1

THE SHIFTING LANDSCAPE OF GLOBAL FOREIGN DIRECT

INVESTMENT: EMERGING PATTERNS

Kalman Kalotay 1

This presentation examines changes in global foreign direct investment (FDI) flows and stocks in

the past quarter of century (1991–2016) for countries and territories for which data are available.

It also asks how those changes affect our way of explaining FDI and other activities of

multinational enterprises (MNEs), and draws basic public policy conclusions from the evolving

landscape of FDI and our understanding about the main driving forces behind those changes.

[Key words: foreign direct investment, multinational enterprise, Triad, emerging markets,

indirect FDI, productive investment

JEL codes: F01; F10; F23; M16]

Introduction

In this presentation, we highlight various changes over the recent period observed,

including a very fast rise of FDI, its geographical spreading, especially in and from emerging

markets, its sectoral shift towards services, the parallel expansion of multinational enterprises

(MNEs),2 the growing financialization of FDI,3 resulting in such phenomena as FDI in offshore

financial centres, round tripping, 4 transhipment 5 and tax inversion, 6 the de-linking of a

2 Until 2015, the United Nations called these firms transnational corporations (TNCs). It then switched to

multinational enterprises. This study follows that change of standard. However, the term of TNC found in previous

publication is to be treated as a full synonym for MNE. 3 The term “financialization” of FDI is used when transactions are no longer bringing a package of resources to host

countries, such as job creation, transfer of technology, access to new markets, skills development, opportunities for

linkages etc., as it would be usual in traditional FDI, but are technical flows of money aimed at providing financial

gains for the MNEs carrying out those operations. However, as FDI is a balance of payments concept measured by

their financial aspect, these transactions are technically still FDI. 4 In round tripping, there are typically two interrelated transactions: one that leaves country A for country B, and

another one, in which FDI returns from B to A. In substantive terms, for country A, this transaction does not add to

its capital or technological base, as the ultimate beneficiary owner is located within the country.

2

substantive part of FDI from productive capacity building, and increasing volatility in FDI.

These phenomena raise questions not just about the extent of development gains to be derived

from FDI, but also about our capacity to explain why one can talk about a new world of FDI,

which affects the capacities of public policy makers to react properly to changing patterns of FDI.

In our building of explanatory theories, challenges include, on the one hand, the need for re-

focusing our attention to the various push factors of FDI that are to be found in the home

countries of MNEs, and on the other hand, the need for combining the FDI paradigm with

elements of financial flow theories.

The selection of the period (1991–2016) is intentional. Its initial year was marked with

the fall of communism in the Soviet Union and the fall of the Soviet Union proper. Naturally

some people may argue that it would have been better to choose 1989, the year of the fall of the

Berlin Wall. However, the disappearing of the Soviet Union two years later was the event that

made the developments started in 1989 irrevocable on a global scale. This was the “end of

history” [Fukuyama 1992], the start of a new age of liberalism. That period later on turned out to

be also the age of strong reactions to liberalism, but that is already another story. As the last

section will show, in any case, the influence of liberalism on national policies towards FDI was

stronger than the impact of the anti-liberal reactions. The end of this period (2016) is

characterized by increasing policy uncertainty tilting towards more protectionist tendencies

epitomized by the surprise vote of the United Kingdom of Great Britain and Northern Ireland to

leave the European Union (Brexit) and the election of Donald Trump to the presidency of the

United States of America. Especially the latter one may have a major impact on international

economic relations, as the new administration forced the Group of Twenty to abandon the

condemnation of protectionism (a rather symbolic measure), announced its departure from the

Trans-Pacific Partnership (a real measure) and threatened to renegotiate the North American Free

Trade Agreement.

The shifting global FDI landscape

This section sums up the changing global landscape for FDI in 1991–2016 in seven

observations. These observations are sometimes linked with each other, and, taken together,

5 In transhipped FDI, capital country A for country B (the intermediary country) to be further re-invested in country

C (the final destination country). Transhipment inflates the inward FDI numbers of the intermediate country (B),

and can provide misleading information to the final destination country (C) about the origin of the capital (which in

reality is A, and not B as ordinary statistics would show). 6 Tax/corporate inversion is a complex transaction under which a former affiliate becomes the new parent company

of the MNE and the former parent becomes an affiliate, at least for the reporting of corporate finances and profits.

In reality, it is unclear which of the headquarter functions are de facto transferred to the new “parent” company.

3

provide a picture of increasing complexity, especially from the point of view of identifying

development gains.

Observation № 1: spectacular rise

FDI is far from being a new phenomenon; its spectacular expansion and geographical

dispersion between 1991 and 2016 are. In a generation, we witnessed enormous growth and

structural change. In 1991, the global FDI stock stood at around $2.5 trillion.7 By 2016, the

world stock of inward FDI expanded to $27 trillion – an eleven times increase in 25 years (figure

1).8 In 1991–1995, annual FDI flows averaged $226 billion. In 2012–2016 (to use another five-

year term for comparison), they were practically seven times higher: close to 1.6 trillion. Note

that the latter period came just after the Great Recession, in which FDI flows suffered seriously.9

For a comparison with another main channel of international economic transactions, trade,

over the same period of time, world merchandise exports grew less than five times, from $3.5

trillion to $16 trillion, and world GDP only slightly more than three times, from $24 to $75

trillion (figure 1).10

7 Unless otherwise stated, FDI data are derived from the UNCTAD FDI/MNE database. Please also note that due to

reasons of imperfections in statistics, the world totals of inward and outward FDI do not necessarily match. 8 Readers should be aware that the use of stocks for measuring the spread of FDI is deliberate; it is linked to the

lumpiness of flows (see, e.g. [Head & Ries 2008]; [Blobel et al. 2012]), which makes it more difficult to draw

conclusions from the flows of individual years. This choice is also related to the fact that new flows add to

capacities created in previous years. In this respect, FDI is very different from trade; the latter is typically a one-off

phenomenon. 9 It is strongly counter-indicated though for econometricians to compare the three series because they are very

different in nature: one of them is stock, the other two are flows; trade in a gross concept, while GDP is a value

added concept. The aim of figure 1 is just to give an idea about their scale. 10 Source: UNCTADstat database.

4

Figure 1. World GDP, merchandise exports and inward FDI stock, 1991 and 2016

(Billions of dollars)

Source: Author’s calculations, based on data from UNCTAD.

Note: 2016 GDP is an estimate.

0

10.000

20.000

30.000

40.000

50.000

60.000

70.000

80.000

1991 2016

GDP Merchandise exports Inward FDI stock

5

Observation № 2: geographical diffusion

An equally major change in FDI is its geographical dispersion since 1991. In the late

1980s, the so-called Triad consisting of the United States, the European Community (today’s

European Union (EU)-12)11 and Japan accounted for the bulk of total outward stocks and flows

[UNCTC 1991: 31]. Since then, their share in both inward and outward stocks has declined,

although unevenly. It is possible to establish a comparable graphic presentation with a consistent

series between 1991 and 2016. They confirm the overall, but not linear, decline in the Triad’s

share in both inward and outward stocks, but through different trajectories (figure 2). In inbound

FDI, the Triad’s share fluctuated, or even increased in certain years (e.g. in 1995–1998), before

declining dramatically in the aftermath of the Great Recession, to around 45 per cent in 2012,

and recovering slightly afterwards. In summary, we can say that since 2009, we have been living

in a world in which the Triad no longer attracts the bulk of FDI inflows of the world (for more

details, see also below).

Figure 2. World FDI inward and outward stock by main host and home groups, 1991–2016

(Per cent)

(a) Inward stock

11 Belgium, Denmark, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain and

the United Kingdom of Great Britain and Northern Ireland. This chapter covers the period up to 2016, and

therefore treats the United Kingdom as still part of the EU. It is to be noted that, at the moment of closing this

manuscript (September 2017), negotiations on the application of Article 50 of the Treaty on European Union were

still ongoing. Therefore no statement about the details of the exit of the United Kingdom can be used in this study.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

19

91

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

United States Japan European Union-12 Others

6

(b) Outward stock

Source: Author’s calculations, based on statistics from the UNCTAD FDI/MNE database.

There were also differences between the three main Triad centres. The share of Japan in

inward FDI always remained marginal. The EU-12 accounted for most of the fluctuations while

the United States saw a slight increase in its share in most of the 1990s, followed by a consistent

decline.

In outward FDI, the shift is equally clear but from a more pronounced dominance of the

Triad in the base year of 1991 when non-Triad accounted for only 18 per cent of global FDI

stocks (figure 2b). Its share grew in practically all years, and came close to 40 per cent in 2016.

