Battisti and Stoneman, 2003). Policies tend tofocus mainly on the adoption of innovations by firms rather than ontheir intensity of use. Yet, once beyond the early stages of adoption,the intensity of use becomes extremely important in the generationof benefits from adoption, in general, and of eco-innovations inparticular (e.g. in increasing learning and the absorptive capacity ofthe company) (Battisti and Stoneman, 2003). Therefore, we highlightthe necessity for designing environmental regulations that do notonlymotivate the adoption of eco-innovations but also stimulate theirintensity of use.

In contrast to environmental regulations, environmental taxes donot affect any type of eco-innovations. UK is currently adopting agrowing number of market-based instruments in lieu of the EU. Policymakers should carefully consider the effectiveness of these instrumentsnot only for reducing the immediate pollution but also for stimulatingeco-innovations that will lead to a greener economy in the UK. Kempand Pontoglio (2008) highlight the need for policy instruments thatensure the required policy stringency through appropriate design andeffective enforcements. Additionally, our results suggest that market-based instruments cannot be solely relied on and they should becombined with the necessary regulations.

Our findings also point to the presence of an important number offirm-based factors that determine the adoption and/or creation ofeco-innovations. Firstly, cost savings resulting from environmentalefficiency appear to be only significant in driving investments inenvironmental R&D which are the most advanced in the spectrum ofeco-innovations. Cost savings come from eliminating or re-usingwaste. Less advanced eco-innovations, expectedly, have a lowerpotential for creating such savings for companies. Therefore, it ispossible that cost savings are most closely associated with the mostadvanced eco-innovations. Improvements to the eco-infrastructureplay a significant role in ensuring that even firms with less advancedeco-innovations can reap cost benefits. For example, ensuring thatfirms have accessible outlets for the sale of their by-products can behighly effective in encouraging cost savings and increasing eco-innovations.

The least and medium impact eco-innovations (i.e. EOP and INT),on the other hand, are motivated by increased environmentalefficiency due to equipment upgrades, which allow firms to shift tomore energy efficient means of production. We note the benefits ofpolicy incentives, such as ‘scrappage’ style programs, that will make iteasier for firms to trade-in their machinery for more energy efficientalternatives.

Second of firm-based factors, environmental organisational capa-bilities, also impact the least and most advanced eco-innovations,possibly providing the new starters with the required organisationalstructure on their first attempt to eco-innovate and also the mostcapable innovators with a sound framework to enable them to applytheir innovation skills and knowledge to environmental matters. Wefind that ISO14001 certification is effective in strengthening the


positive impact of environmental management systems on both end-of-pipeline technologies and environmental R&D.

The third firm based factor, CSR, is interestingly not a significantdriver of any type of eco-innovation. This casts doubt on the effec-tiveness of voluntary agreements by companies to reduce their envi-ronmental impact. CSR policies can most effectively be used as asupporting mechanism to environmental policies which underline theminimum basis for environmental compliance.

Acknowledgements

The authors would like to thank Rocky Harris and SteveWellingtonfrom DEFRA for providing us access to the dataset and clarifications onthe data collection methods. We are grateful to Giuliana Battisti, PeterSwann, Sourafel Girma and two anonymous referees for usefulcomments and suggestions on earlier drafts. We also thank EskandarRashid Mohamed for excellent research assistance.

References

Aggeri, F., 1999. Environmental policies and innovation: a knowledge-based perspec-tive on cooperative approaches. Research Policy 28 (7), 699–717.

Anton, W.R.Q., Deltas, G., Khanna, M., 2004. Incentives for environmental self-regulation and implications for environmental performance. Journal of Environ-mental Economics and Management 48 (1), 632–654.

Arimura, T.H., Hibiki, A., Katayama, H., 2008. Is a voluntary approach an effectiveenvironmental policy instrument? A case for environmental management systems.Journal of Environmental Economics and Management 55 (3), 281–295.

Ashford, N.A., 1994. Government strategies and policies for cleaner production. UnitedNations Environment Programme: Industry and Environment.

Bansal, P., Hunter, T., 2003. Strategic explanations for the early adoption of ISO 14001.Journal of Business Ethics 46 (3), 289–299.

Battisti, G., Stoneman, P., 2003. Inter- and intra-firm effects in the diffusion of newprocess technology. Research Policy 32 (9), 1641–1655.

Baumol, W.J., Oates, W.E., 1988. The Theory of Environmental Policy, Second edition.Cambridge University Press, Cambridge, UK.

Biondi, V., Iraldo, F., Meredith, S., 2002. Achieving sustainability through environmentalinnovation: the role of SMEs. International Journal of Technology Management 24(5–6), 612–626.

Boiral, O., 2007. Corporate greening through ISO 14001: a rational myth? OrganizationScience 18 (1), 127–146.

Brouhle, K., Griffiths, C., Wolverton, A., 2009. Evaluating the role of EPA policy levers:an examination of a voluntary program and regulatory threat in the metal-finishing industry. Journal of Environmental Economics and Management 57 (2),166–181.