In the meantime, the share of Japan fell from 9 to 5 per cent, that of the US from 33 to 24 per

cent, while the EU-12 had a rather stable over 40 per cent share in the beginning of the period,

with a decline to 31 per cent by the end of the period (due to the Great Recession).

Shifting geography for FDI has major implications for how we explain FDI in our

economic theory. On the one had the recent shift of inbound FDI to new locations shows that

some newcomers are particularly successful in leveraging their national competitiveness. The

emergence of new sources of FDI in turn indicates that we have to revise our explanations on

what drives and determines FDI and other types of MNE activity.

By 2016, the 25 largest recipients of FDI accounted for more than four-fifths of global

inward stocks (table 1). Of these 25, only 10 were Triad countries (shown in italics in the table).

The top positions of the list showed a sandwich of Triad and newcomer countries: while the

United States, the United Kingdom remained the first and fourth largest recipients of FDI,

respectively (in that order), the second, third and fifth positions were occupied by newcomer

economies: Hong Kong (China), China and Singapore, respectively. Moreover, the FDI stocks of

newcomer countries were growing faster than that of Triad economies. Over the whole period,

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

19

91

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

20

13

20

14

20

15

20

16

United States Japan European Union-12 Others

7

the inward stocks of India, the Russian Federation and China, in that order, showed the highest

dynamism. They also grew phenomenally in countries starting from a very low base such as 27th-

placed Poland (therefore not yet shown in the table), exhibiting a growth of 43 000 per cent. It is

to be noted that Brazil, the last of the "BRIC" also showed above-average dynamism, but much

its growth was far slower than that of the other three. As for the period since 2001, India was still

the top winner, but the other "BRICs" were only in the middle range of growth (the Russian

Federation 6th, China 7th, and Brazil 12th).12

Table 1. World inward FDI stock in 2016, and its growth since 1991 and 2001

(Billions of dollars and per cent)

Rank Economy FDI stock in 2016 ($ billion) Growth since 1991 (%) Growth since 2001 (%)

- World 26 728 981 258

1 United States 6 391 855 150

2 Hong Kong, China 1 591 685 297

3 China 1 354 5 305 567

4 United Kingdom 1 197 474 149

5 Singapore 1 096 2 980 690

6 Canada 956 717 201

7 Ireland 840 2 034 526

8 Netherlands 801 962 183

9 Switzerland 793 2 119 794

10 Germany 771 206 82

11 France 698 498 257

12 Brazil 626 1 522 432

13 Australia 576 606 366

14 Spain 557 600 214

15 Belgium 475 n.a. 107 a

16 Mexico 474 1 438 186

17 Russian Federation 379 11 458 b 650

18 Italy 346 463 202

19 India 319 18 291 1 519

20 Sweden 290 1 502 215

21 Luxembourg 245 n.a. 528 a

22 Chile 239 1 309 449

23 Indonesia 235 2 200 1 445

24 Saudi Arabia 232 1 407 1 240

25 Thailand 189 1 733 453

Source: Author’s calculations, based on statistics from the UNCTAD FDI/MNE database.

Note: Ranking excludes offshore financial centres. Triad countries are shown in italics. a 2002. b 1994.

In 2016, the concentration of FDI was even higher in terms of outward FDI stocks than in

terms of inward FDI stocks. The top 25 source countries accounted for more than nine-tenth of

the world total in 2016 (table 2). In that group, the number of Triad economies was 12, and

12 In this global list, Colombia is 31st, with an FDI inward stock of $164,249 billion in 2016. Its growth since 1991

(4'460 per cent) is slightly less than those of the BRICs, but still more than four times higher than the world

average. And since 2001, growth was very high again (968 per cent), more than three times of the world average,

ahead of all BRICs except India.

8

among the top five, there was only one non-Triad: Hong Kong (China). The rest of the top

positions were occupied by the United States, the United Kingdom, Japan and Germany, in that

order. In this group again, emerging economies tend to be more dynamic than Triad economies.

Growth of outward FDI stock from China, the Russian Federation and Hong Kong (China), in

that order, was the highest since 1991. After 2001, the fastest growing sources of FDI have been

China, Ireland and the Republic of Korea, followed by South Africa, Luxembourg and the

Russian Federation, in their majority non-Triad economies.

Table 2. World outward FDI stock in 2016, and its growth since 1991 and 2001

(Billions of dollars and per cent)

Rank Economy FDI stock in 2016 ($ billion) Growth since 1991 (%) Growth since 2001 (%)

- World 26 160 935 259

1 United States 6 384 671 176

2 Hong Kong, China 1 528 10 832 339

3 United Kingdom 1 444 522 60

4 Japan 1 401 504 367

5 Germany 1 365 288 146

6 China 1 281 23 763 3 596

7 France 1 259 797 212

8 Netherlands 1 256 947 278

9 Canada 1 220 1 193 203

10 Switzerland 1 131 1 390 348

11 Ireland 833 5 402 1 940

12 Singapore 682 7 228 657

13 Spain 516 2 414 259

14 Italy 460 554 175

15 Belgium 453 n.a. 126 a

16 Australia 402 1 027 242

17 Sweden 382 598 210

18 Russian Federation 336 12 874 b 676

19 Taiwan Pr. of China 321 890 354

20 Korea, Republic of 306 9 100 1 433

21 Luxembourg 230 n.a. 748 a

22 Austria 200 3 107 602

23 Norway 188 1 445 401

24 Denmark 179 1 045 128

25 South Africa 173 973 1 093

Source: Author’s calculations, based on statistics from the UNCTAD FDI/MNE database.

Note: Ranking excludes offshore financial centres. a 2002. b 1994.

If we cluster the economies (countries and territories) of the world in regions, we can

observe that not all areas have benefitted equally from the spread of FDI. The main beneficiary is

developing Asia, as it can also be deducted from tables 1 and 2. On the list of 25 largest host

economies, there are seven that the United Nations classifies as belonging to developing Asia,

three to Latin America and the Caribbean, none to Africa, none to developing Oceania, and one

to countries in transition. On the list of largest home economies, the respective numbers are five

9

(developing Asia), zero (Latin America and the Caribbean), one (Africa), zero (developing

Oceania) and one (countries in transition).

To have a deeper understanding of regional shifts, we are looking at the dynamism of

inward FDI stocks by emerging (developing and transition) areas between 1991 and 2016. Over

this period, the share of developing Asia in world inward FDI stock increased from 15 to 23 per

cent, while that of Latin America and the Caribbean from 5 to 7 per cent only (figure 3a). A

salient difference between the two regions can be observed since the end of the Great Recession:

since 2010, the rise of Asia accelerated while the share of Latin America and the Caribbean has

stagnated and declined. Africa and transition economies have remained even more marginal

recipients of inward FDI. The share of the former was around 3 per cent at the beginning and at

the end of the period; that of the latter rose fast until the Great Recession, to fall back to around 3

per cent afterwards. The share of developing Oceania, which was not possible to show in the

figure due to its absolute marginality, hovered around 0.1 per cent.

One of the objections to that calculation is that the definition of developing Asia includes

high-income countries, distorting the picture. However, if we exclude those 12 economies, the

rise of Asia remains equally salient. In 1991–1994, the share of Latin America and the Caribbean

eventually exceeded the share of developing Asia excluding the high-income economies. The

latter caught up with the former in the mid-1990s. The difference between the two in favour of

developing Asia started to grow dramatically in 2008. Using the measures of excluding high-

income countries, by 2016, the inward FDI stock of Asia without high-income economies (11 per

cent) was almost one and a half time higher than that of Latin America and the Caribbean (7 per

cent).

Regional trends in outward FDI were fairly similar to patterns inward FDI but the rise of

developing Asia was much steeper. Between 1991 and 2016, its share in world inward FDI stock

rose more than six times, from around 3 to around 19 per cent, while the shares of Latin America

and the Caribbean (around 2 per cent), Africa (around 1 per cent) and developing Oceania (less

than 0.01 per cent) remained unchanged, with some fluctuations, while the share of transition

economies rose from zero to 2 per cent in 2007, to fall back slightly to 1.5 per cent afterwards

(figure 3b). If we exclude the high-income economies, the share of developing Asia rose from

0.4 to 6 per cent. Notably, the outward FDI stock of Latin America and the Caribbean in 1991

was almost four times higher than developing Asia excluding the high-income economies. It was

only in 1995 that the share of those Asian economies exceeded the share of their Latin American

peers for the first time. By 2015, the ratio between the two groups rose to close to 3:1, in favour

of the Asian group.

10

Figure 3. Share of emerging regions in world inward and outward FDI stock, 1991–2016

(Per cent)

(a) Inward stock

(b) Outward stock

Source: Author’s calculations, based on data from UNCTAD.

Note: These figures follow the United Nations classification of regions. Developing Asia includes all Asian economies except for

Israel and Japan; the latter are classified as developed countries. The United Nations list of developing Asia includes 12

economies that the World Bank classifies high income: Bahrain, Brunei Darussalam, the Hong Kong Special Administrative

Region of China, the Republic of Korea, Kuwait, the Macao Special Administrative Region of China, Oman, Qatar, Saudi Arabia,

Singapore, Taiwan Province of China, and the United Arab Emirates. The United Nations considers Turkey to be part of the

region of developing Asia. Developing Oceania excludes Australia and New Zealand; the latter are classified as developed

countries. The data for Latin America and the Caribbean exclude FDI of the offshore financial centres: Anguilla, Antigua and

Barbuda, Aruba, the Bahamas, Barbados, the British Virgin Islands, the Cayman Islands, Curaçao, Dominica, Grenada,

Montserrat, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Sint Maarten, and the Turks and Caicos Islands.

The United Nations do not classify the countries in transition as part of developing economies but as a separate, third group,

additional to the developed and developing categories. This list of countries in transition does not include those countries that

0

5

10

15

20

25Africa

Developing Asia

Developing Asiaexcluding high-incomeeconomiesLatin America andthe Caribbean

Transitioneconomies

0

2

4

6

8

10

12

14

16

18

20 Africa

Developing Asia

Developing Asiaexcluding high-incomeeconomiesLatin Americaand theCaribbean

Transitioneconomies

11

have become members of the European Union. The table does not show developing Oceania, whose share in global FDI is

marginal.

The dispersion of FDI across the globe has been largely facilitated by the increasing

fragmentation of international production and the rise of value chains in both manufacturing and

in services. In other words, countries attracting FDI no longer specialize on the preparation of

final products or services from the "raw material" to the consumption-ready part but only one

segment of the value chain. Once they completed their value-adding activities to the semi-

finished product or service, they ship it to the final location for "assembling". In turn, assembling

locations (e.g. China, increasingly the workbench of the world during the period observed)

specialize only on that phase of production. The extent of value chain-related activities is large.

It was estimated that in 2010, 28 per cent of global trade was related to value-chain related

activities [UNCTAD 2013: 125]. In East and South-East Asia (the region that includes China, the

Republic of Korea, Malaysia, Singapore, Taiwan Province of China, Thailand, Viet Nam etc.)

that share was 30 per cent, in Central America (including Costa Rica and Mexico) 31 per cent,

and in the European Union (including both the large leading economies such as Germany and

new members such as the Czech Republic and Poland) 39 per cent.

Observation № 3: shift towards services

In global FDI, important sectoral changes took place over the past generation, with a shift

taking place mostly in favour of services. In 1991, they represented about one half of the global

FDI stock, in 2001 around 58 per cent, and in 2015, around two-thirds [UNCTAD 2017: 21],

although this may be an overestimation as many of the services activities are related to

manufacturing, carried out by industrial MNEs. Even keeping this reservation in mind, for

corporate strategies and for investment promotion efforts following trends in MNE activities, this

shift meant not only that manufacturing has lost its dominant position in FDI and other forms of

international production such as contract manufacturing and farming, service outsourcing,

franchising and licensing [UNCTAD 2011], but also that the sustainability and value adding

capabilities of manufacturing projects depended more and more on the adjacent services

activities such as research and development (R&D), marketing or aftercare services. All this has

to be viewed in the light of value chain-related activities, in which indeed the division line

between manufacturing and services becomes blurred. It is particularly clear in the automotive

sector, one of the usual priorities of investment attraction in middle-income countries due to its

broad-based economic linkages, spillover effects and technological and knowledge content.

Observation № 4: spread of multinationals and affiliates

The number of MNEs, the main agents of FDI transactions, and of their affiliates located

in foreign countries, exploded between 1991 and 2016. Indeed, if we discount the marginal effect

12

of suitcase13 and diaspora investors,14 whose importance is minimal, except in some special cases

like in Africa, where “foreign-owned and operated firms that are not subsidiaries of a foreign

based enterprise but are owned and operated by a foreign entrepreneur” [UNIDO 2005: x],

including diaspora investors, play an important role, practically all remaining FDI flows and

stocks are to be attributed to MNEs. And it is possible to use the term explosion about the

number of MNEs despite that fact that this presentation uses a restrictive definition of the term,

meaning companies owning assets in foreign countries, not just firms carrying out trade or other

transactions with foreign partners, but without presence in foreign locations through the

possession of assets.

Even if we stick to this definition, the current number of MNEs is surprisingly big,

witnessing a massive spread of MNEs around the globe. In UNCTAD’s latest counting

[UNCTAD 2017], there are around 100,000 MNEs around the globe, controlling and managing

approximately 860,000 affiliates in foreign countries, including subsidiaries, associated

companies, branches and representative offices. In the early 1990s, the estimate was about 37

500 MNEs and 207 000 affiliates [UNCTAD 1994: 4]. In other words, thanks to probably partly

better reporting, but mostly to a real spread of MNEs, the number of parent firms increased more

than 2.5 times, and the number of affiliates more than four times (figure 4).

13 We talk about suitcase FDI when is the home country (country of origin), there is now parent MNE involved in

the FDI transaction. The investment is carried out directly by a person having her/his residence in the home

country (see [UNIDO 2005: x]). 14 Diaspora FDI is FDI carried out by current or former citizens of the host country (the “Diaspora”) having

established their permanent residence or their firms, or both, in the home country. It is still FDI because the test of

foreignness hinges on the place of residence of the investing person or company, not the colour of the owner’s

passport.

13

Figure 4. Estimated number of multinationals and their foreign affiliates, early 1990s and 2017

Source: Author’s calculations, based on data from [UNCTAD 1994 and 2017].

A search for explanations of the spread of MNEs and affiliates can naturally point out

that the rise in their numbers was slower than the increase in global FDI stocks, which were

multiplied by ten times over the same period. In other words, the average size of MNEs

increased fourfold, and the average size of affiliates two and the half times, while the compound

inflation of the US dollar, in which data are expressed, was only 74 per cent. Therefore, we can

conclude that the increase in the number of MNEs and affiliates is not surprisingly big but

surprisingly moderate, especially if we consider the rise of new source countries, whose MNEs

are additional to the MNEs of the Triad (observation № 2). There may have been other factors at

play: for instance, with the disintegration of federal countries (former Yugoslavia, former Soviet

Union, former Czechoslovakia etc.), some firms became MNE overnight as part of their assets

remained in territories that became foreign countries. However, we have to consider also the fact

that even in these former federal countries under communist rule, there existed a non-negligible

number of MNEs, mostly dealing with foreign trading operations (cf. [Hamilton 1986], and

[Bulatov 1998] for the specific Soviet case). After transition to the market economy, some of

these MNEs disappeared, while others transformed themselves to new MNEs adapted to the

changing situation.

It is also possible that the creation and expansion of integration groupings has had an

impact on the number of MNEs and affiliates. However, we lack systematic evidence in this

respect, and the theoretical link between integration and FDI is ambiguous: deepening

integration could stimulate more FDI (“FDI creation”) but also discourage it (“FDI diversion”)

as the intra-grouping locations can be served by trade or other non-FDI transactions, too (cf.

[Dunning & Robson 1998].)

0

100.000

200.000

300.000

400.000

500.000

600.000

700.000

800.000

900.000

1.000.000

Early 1990s 2017

Multinationals

Foreign affiliates

14

There are also increasing size differences between MNEs. In 2015, the 100 largest ones

controlled $13 trillion in total assets and $8 trillion in assets abroad. As the “total assets of

foreign affiliates” was reported to be $113 trillion in 2016 [UNCTAD 2017: 26], the foreign

assets of the remaining 99 900 MNEs should stand at $105 trillion. On average, that gives us

about $1 billion per MNE. If this universe then follows a normal distribution, an important part

should be small and medium-sized enterprises according to their national definitions. In other

words, the two universes are not distinct from each other but overlap, an observation already

made in the 1990s [Fujita 1995], but often ignored.