Carraro, C., 2000. Environmental technological innovation and diffusion: modelanalysis. In: Hemmelskamp, J., Leone, F., Rennings, K. (Eds.), Innovation-OrientedEnvironmental Regulations: Theoretical Approaches and Empirical Analysis.Physica, Heidelberg, New York, pp. 269–297.

Cleff, T., Rennings, K., 2000. Determinants of environmental innovation—empiricalevidence from theMannheim Innovation Panel and an additional telephone survey.In: Hemmelskamp, J., Leone, F., Rennings, K. (Eds.), Innovation-Oriented Environ-mental Regulations: Theoretical Approaches and Empirical Analysis. Physica,Heidelberg, New York, pp. 269–297.

Cohen, W., Levinthal, D., 1990. Absorptive capacity: a new perspective on learning andinnovation. Administrative Science Quarterly 35 (1), 128–152.

Community Innovation Survey UK, 2010. UK Innovation Survey 2009: Science andInnovation Analysis, available at: http://www.bis.gov.uk/assets/biscore/science/docs/u/10-p107a-uk-innovation-survey-2009-science-and-innovation-analysis.pdf.

Cooke, P., 2008. Cleantech and an analysis of the platform nature of life sciences: furtherreflections upon platform policies. European Planning Studies 16 (3), 375–393.

Darnall, N., 2006.Why firmsmandate ISO 14001 certification. Business & Society 45 (3),354–381.

DEFRA, 2006. Environmental expenditure by industry: 2006 UK survey, available at:http://archive.defra.gov.uk/evidence/statistics/environment/envsurvey/expn2007/index.htm.

Frondel, M., Horbach, J., Rennings, K., 2007. End-of-Pipe or cleaner production? Anempirical comparison of environmental innovation decisions across OECDcountries. Business Strategy and the Environment 16 (8), 571–584.

Geels, F.W., Schot, J., 2007. Typology of sociotechnical transition pathways. ResearchPolicy 36 (3), 399–417.

Greene, W.H., 2003. Econometric Analysis, 5th edn. Prentice Hall, New York.Hamschmidt, J., Dyllick, T., 2006. ISO14001:profitable? Yes! But is it ecoeffective? In:

Schaltegger, S.; Wagner, M.: Managing the Business Case for Sustainability. Sheffield :Greenleaf Publishing.

Hart, R., 2004. Growth, environment and innovation—amodel with production vintagesand environmentally oriented research. Journal of Environmental Economics andManagement 48 (3), 1078–1098.

Hitchens, D., Trainor, M., Clausen, J., Thankappan, S., De Marchi, B., 2003. Small andMedium Sized Companies in Europe—Environmental Performance, Competitivenessand Management. Springer Verlag, Heidelberg.

Jaenicke, M., Blazejczak, J., Edler, D., Hemmelskamp, J., 2000. Environmental policy andinnovation—an international comparison of policy frameworks and innovationeffects. In: Hemmelskamp, J., Leone, F., Rennings, K. (Eds.), Innovation-orientedEnvironmental Regulations: Theoretical Approaches and Empirical Analysis.Physica, Heidelberg, New York, pp. 269–297.

Johnstone, N., 2007. Environmental Policy and Corporate Behaviour. Edward Elgar/OECD, Cheltenham, UK; Northampton, MA.

Jordan, A., Wurzel, R.K.W., Zito, A.R., 2003. Comparative conclusions — ‘new’ environ-mental policy instruments: an evolution or a revolution in environmental policy?Environmental Politics 12 (1), 201–224.

Kemp, R., 1997. Environmental Policy and Technical Change. Edward Elgar, Cheltenham,UK.

Kemp, R., Pontoglio, S., 2008. The Innovation Effects of Environmental PolicyInstruments—A Typical Case of the Blind Men and the Elephant, Paper for DIMEWP 2.5 Workshop on Empirical Analyses of Environmental Innovation. FraunhoferInstitute for Systems and Innovation Research (ISI), Karlsruhe. January 17–18.

Kesidou, E., Demirel, P., 2010. On the drivers of eco-innovations: empirical evidencefrom the UK. NUBS Research Paper Series No. 2010-03. Nottingham UniversityBusiness School, Nottingham, UK.

Kollman, K., Prakash, A., 2002. EMS-based environmental regimes as club goods. PolicySciences 35 (1), 43–67.

Lanoie, P., Laurent-Lucchetti, J., Johnstone, N., Ambeck, S., 2007. Environmental Policy,Innovation and Performance: New insights on the Porter Hypothesis, WorkingPaper GAEL; 2007-07.

McDonald, J.F., Moffitt, R.A., 1980. The uses of Tobit analysis. Review of Economics andStatistics 62 (2), 318–321.

Melnyk, S.A., Stroufe, R.P., Calantone, R., 2003. Assessing the impact of environmentalmanagement systems on corporate and environmental performance. Journal ofOperations Management 21 (3), 329–353.