Observation № 5: financialization

In recent years, and especially after the Great Recession, it has become more visible that

part of what is considered to statistically FDI is in fact differs from the classical case of an FDI

transaction, in which investment in companies and control over them is accompanies by a

package of “resources”, such as jobs, skills and technology transfer, transfer of managerial skills,

and opening access to foreign markets, to mention probably the most important ones. Counted

technically as FDI, these transactions are in fact closer to portfolio investment. World Investment

Report 2016 [UNCTAD 2016] has recounted that in certain MNEs, it takes up to seven layers of

intermediation before we arrive from the affiliate to the ultimate beneficiary owner. The main

reason of this layering is tax optimization, although sometimes other motivations related to the

regulation of companies can play a role.

Hence, the increasing financialization of FDI results in a phenomenon that can be

generally called indirect FDI [Altzinger & Bellak 1999]; [Kalotay 2012], as opposed to “direct”

FDI, in which the relationship between the parent and the affiliate is without intermediation. This

phenomenon uses at least three types of locations: “ordinary ones”, offshore financial centres and

centres for special purpose vehicles (SPVs). For the phenomenon of financialization, the latter

are of particular importance. It is interesting to observe that tax havens are more in the spotlight

than SPV centres due to the role they play in the management of private fortunes, although SPVs

may be bigger in size. A recent estimate [UNCTAD 2015] has found that in 2012, about 10 per

cent of global FDI stock was routed via tax heavens, and 20 per cent via SPV centres. Exact

counting is further complicated by the fact that, although countries are asked to report their FDI

without SPVs (and in principle UNCTAD data exclude them from the world total), that

recommendation is implemented only partly, making the data even less transparent. Furthermore,

not all indirect FDI transactions require the mediation of offshore financial or SPV centres,

making it further possible that the extent of indirect FDI is underestimated. Indeed, the three

main financial constructs, namely, round-tripping (in which FDI returns to the country of origin),

transhipment (in which the final destination is different from the country of origin) and

tax/corporate inversion (in which the parent and the affiliate exchange places) can very well use

“ordinary” FDI locations.

15

The taking into account of the transactions described above is further complicated by the

fact that not all indirect FDI belongs to the category of financialization. Often indirect FDI is

used for such reasons as more proximity to target markets, cultural affinity, logistical

consideration etc. For instance, if General Motors uses its German affiliate Adam Opel to invest

in Poland, it has little to do with financialization, and more with the reasons mentioned in this

paragraph.

Even the statistical monitoring of the direction of FDI is very difficult. A major

uncertainty stems from the fact that countries can and should register the immediate source or

destination country of FDI (depending on whether it is inward or outward), and not necessarily

the country of the ultimate beneficiary owner or of the ultimate destination country. As a

consequence, a large part of FDI is registered in centres of round tripping, transhipment and

special purpose vehicles. It is possible to illustrate this point with the outward FDI stocks of two

countries that are usually considered to be emerging and which are producing regular reports.

In both Brazil and the Russian Federation, two countries figuring among the top 25

source countries (table 2), and whose outward FDI stock grew faster than the world average over

the period 1991–2005,15 offshore locations often figure among the top destinations of their FDI

according to the immediate investment concept (table 3). There is no doubt, looking at the

Brazilian list, that at least the Cayman Islands (1st), the British Virgin Islands (4th), the Bahamas

(5th), Luxembourg (6th), Panama (12th) and Curaçao (21st) fall into that category. But even when

we consider other top destinations such as Austria (2nd), the Netherlands (3rd) and Switzerland

(13th) are also among countries in which major financial engineering transactions take place, for

instance, via SPVs. In the case of the Russian Federation, the main offshore locations are Cyprus

(1st), the British Virgin Islands (3rd), Luxembourg (6th), the Bahamas (12th), the Cayman Islands

(19th) and Bermuda (24th). Russian investors also frequently use the Netherlands (2nd), Austria

(4th), and Switzerland (5th) as frequent investment destination.

Table 3. Top 25 destination countries of Brazilian and Russian outward FDI stocks, 2015

(Billions of dollars and per cent of total) Brazil Russian Federation

Rank Destination economy Value Share in total Destination economy Value Share in total

Total OFDI 282'950 100.0

Total OFDI 371'719 100.0

1 Cayman Islands 61'385 21.7

Cyprus 112'582 30.3

2 Austria 40'824 14.4

Netherlands 65'467 17.6

3 Netherlands 38'228 13.5

British Virgin Islands 38'085 10.2

4 British Virgin Islands 26'819 9.5

Austria 21'848 5.9

5 Bahamas 21'436 7.6

Switzerland 17'978 4.8

6 Luxembourg 17'668 6.2

Luxembourg 14'661 3.9

7 Spain 13'790 4.9

Germany 10'990 3.0

15 Between 1991 and 2005, world outward FDI stock grew by 242 per cent, that of Brazil somewhat faster (300 per

cent), and that of the Russian Federation particularly fast (we have reliable data only from 1993 on; between 1993

and 2015, the country’s outward FDI stock grew by 483 per cent).

16

8 United States 12'451 4.4

United Kingdom 8'243 2.2

9 Argentina 5'896 2.1

United States 7'183 1.9

10 Chile 4'237 1.5

Spain 6'223 1.7

11 United Kingdom 4'027 1.4

Turkey 6'015 1.6

12 Panama 3'988 1.4

Bahamas 4'282 1.2

13 Switzerland 3'704 1.3

France 3'303 0.9

14 Portugal 3'361 1.2

Ireland 3'248 0.9

15 Uruguay 3'115 1.1

Bulgaria 3'111 0.8

16 Hungary 2'809 1.0

Belarus 3'044 0.8

17 Venezuela, Bolivarian Rep. of 2'449 0.9

Kazakhstan 2'583 0.7

18 Belgium 1'614 0.6

Finland 2'514 0.7

19 Sweden 1'427 0.5

Cayman Islands 2'311 0.6

20 Mexico 1'122 0.4

Italy 2'309 0.6

21 Curaçao 939 0.3

Latvia 1'912 0.5

22 Peru 867 0.3

Czech Republic 1'876 0.5

23 Denmark 787 0.3

Ukraine 1'861 0.5

24 Paraguay 773 0.3

Bermuda 1'557 0.4

25 France 639 0.2

Canada 1'508 0.4

Top 25 274'356 97.0

Top 25 344'693 92.7

Source: Author’s calculations, based on national statistics. Note: Typical offshore locations are shown in italics.

The prominence of offshore and financialization centres among immediate investment

locations makes it more difficult to gauge the real geographical spread of Brazilian and Russian

MNEs. If we discard the offshore centres, the share of remaining Latin America and the

Caribbean stands at around 7 per cent (with Argentina, Chile, Uruguay and the Bolivarian

Republic of Venezuela among the top Latin American destinations). In the case of the Russia

Federation, if all countries in transition are counted, including the ones that have joined the EU,

the intra-regional share is slightly over 5 per cent, with Belarus, Kazakhstan, Latvia, the Czech

Republic and Ukraine figuring among the top locations. Naturally, knowing the extent of

offshore locations and the potential for consequent transhipment, these numbers in both countries

have to be treated as surely underestimations of the regional links.

As for the share of developed economies, it is higher in both cases (around 42 per cent in

each case) but includes the ambiguous locations of Austria, the Netherlands and Switzerland. In

any case, the geography of outward FDI of these economies is an indication of a mixed strategy,

in which intra-regional links still matter, but MNEs form these countries already have major

implantations in developed economies, too, and use actively the services of offshore financial

centres.

The main consequences of financialization are blurred information about the nationality

of the investor, a de-linking of transactions from the real economy, and increasing volatility of

FDI. The latter two are analysed briefly in the subsequent two observations.

17

Observation № 6: partial de-linking from productive capacity building

Due to various factors, but especially those described in our observation number 5 about

financialization, we can observe a growing de-link of FDI from productive capacity building in

host countries. To recall, many transactions that MNEs undertake under the title of FDI do not

add to job creation or technology transfer, or other effects that host countries usually can expect

from such transactions. But it is also true that MNEs undertake activities under the general

leading of non-equity modes (NEMs) of investment (such as contract manufacturing, business

process outsourcing, franchising or licensing) that affect production in host countries but is not

counted as FDI because form a formal point of view does not imply the ownership of assets. As

for the scale of those transactions, it has been estimated [UNCTAD 2011: 132] that sales related

to those NEMs amounted to $2 trillion in 2010, compared with $33 trillion in terms of sales of

foreign affiliates. Therefore, NEMs further make FDI deviate from productive investment, but to

the opposite direction as financialization; as a result, FDI data become even dimmer.

We have to stress that by productive capacity we mean a broad range of activities, not

just the manufacturing of nuts and bolts. One difficulty here is the fact that productive capacity is

a very popular term both in public discussion and in academic literature but it is never defined

explicitly. A reading of the literature gives us the impression that most of the authors imply an

investment that is accompanies by such gains as job creation, skills development, transfer of

technology, supplier opportunities for local firms, to mention a few. These are in fact impacts

that figure prominently on the list of desired effects of FDI, too. It includes not just primary and

manufacturing activities, but also most of the services activities, even in finances (if the aim is to

sell financial products to clients). That leaves out a relatively small part of activities outside the

cases of financial engineering proper, such as real estate investment.

Keeping that tentative definition in mind, one of the main issues is the measurement of

productive versus non-productive FDI. In reality, it is impossible to divide FDI transactions into

these categories. However, it is possible to use proxies to gauge the extent of the phenomenon.

One way is to compare the effective capital expenditures of large MNEs (accounting for the bulk

or the investment outlays of the MNE universe) with FDI flows. Here the issue is not so much to

compare the sizes of the two phenomena, but to see commonalities and differences in their trends.

A look at the period of 2007–2015 (figure 5) proves the existence of a strong de-link

between the two: in four years, the tow moved together but in four other years, including 2008,

2012, 2013 and 2015, they moved towards the opposite direction. In other words, when one of

them increased, the other decreased.

18

Figure 5. Comparison of world FDI inflows and capital expenditures of top 5,000 MNEs, 2007–2015

(Billions of dollars)

Source: Author’s calculations, based on data from UNCTAD.

Observation № 7: volatility

A major corollary of the previous two observations is the increasing volatility of FDI. In

variance to the traditional paradigm about the (relative) stability of FDI compared with portfolio

and other investment, FDI has recently become more and more volatile. In other words, there

was such an amount of instable transactions that even at the global level it was impossible to

cancel out the lumpiness of FDI in individual economies. For instance, 2011–2016 was a six-

year period in which the direction of growth and contraction of FDI changed no less than three

times, and a change (from growth to contraction) is foreseen for 2017 again [UNCTAD 2017: 4].

This is the highest level of volatility observed since 1991. Moreover, in this period, inflows

decline three times. In 2001–2005 and in 2006–2010, the number of volatile years was only two

(despite the fact that the Great Recession fell in the middle of the latter period), and in 1996–

2000 that number was zero.

Challenges for economic and business theories16

It is not easy to create and maintain a paradigm of FDI that is specific enough for the

phenomenon, and takes into consideration all the dynamics and changes that the previous section

has just described. Just enough to consider the fact nine-tenth of the FDI stock with which we are

16 The arguments of this section draw partly on [Kalotay et al. 2014]. The author is grateful to Andrea Éltető,

Magdolna Sass and Csaba Weiner for their insights on FDI/MNE theories.

0

500

1.000

1.500

2.000

0

500

1.000

1.500

2.000

2.500

2007 2008 2009 2010 2011 2012 2013 2014 2015

World FDI inflows (right scale)

Capital expenditures of top 5,000 MNEs (left scale)

19

currently dealing was accumulated over the past generation. No wonder that our theories and

paradigms are notoriously late. They were created for a world in which FDI was much smaller,

more homogeneous in terms of main home and host economies, less complex, and more directed

towards manufacturing. At that time, we knew much less about value chains or production

systems (although the first studies on integrated international production were published in the

early 1990s [UNCTAD 1993]; however, it took time till they became mainstream). It was easier

in the 1990s to imagine that FDI was just one more factor movement, and not so a key organizer

of economic activities. We also have to take into consideration the inherent conservativism of

science, and its aversion to deal with complications and nuances. No wonder that our

international economics are moving away so slowly from the world based on classical trade.

In this respect, it has to be emphasized that the first studies describing and attempting to

explain the FDI phenomenon were prepared on the basis of traditional theories of the world

economy, in which trade was the main glue of interactions between countries, and if there were

factor movements, they could be explained by factor endowments. That was the case of the

Heckscher-Ohlin-Samuelson paradigm [Heckscher 1919]; [Ohlin 1933]; [Samuelson 1948,

1949], which posited that capital should flow from rich to poor economies. To its merit, it paid

equal attention to push and pull factors; to their criticism, it has to be mentioned that they treat

countries as single macroeconomic entities. It is also to be recalled that in the original theory of

the international economy, from David Ricardo [1817] via Eli Heckscher [1919] and Bertil Ohlin

[1933] worked with the assumption that capital was immobile. The main difference between

Ricardo and his followers was that the latter allowed for differences in capital endowments as a

key reason for international trade (alongside with endowments in land and labour). We had to

wait till the 1940 and the additions of Samuelson [1948, 1949] to the Heckscher–Ohlin theorem

to accept and integrate the existence of capital movements. However, logically, capital had to

flow from capital-rich to capital-scarce countries, that is, from the developed to the developing

world.

In the 1970s, it was the existence of the Triad that contradicted this theory, as capital was

flowing from one developed country to another developed country, instead of targeting the

developing world. In other words, this was a world that resembled the world of the new trade

theory, under which the exchange of goods took place within industries, and among developed

economies producing similar goods [Krugman 1981, 1983].

In the 1970s and 1980s, the first researchers who noted the importance of MNEs and the

need for economic theory to explain their activities (for example [Dunning 1977] and [Vernon,

1971]), started to build their explanations based on their empirical observations, and focused on

the most salient factors such as the competitiveness of investing firms, and stressed that FDI is

not simply a financial phenomenon but a bunch of resources, and it is about the control and

management of assets. In other words, the early FDI theories wanted to “divorce” from financial

20

theories of capital movements. The three most salient ones are the investment-development path

(IDP), the Uppsala theory and the eclectic paradigm.

The theory of the investment development path (IDP) [Dunning 1981, 1986] is a typical

stages paradigm, which attempts to explain the ratio between inward and outward FDI stocks in

function of the GDP per capita of countries, is yet another theorem which suffers from the

weakness of black box (alleged homogeneity of countries). Apparently the poorest countries of

the world would have no inward or outward FDI (stage 1), and then, with the growth of income,

inbound FDI would grow faster than outbound FDI (stage 2). As countries reach middle income,

outbound FDI grows faster and catches up with inbound FDI (stage 3); beyond middle income,

they become net capital exporters (stage 4), and in the richest countries of the world, the ratio

between inward and outward FDI stocks again becomes uncertain and fluctuates around 1 (stage

5). Simplicity makes this theory rather attractive; it is however very difficult to translate it into

concrete numbers (quantify the GDP per capita belonging to each stage).

The Uppsala School ([Johanson & Vahlne, 1977, 1990] [Johanson & Wiedersheim-Paul,

1975]), is yet another stages theory, but this time limiting its observations to internationalising

firms. Firms following the Uppsala theory would start operating with limited experience and face

uncertainty on foreign markets; they would internationalise via international trade at best. They

envisage investing abroad gradually. Then they gain experience in FDI, and become major

global players in the longer run. The Uppsala School builds on Raymond Vernon’s product cycle

hypothesis [Vernon 1966, 1979], according to which a new merchandise is produced first in the

country in which it was invented, then exported, and its production is gradually relocated

overseas when the product become mature, and its home country production is no longer

economical. 17 One of the main merits of the Uppsala School is that it provides ideas for

explaining how MNEs are overcoming their liability of foreignness in overseas markets

[Johanson & Vahlne 2009]; [Luo & Mezias 2002].

The eclectic – or OLI – paradigm of Dunning [1977] covers a broader range of factors. In

this presentation, we only recall its main lines. The eclectic paradigm posits that for firms to

successfully invest abroad, they must possess ownership advantages (O), which enables it to

invest successfully in a foreign country. That covers mostly the competitive advantages of MNE

parent firms, and is yet another answer to the question of how the liability of foreignness is

overcome: by way of building strong competitive advantages.

The second leg of the eclectic paradigm, the concept of location advantages (L), refers to

the situation of host countries. That allows us to cover various stakeholders in the host country,

probably the government (through policies) or the local business partners (through linkages),

17 However, the writer of these lines had the opportunity to exchange views with Vernon, who insisted that it was

really just a hypothesis applicable only to some Triad economies (mostly the United States).

21

although not necessarily the bulk of civil society. As a third leg of the paradigm, the firm in

question not only invests but also owns assets permanently abroad, i.e. internalizes foreign

transactions (I) when it is profitable. That covers the local affiliate and its relationship with the

parent firm.

The main merit of this eclectic paradigm is not its originality but its success in weaving

together different strands of previous analysis. For example, the ownership advantages build on

insights on firm- specific competitive advantages by Stephen Hymer [1960] and Edith Penrose

[1968]. The theory of locations advantages draws on such previous studies as [Davidson 1980]

and especially on Michael Porter [1985]. John H. Dunning stressed in many informal discussions

his close link and dialogue with Porter. Their only slight divergence was about the role of MNEs.