Milliman, S.R., Prince, R., 1989. Firm incentives to promote technological change inpollution control. Journal of Environmental Economics and Management 17 (3),247–265.

Montero, J.-P., 2002. Permits, standards, and technology innovation. Journal ofEnvironmental Economics and Management 44 (1), 23–44.

Nakamura, M., Takahashi, T., Vertinsky, I., 2001. Why Japanese firms choose to certify: astudy of managerial responses to environmental issues. Journal of EnvironmentalEconomics and Management 42 (1), 23–52.

Nill, J., Kemp, R., 2009. Evolutionary approaches for sustainable innovation policies:from niche to paradigm? Research Policy 38 (4), 668–680.

OECD, 2009. Sustainable Manufacturing and Eco-Innovation: Framework, Practices andMeasurement— Synthesis Report. OECD, Paris.

Palmer, K., Oates, W.E., Portney, P.R., 1995. Tightening environmental standards: thebenefit–cost or the no-cost paradigm? The Journal of Economic Perspectives 9 (4),119–132.

Popp, D., 2005. Uncertain R&D and the Porter hypothesis. Contributions to EconomicAnalysis and Policy 4 (1), available at: bhttp://www.bepress.com/bejeap/contri-butions/vol4/iss1/art6/N.

Porter, M.E., 1991. America's green strategy. Scientific American 264 (4).Porter, M.E., van der Linde, C., 1995a. Toward a new conception of the environment-

competitiveness relationship. The Journal of Economic Perspectives 9 (4), 97–118.Porter, M.E., van der Linde, C., 1995b. Green and competitive: ending the stalemate.

Harvard Business Review 73 (5).Rammel, C., van den Bergh, J.C.J.M., 2003. Evolutionary policies for sustainable

development: adaptive flexibility and risk minimising. Ecological Economics 47(2–3), 121–133.

Rehfeld, K.-M., Rennings, K., Ziegler, A., 2007. Integrated product policy andenvironmental product innovations: an empirical analysis. Ecological Economics61 (1), 91–100.

Renning, K., Ziegler, A., Ankele, K., Hoffman, E., 2006. The influence of differentcharacteristics of the EU environmental management and auditing scheme ontechnical environmental innovations and economic performance. EcologicalEconomics 57 (1), 45–59.

Rennings, K., 2000. Redefining innovation — eco-innovation research and thecontribution from ecological economics. Ecological Economics 32 (2), 319–332.

Requate, T., 2005. Dynamic incentives by environmental policy instruments— a survey.Ecological Economics 54 (2–3), 175–195.

Requate, T., Unold, W., 2003. Environmental policy incentives to adopt advancedabatement technology: will the true ranking please stand up? European EconomicReview 47 (1), 125–146.

Rhothfels, J., 2002. Environmental policy under product differentiation and asymmetriccosts: does leapfrogging occur and is it worth it? In: Marsiliani, L., Rauscher, M.,Withagen, C. (Eds.), Environmental Economics and the International Economy.Kluwer Academic Publishers, Drodrecht.

Rondinelli, D., Vastag, G., 2000. Panacea, common sense, or just a label? The value of ISO14001 environmental management systems. European Management Journal 18 (5),499–510.

Russo, M.V., Fouts, P.A., 1997. A resource-based perspective on corporate environmentalperformance and profitability. Academy of Management Journal 40 (3), 534–559.

Russo, M.V., Harrison, N.S., 2005. Organizational design and environmental perfor-mance: clues from the electronics industry. Academy ofManagement Journal 48 (4),582–593.

Sigelman, L. and Zeng, L. 1999. Analysing censored and sample selected data with Tobitand Heckit models Political Analysis 8(2), 167–182.


Stern, N., 2006. Stern Review: The Economics of Climate Change. Office of ClimateChange, UK.

van den Bergh, C.J.M., Faber, A., Idenburg, A.M., Oosterhuis, F.H., 2007. EvolutionaryEconomics and Environmental Policy: The Survival of the Greenest. Edward Elgar,Cheltenham.

van Dijken, K., Prince, Y.,Wolters, T., Frey, M., Mussari, G., Kalff, P., Hansen, O., Kerndrup,S., Son-dergard, B., Rodrigues, E.L., Meredith, S., 1999. The Dynamics of Innovationas Interplay between Business Competence. Environmental Orientation andNetwork Involvement, Kluwer, Dordrecht.

Wagner, M., 2008. Empirical influence of environmental management on innovation:evidence from Europe. Ecological Economics 66 (2–3), 392–402.

Williamson, D., Lynch-Wood, G., Ramsay, J., 2006. Drivers of environmental behaviourin manufacturing SMEs and the implications for CSR. Journal of Business Ethics 67(3), 317–330.

Wooldridge, J.M., 2002. Econometric Analysis of Cross Sectional and Panel Data. TheMIT Press, London.

Documents

Stimulating different types of eco-innovation in the UK: Government policies and firm motivations