Dunning saw them as privileged agents, while Porter thought they had to be seen as one of the

many actors in competitiveness. As for internalisation, it draws mostly on Ronald Coase [1937].

The original OLI framework has been extended and modified several times to adjust to

new sources and recipients of FDI. In the latest updated version of the theory published before

Dunning’s death ([Dunning & Lundan 2008]), the most important change to early versions was

the division of ownership advantages into asset-based advantages (Oa) such as cutting-edge

technologies, marketing strength or powerful brand names, and transaction-based advantages

(Ot) such as common governance of assets and interaction with other corporate networks. The

latter is important because for the first time it covers relationship with home country business

partners.

The flipside of these proposed amendments is that the theory may become too complex.

Narula [2010] for example warned that the creation of too many extensions and sub-categories of

the eclectic paradigm could endanger the integrity of the theory. Rugman [2010] also considered

that the paradigm had become too eclectic and too broad. At the same time, despite these

advances, the emergence of new sources of FDI is still not very well explained ([Child &

Rodrigues 2008]). It is evident that the new MNEs do not possess the same ownership

advantages as their traditional counterparts, if they possess any advantage of all. It may well be

that they are pushed to go abroad because of their relative disadvantages. To deal with this

question, Moon and Roehl [2001] suggested an “imbalance theory” for new FDI, claiming that a

firm wants to search for a kind of balance between ownership advantages and disadvantages

when investing abroad.

As shown in the previous section, with the rise of new sources of FDI, first some

emerging developing economies, then from countries in transition, FDI became multi-

directional, resulting in an even more complex world than that of international trade (although

there, too, the rise of emerging economies represents a challenge for the theory of intra-industry

trade). It is particularly challenging to explain why new MNEs are emerging from new locations,

especially from home countries that at first sight are not “ready” for capital exports. That is the

22

case of the Heckscher-Ohlin-Samuelson paradigm, but also of the main FDI theories in their

original form. For instance, once the new sources of outward FDI have to be explained, the IDP

model is hardly applicable. For instance, Russia’s investment position turns into balance (and

since 2009 FDI outflows have been exceeding inflows) prematurely. The main reason for the

black box approaches limited power of explanation for real FDI flows and stocks is indeed their

aggregate macroeconomic approach, which does not for instance consider such structural

elements as the split of new capital exporting countries into high and low-income segments, and

the accumulation of capital by the high-income group, used in part for international business

expansion.18 As for the Uppsala theory, leapfrogging to international prominence, a common

strategy of Chinese or Russian firms, remains largely unexplained. Chinese and Russian firms

are not the typical technology-based small upstarts, but mostly giant firms deriving large income

from monopolistic power at home or natural resources, and transform their rents to foreign

expansion without regard to traditional technological learning. The overcome their liability of

foreignness even under conditions of apparent competitive disadvantages (for the Russian case,

see [Panibratov et al. 2015]).

The dispersion of sources of FDI, in particular, is challenging for the coherence of

FDI/MNE theories. In a nutshell, there is an increasing tension between the need to keep them

relatively simple and the need to reflect increasing diversity. Analysts should indeed resist in this

context the temptation to create a special theory for each and every new major source country:

one for the Dragon MNEs (it already exists, see [Mathews 2002]), one for the Russian “Eagles”

(it does not yet exist), etc. Such a fragmentation of theory would make cross-country (and over-

time) comparisons very difficult. However, if extant paradigms do not develop together with

time, they risk becoming increasingly irrelevant.

One possibility for reviving FDI theories in the light of new developments is to integrate

the insights on MNE activities in terms of fragmentation of production ([Markusen & Venables

1998] [Venables 1999]) and global value chains ([Gereffi et al. 2001] [Rugman & Verbeke

2004]) organised along “global factories” [Buckley 2011]. However, it is resolving the issue of

how to reflect the behaviour of firms better, not necessarily how to deal with the multiplicity of

stakeholders.

The multiplicity of stakeholders and transactions is probable the most difficult challenge

for economic theories attempting to explain FDI. For sociologists and political scientists looking

at FDI this may be important evidence and a starting point for analysis. However, in economic

theory, in which the micro and the macro are usually separated from each other, the

consideration of this stakeholder phenomenon for an aggregate phenomenon would blur the

division line between micro and macro, making economic explanations even more complex.

18 For the Russian case, see [Kalotay 2008].

23

Other than the parent companies and the affiliates, other main stakeholders in FDI

include business partners in the home and host economies, home and host country governments

with their own legislative, executive and judiciary branches, all their specialised agencies

(including naturally the ones dealing with investment promotion, taxation, competition etc.), at

all levels (federal, subnational and local), other political forces in the home and host countries

such as opposition parties, which may agree or disagree with government policies vis-à-vis

inward and outward FDI, and other civil society in the home and host countries (as business

itself also has to be counted as part of civil society) [Kalotay 2016]. These key stakeholders then

have interactions of various types among them. Naturally the most important ones are the ones

between the parent company and the host country affiliate and the host government and the two

components of the MNE (parent and affiliate). But we should not forget about the interactions

with other business in the home and host country (usually undertaken via business deals and

business associations), the interactions between home and host governments leading to

international legal instruments (e.g. bilateral investment treaties or double taxation treaties), or

the interactions with civil society, especially in the host country, where it is an essential part of

the social license of the foreign investor (social acceptance of its activities, in addition to an

official government license), but also in the home country in which local civil society is often the

driver of a push for corporate social responsibility (figure 6).

Figure 6. Main home and host country stakeholders involved in FDI transactions: a simplified

scheme

Source: the author.

In FDI theories, the stakeholder question is treated only marginally. This is so because it

is a really interdisciplinary issue, on which probably social and political scientists are more at

ease than economists and business scholars more traditionally dealing with MNEs. Among the

three main strains, the OLI is the most developed, or has at least the best potential to deal with it.

24

The locational (L) leg can be in principle split, into various host country stakeholders and their

impact. Ad for the home country, the biggest unknown is if it covers the effects of the home

country government. It may well be that the Ot advantages implicitly cover some of these

interactions. However, it is still does not satisfactorily explain the effects of state capitalism,

typical for China and Russia. State influence can be direct (easy to document) or indirect (more

informal). The latter can become a norm in state capitalism (see [Grätz, 2014] for Russia and

[Wei et al. 2015] for China). The prominent role of the State and the policy environment in

prompting OFDI in these cases may indicate that home country influence can no longer be

assimilated under the Ot factor. Let us also stress that the rest of home country stakeholders are

not even implicit in the Ot (unless the impact of local partners is assumed under the

competitiveness of the MNE, which is probably already much overstretched as a hypothesis).

Despite the further complications it could engender in the integrity of theory, a home-

country (H) factor has to be added to the OLI legs. The OLIH hypothesis [Kalotay, 2010]19 can

cover an Hb leg encompassing the home country business environment, which would go beyond

the relationship between the parent company and its home country partners, and cover the

interaction of business with the rest of society and politics, an Hd leg on the development

strategy of the home government, and an Hs leg explaining the home country State involvement

in outward FDI. All this requires further testing in the future. A first econometric exercise on the

Russian case [Kalotay & Sulstarova 2010] has found promising leads on the importance of the

home country. More analysis based on more (and more detailed) data would be needed to follow

it up.

There is also a need to relate FDI theory more closely with the theorems of foreign

portfolio investment, too. Dunning and Dilyard [1999] made the first steps towards that direction.

However, the limitation of their study was that they mostly attempted to extend the FDI theory to

portfolio investment, with limited efforts towards integrating the findings of the latter. In this

area, there would be a scope for a deeper consideration of main concerns of international

portfolio theory such as the issue of different types of risks and risk taking, currency and

exchange rate issues, the impact of taxation [Bartram & Dufey 2001], the home country

preference of investors, the problem of cross-investments, return on investment, and naturally,

factors of volatility [Gokkent 1997], to mention some of those that could be equally relevant for

FDI, and whose application would enrich our understanding of both FDI and its relationship with

foreign portfolio investment.

Public policy issues

For public policy makers, the main challenge remains the adjustment of their action to the

changing face of FDI, with the aim of ensuring development gains for the population. This

19 See also the contributions of Álvarez & Torrecillas [2013] and Stoian [2013].

25

challenge is nothing new in the current generation, although adjustment to growing complexities

is increasingly difficult. There is no doubt that the majority policy makers of the world are aware

of the growing importance of FDI, they have some ideas about its main patterns, and usually they

wish to attract it. However, in many cases the quantitative aspect dominates (success is measured

by the amount of FDI is attracted), and strategies aimed at maximizing FDI are not well

coordinated with national development goals or other policy areas, such as trade or

competitiveness policies, and even less with such structural policies and industrial or

technological policies.

It is to be stressed that the policies aimed at providing a better regulatory environment for

FDI also have certain inclinations towards formal gains. For instance, these days many countries

set targets to move up in the Doing Business rankings of the World Bank, without asking if that

would have any measurable effect on FDI attraction, and any link with development goals. There

are also certain problems related with the sustainability and social acceptance (social license) of

FDI policies. Only some countries of the world apply impact assessment systematically on their

policy measures. There is also a lack of effective stakeholder involvement in policy making

(including public-private dialogue), which consistently reduces the acceptability of public

policies.

In quantitative terms, most policy changes affecting FDI are directed towards

liberalization and attraction, and only a smaller fraction towards regulation and restriction,

confirming the dominance of liberalism in policies. However, there seems to be a rather uneven

rise of restrictive/regulatory measures over time, indicating a half-shift in policy stances in recent

years. In 2001, the share of those measures more favourable accounted for an overwhelming 88

per cent. Then that share declined to a low of 66 per cent in 2010 but recovered to 75 per cent by

2015, to decline again to 68 per cent in 2016 (table 4). The decline could be attributed mostly to

the Latin American region, where certain Governments changed their stance on FDI relatively

radically, and to two sensitive industries: infrastructure and natural resource extraction.

Table 4. National regulatory changes affecting FDI, 2001–2016

(Number of cases and per cent)

Year

Number of

regulatory changes

worldwide

Restriction/

regulation Indeterminate Liberalization/promotion

Share of

liberalization/promotion in

world total (%)

2001 97 2 10 85 87.6

2002 94 12 3 79 84.0

2003 125 12 0 113 90.4

2004 164 20 2 142 86.6

2005 144 25 1 118 81.9

2006 126 22 0 104 82.5

2007 79 19 2 58 73.4

2008 68 15 2 51 75.0

2009 89 24 4 61 68.5

2010 116 33 6 77 66.4

2011 87 21 3 63 72.4

26

2012 92 21 6 65 70.7

2013 88 21 3 64 72.7

2014 74 12 10 52 70.3

2015 99 14 11 74 74.7

2016 124 22 18 84 67.7

Source: the author's calculations, based on [UNCTAD 2017: 99].

It is to be noted that the political changes of 2016 described in the introduction may also

have a further impact on national policy changes in the future. However, the recent past we have

to describe has still been a period dominated by intentions of FDI attraction. The focus of

policies on attraction seems to make sense, especially if they are accompanied by policies of

beneficiation. Beneficiation can be promoted through various policies, among which particularly

important is the promotion of supplier linkages due to its multiplier impact on local suppliers

[UNCTAD 2001], but equally important are the various incentives and performance

requirements aimed at stimulating such key impacts as job creation, transfer of knowledge, skills

development etc. (see [UNCTAD 1999]), policies having a major impact on economic actors

such as tax policies [UNCTAD 2015] and policies aimed at ensuring the proper functioning of

markets [UNCTAD 1997]. The idea behind active beneficiation policies is the fact that the

benefits of FDI are far from being automatic (cf. [UNCTAD 1999]).

In principle, the fast growth of FDI opens opportunities for many new countries to attract

FDI; however, its spread also means increasing competition for FDI. Targeting becomes more

and more complicated as the number of potential MNE investors increases, and policy makers

should not forget the sectoral shift of FDI towards services. Indeed, many countries still give

special priority to FDI in manufacturing.

One of the main limitations of policy responses to the currently emerging world of FDI is

in its international dimension. First of all, policy makers consider their national actions usually in

relative isolation from other countries. In reality, aggressive measures attracting FDI carry a very

high probability of beggar-thy-neighbour effects. The results are quite negative, especially in the

field of taxation of corporations or incentives offered to them. It seems that the challenges related

to the financialization of FDI in terms of negative development impact can be overcome only

through concerted international action. It concerns mostly taxation but also ownership and

registration rules, the regulations on corporate inversions.

The biggest tension however is between two facts: that FDI is as important as trade, and

that there is no international mechanism or agreement similar to the World Trade Organization

(WTO) in trade. As a result, there are more than 3,300 treaties signed, mostly at the bilateral

level. The popularity of bilateral treaties is surprising because evidence suggests that they play

only limited role in helping to attract FDI [Hallward-Driemeier 2003]; [Poulsen 2009], while

their consequences in terms of costs of international investor-State disputes can be high.

27

It is to be added to that observation that, and as a consequence of financialization and

layers of indirect FDI, the system of 3,300 treaties is de facto becoming multilateralized, but

without the necessary control of an official negotiation [UNCTAD 2016]. As MNEs have the

capacity to locate some of their activities in almost any economy, at least in terms of formal

presence, they can open the door to invoking any international treaty they wish. At the same

time, the even raising the question of a multilateral system regulating FDI is expected to meet

strong political opposition in many countries and social groups of the world, making the need for

muddling through with the current system, perhaps partly reformed, highly probable. However,

the world is at a very initial stage in terms of international treaty reform. There is no consensus

(yet) about what the exact contents of changes should be. It is easy to foresee that, given the

multiplicity of interests, negotiations about common ground may turn out to be arduous.

Conclusions

This presentation has found that we are living in a world of fast growing (although also

volatile) FDI flows. It has become a commonplace to add that the speed of changes is

accelerating. Our vision of global economic geography has to be revised more often than in any

period of the past. It is naturally a challenge for the policy makers who attempt to improve their

country’s competitiveness, which is the basis of job creation and welfare. They have to keep up

with new developments, and should have an unbiased view on them. It means that the

information provided to them has to be truthful and reliable, and should contain indications to

what degree policy makers can influence trends. There is also homework for the international

community: to improve the international regulatory system in order to be able to cope with the

increasing complexity of FDI. In the meantime, FDI continued changing. It may well be that,

some time from now, we will have to report completely new emerging patterns.

References

Altzinger, W. and Bellak, C. (1999): “Direct Versus indirect FDI: impact on domestic exports and

employment” http://papers.ssrn.com/sol3/papers.cfm?abstract_id=230006.

Álvarez, I. and Torrecillas, C. (2013): What does it matter about the home countries of emerging

multinationals? Institute for International Integration Studies, Trinity College Dublin, Discussion Paper 434.

Bartram, S.M. and Dufey, G. (2001): “International Portfolio Investment: Theory, Evidence, and

Institutional Framework” Financial Markets, Institutions & Instruments 10(3): 85–155.

28

Blobel, D.; Görlach, B. and Ingwersen, W. (2012): “Investment: The Context Matters.” In: Ekins, P. and

Voituriez, T. (eds.), Trade, Globalization and Sustainability Impact Assessment: A Critical Look at

Methods and Outcomes, London: Earthscan, pp. 101–122.

Buckley, P. J. (2011): “International integration and coordination in the global factory” Management

International Review 51(2): 269–283.

Bulatov, A. (1998): “Russian direct investment abroad: main motivations in the post-Soviet period”

Transnational Corporations 7(1): 69–82.

Child, J. and Rodrigues, S. B. (2005): “The Internationalization of Chinese Firms: a case for theoretical

extension?” Management and Organization Review 1(3): 381–410.

Coase, R. H. (1937): “The nature of the firm” Economica 4(16): 386–405.

Davidson, W. H. (1980): “The Location of Foreign Direct Investment Activity: country characteristics

and experience effects” Journal of International Business 12: 9–22.

Dunning, J. H. (1977): “Trade, location of economic activity and the multinational enterprise: a search for

an eclectic approach” in: Ohlin, B.; Hesselborn, P. O. and Wijkman, P. E. (eds.): The International

Allocation of Economic Activity. London: Macmillan, pp. 395–418.

Dunning, J. H. (1981): “Explaining the international direct investment position of countries: towards a

dynamic or developmental approach” Weltwirtschaftliches Archiv 119(1): 30–64.

Dunning, J. H. (1986): “The investment development cycle revisited” Weltwirtschaftliches Archiv 122(4):

667–677.

Dunning, J. H. and Dilyard, J.R. (1999): “Towards a general paradigm of foreign direct and foreign

portfolio investment” Transnational Corporations 8(1): 1–52.

Dunning, J. H. and Lundan, S. M. (2008): Multinational Enterprises and the Global Economy, 2nd edition.

Cheltenham: Edward Elgar.

Dunning, J. H. and Robson, P., eds. (1988): Multinationals and the European Community. Oxford: Basil

Blackwell.

Fujita, M. (1995): “Small and medium-sized transnational corporations: Trends and patterns of foreign

direct investment” Small Business Economics 7(3): 183–204.

Fukuyama, F. (1992): The End of History and the Last Man. New York: Free Press.

Gereffi, G.; Humphrey, J. and Kaplinsky R. (2001): “Introduction: Globalisation, value chains and

development” IDS Bulletin 32(3): 1–8.

29

Gokkent, G. M. (1997): Theory of foreign portfolio investment. Ph.D. thesis, Florida International

University.

Grätz, J. (2014): “Russia’s Multinationals: network state capitalism goes global” in: Nölke, A. (ed.):

Multinational Corporations from Emerging Markets: State Capitalism 3.0. Houndmills: Palgrave

Macmillan, pp. 90–108.

Hallward-Driemeier, M. (2003): Do Bilateral Investment Treaties Attract Foreign Direct Investment?

Only a Bit – And They Could Bite. World Bank Policy Research Working Paper 3121.

Hamilton, G., ed. (1986): Red Multinationals or Red Herrings? The Activities of Enterprises from

Socialist Countries in the West. London: F. Pinter.

Head, K. and Ries, J. (2008): “FDI as an Outcome of the Market for Corporate Control: Theory and

Evidence.” Journal of International Economics 74(1): 2-20.

Heckscher, E. F. (1919): “Utrikeshandelns verkan på inkomstfördelningen” Ekonomisk Tidskrift 21: 497–

512. (English version: “The effect of foreign trade on the distribution of income” in: Flam, H. and

Flanders, M. J. (eds.), Heckscher–Ohlin Trade Theory. Cambridge, MA: MIT Press, 1991, pp. 43–69.)

Hymer, S. (1960): The International Operations of National Firms: a study of direct investment. Ph.D.

thesis, Massachusetts Institute of Technology (published in book format in 1976 by the MIT Press,

Cambridge, Massachusetts).

Johanson, J. and Vahlne, J. E. (1977): “„The internationalization process of the firm: A model of

knowledge development and increasing foreign market commitments” Journal of International Business

Studies 8(1): 23–32.

Johanson, J. and Vahlne, J. E. (1990): “The mechanism of internalisation” International Marketing

Review 7(4): 11–24.

Johanson, J. and Vahlne, J. E. (2009): “The Uppsala internationalization process model revisited: From

liability of foreignness to liability of outsidership” Journal of International Business Studies 40(9): 1411–

1431.

Johanson, J. and Wiedersheim-Paul, F. (1975): “The internationalization of the firm: four Swedish cases”

Journal of Management Studies 12(3): 305–322.

Kalotay, K. (2008): “Russian transnationals and international investment paradigms” Research in

International Business and Finance 22(2): 85–107.

Kalotay, K. (2010): “The Future of Russian Outward Foreign Direct Investment and the Eclectic

Paradigm: what changes after the crisis of 2008–2009?” Competitio 9(1): 31–54.

30

Kalotay, K. (2016): “From Triad to Dispersion: the (almost) irresistible rise of foreign direct investment

in new countries” Köz-Gazdaság 11(3).

Kalotay, K.; Éltető, A.; Sass, M. and Weiner Cs. (2014): Russian capital in the Visegrád countries.

Institute of World Economics, Centre for Economic and Regional Studies, Hungarian Academy of

Sciences, Budapest, Working Paper 210.

Kalotay K. and Sulstarova A. (2010): “Modelling Russian outward FDI” Journal of International

Management, 16(2): 131–142.

Krugman, P. (1981): “Intra-industry specialization and the gains from trade” The Journal of Political

Economy 89(5): 959–973.

Krugman, P. (1983): ‘New theories of trade among industrial countries” The American Economic Review

73(2): 343–347.

Luo, Y. and Mezias, J. (2002): “Liability of foreignness: Concepts, constructs, and consequences”

Journal of International Management 8(3): 217–224.

Markusen, J. R. and Venables, A. J. (1998): “Multinational Firms and the New Trade Theory” Journal of

International Economics 46: 183–203.

Mathews, J. A. (2002): Dragon Multinational: a new model of global growth. Oxford: Oxford University

Press.

Moon, H.-Ch. and Roehl, T. W. (2001): “Unconventional foreign direct investment and the imbalance

theory” International Business Review 10: 197–215.

Narula, R. (2010): Keeping the eclectic paradigm simple: A brief commentary and implications for

ownership advantages. United Nations University, Working Paper 031.

Ohlin, B. (1933): Interregional and International Trade. Cambridge, MA: Harvard University Press.

Panibratov, A; Ermolaeva, L.A. and Abramkov, A.E. (2015): “Bremya inostrantsa v protsesse

inernatsionalizatsii rossiyskikh energeticheskikh kompanii” Vestnik St. Petersburskogo Universiteta,

Seria 8 Management 2/2015: 39–73.

Penrose, E.T. (1968): The Large International Firm in Developing Countries: the international petroleum

industry. London: Allen and Unwin.

Poulsen, L. S. (2009): “The Effect of Treaties on Foreign Direct Investment: Bilateral Investment

Treaties, Double Taxation Treaties, and Investment Flows” European Journal of International Law 20(3):

935–938.

Porter, M.E. (1985): Competitive Advantage: New York: Free Press.

31

Ricardo, D. (1817): On the Principles of Political Economy and Taxation. London: John Murray.

Rugman, A. M. (2010): “Reconciling Internalization Theory and the Eclectic Paradigm” The

Multinational Business Review 18(1): 1–13.

Rugman, A. M. and Verbeke, A. (2004): “A perspective on regional and global strategies of multinational

enterprises” Journal of International Business Studies 35(1), 3–18.

Samuelson, P. (1948): “International trade and the equalization of factor prices” Economic Journal

58(230): 163–184.

Samuelson, P. (1949): “International factor-price equalization once again” Economic Journal 59(234):

181–197.

Stoian, C. (2013): “Extending Dunning’s Investment Development Path: the role of home country

institutional determinants in explaining outward foreign direct investment” International Business Review

22(3): pp. 615–637.

UNCTC (1991): World Investment Report 1991: The Triad in foreign direct investment. New York:

United Nations.

UNCTAD (1993): World Investment Report 1993: transnational corporations and integrated

international production. New York–Geneva: United Nations.

UNCTAD (1994): World Investment Report 1994: transnational corporations, employment and the

workplace. New York–Geneva: United Nations.

UNCTAD (1997): World Investment Report 1997: transnational corporations, market structure and

competition policy. New York–Geneva: United Nations.

UNCTAD (1998): World Investment Report 1998: trends and determinants. New York–Geneva: United

Nations.

UNCTAD (1999): World Investment Report 1999: foreign direct investment and the challenge of

development. New York–Geneva: United Nations.

UNCTAD (2001): World Investment Report 2001: promoting linkages. New York–Geneva: United

Nations.

UNCTAD (2009): World Investment Report 2009: transnational corporations, agricultural production

and development. New York–Geneva: United Nations.

UNCTAD (2011): World Investment Report 2011: non-equity modes of international production and

development. New York–Geneva: United Nations.

32

UNCTAD (2013): World Investment Report 2013. Global value chains: investment and trade for

development. New York–Geneva: United Nations.

UNCTAD (2015): World Investment Report 2015: reforming international investment governance. New

York–Geneva: United Nations.

UNCTAD (2016): World Investment Report 2016. Investor nationality: policy challenges. New York–

Geneva: United Nations.

UNCTAD (2017): World Investment Report 2016: Investment and the digital economy. New York–

Geneva: United Nations.

UNIDO (2005): Africa Foreign Investor Survey 2005. Understanding the contributions of different

investor categories to development: implications for targeting strategies. Vienna: United Nations

Industrial Development Organization.

Venables, A. J. (1999): “Fragmentation and multinational production” European Economic Review 43(4):

935–945.

Vernon, R. (1966). “International trade and international investment in the product cycle” Quarterly

Journal of Economics 80(2): 190–207.

Vernon, R. (1971): Sovereignty at Bay: The Multinational Spread of U.S. Enterprises. New York: Basic

Books.

Vernon, R. (1979): “The product cycle hypothesis in a new international environment” Oxford Bulletin of

Economics and Statistics 41(4): 255–267.

Wei, T.; Clegg, J. and Ma, L. (2015): “The conscious and unconscious facilitating role of the Chinese

government in shaping the internationalization of Chinese MNCs” International Business Review 24(2):

331–343.