substitutes if the output is to be sold in the export market.
This is especially true as manufacturing production goes beyond simple non-durable
consumer goods, such as textiles, clothing, and toys, and moves up the product ladder to
more complex goods, such as electronics. It is a bonus, perhaps, that exporting can contribute
to domestic economic growth, but a fundamental motivation for encouraging firms to export
and for rewarding such behavior is the contribution to the essential technological transforma-
tion that the ability to export imparts to the economy. As noted in Chapter 10, openness to the
international economy, which can be measured by either the share of total exports to GDP or
by the share of manufactured exports to total exports, tends to be positively correlated with
economic growth. No doubt the effect of additional exports on total income directly contrib-
utes something to this positive relation through the income multiplier acting on increased
demand. But just as important as the direct export-income link is the domestic technological
learning capacity that the ability to export enforces on producers and which is then trans-
mitted to other sectors of the economy via positive spillover effects that extend the efficiency
gains economy-wide (Edwards 1992; also see Easterly 2001).
It is this positive effect on national technology acquisition which makes manufactured good
exporting so important. A strategy switch toward some kind of export substitution policy by
the state will, if it is to be successful, compel state policy to carefully evaluate its spending
priorities. Greater attention must be paid to human capital accumulation, to the monitoring
of macroeconomic policies, to improvements in the operation of state decision-making and
the civil service, to evaluation of the legal framework, including intellectual property rights,
to the appropriateness of policies to foster private sector initiative, and so on.
In other words, an import substitution-cum-export substitution strategy along the general
lines followed by the East Asian economies (Chapter 10) tends to oblige decision-makers
to continually upgrade the national technological capacity if economic progress is to be
sustained. Of course, government policies may be poorly conceived and the human capital
endowments and incentives may be insufficient for success, so the outcome of setting out on
such a path is not certain. However, for a government to choose export substitution following
an import substitution phase would seem to imply recognition of the need for firms in the
country to have the capacity to export and be efficient. Since that will require managerial,
442 The Process of Economic Development
financial, and technological competence, the future path-dependent direction of the economy
dictated by such a decision is likely to enforce upon the state a greater degree of commitment
to provide the complementary inputs and policies to make such a strategy switch viable.
Macropolicies and technological change
Policies of the state as they affect the economy have an important impact on the level of total
factor productivity (TFP). Policies which create an environment in which private firms are
enabled and encouraged to produce and invest in technological acquisition and to be effi-
cient are likely to have a greater positive impact on output growth than policies that are less
facilitating. In particular, macroeconomic policies can either encourage or discourage private
entrepreneurs to innovate and change. For example, policies such as easy export substitution
that tend to encourage the hiring of relatively cheap labor, increased expenditures aimed at
expanding access to education and better health care all tend to result in the sharing of the
benefits of industrialization among more members of society, thus increasing the internal
market and domestic sales and growth potential. Other policies that help to keep inflation
rates and the balance of payments in check also would be expected to contribute to growth,
national technological change, and development.
Table 13.4 presents some evidence on how macroeconomic policies relative to exchange
rates, discussed fully in Chapter 15, affect growth and technological progress. The table
shows that when exchange rates are over-valued (‚Äúhigh distortion‚ÄĚ), then this tends to reduce
the returns to education as seen in a lower level of GDP growth and, especially, in lower TFP
rates. ‚ÄúBad policy‚ÄĚ (high distortion) thus reduces the efficiency of the economy‚Ä™s inputs,
while ‚Äúgood policy‚ÄĚ (‚Äúlow distortion‚ÄĚ) raises efficiency all around. In fact, bad policy as
measured by over-valued exchange rates, led to zero or negative TFP, regardless of the level
of human capital accumulation.
The table highlights not only the importance of the level of education but also that increases
in the level of education, particularly in an environment of ‚Äúgood policy,‚ÄĚ add significantly
to national technological capacity, as measured by total factor productivity. This accents an
important issue; though human capital investment is necessary, it is not sufficient for higher
Table 13.4 State policy, growth, and TFP
Average annual Average annual
GDP growth TFP growth
Low distortion/high education level
Low distortion/low education level
High distortion/high education level 0.00
High distortion/low education level
Low distortion/high change in education
Low distortion/low change in education
High distortion/high change in education
High distortion/low change in education
Source: World Bank 1991: 47, Table 2.4.
High distortion refers to an exchange rate over-valued by more than 30 percent; low distortion refers to an exchange
rate over-valued by less than 30 percent. High education means an average of more than 3.5 years; low education is
3.5 years average education or less. TFP is ‚Äútotal factor productivity‚ÄĚ which measures the increased efficiency with
which the labor and capital inputs to production are able to be utilized.
Technology and development 443
rates of economic growth. Likewise, good macroeconomic policies are, by themselves,
insufficient to guarantee progress.
Summary and conclusions
Less-developed countries face the demands not only of initiating the fundamental struc-
tural transformation from agriculture to industry. They also must confront the challenge of
creating a ‚Äúnational technological capacity‚ÄĚ and an ‚Äúindependent technological learning
capacity‚ÄĚ (ITLC) as requisites for sustained progress. Much of the difference in incomes
per capita among nations can be explained by the existence of technology gaps as a result of
distinct capacities of different economies to do technology. Closing these gaps often compels
an economy to completely shift its development strategy, to implement new policies at the
macroeconomic level, to re-order spending priorities of the central government, and so on,
all with the purpose of moving the country to a different path over the future.
What happens in an economy today, this month, and this year is path dependent, being
the result of past decisions and particular historical circumstances that affect macroeco-
nomic policies, spending on education, research and development expenditures, the level of
tariffs, the efficiency of domestic entrepreneurs and workers, and so on. This list of contrib-
uting factors could easily be extended, but the point is that past actions condition outcomes
today. To change the rate of economic growth and the level of development in the future,
it is necessary that countries make choices now that will shift the economy‚Ä™s path depend-
ency to a higher efficiency, higher income track. This requires that greater attention be
afforded those factors which can improve the acquisition and adaptation of the world pool
of knowledge to domestic production processes. An emphasis on technical education and
science and mathematics should increase as the level of development increases. Better state
economic policies can contribute in a complementary fashion to raising both private sector
and public sector efficiency by rewarding technological competency, rather than connec-
tions or power.
In the effort to close the technology gap, decision-makers need to keep firmly in mind
that this can be done only by creating a national technology and a national technological
learning capacity. This requires that there be substantial local control over the produc-
tion and learning processes. There must be an emphasis on forging genuine indigenous
technological autonomy in which it is domestic scientists, domestic entrepreneurs, and
domestic skilled workers who become the carriers and agents of technological knowledge.
This knowledge can then be passed on to the next generation of workers who will become
the future R&D scientists and technicians, the future entrepreneurs, and the future skilled
workers and professionals.
This does not mean that each country must be independent of the rest of the world. Just
the opposite is true. There is much to be gained by integrating with the world economy. each
economy must, however, develop the productive independence that comes with creating a
domestic technological learning capacity that can permit it to utilize the world supply of
knowledge for local development needs. Complete technological independence, in the sense
of not making use of knowledge already created in the other countries and available from the
world supply of technology, would be a foolish and unattainable goal for a less-developed
economy hoping to make genuine progress. A genuine independent technological learning
capacity requires domestic inputs and an effort by the state and private firms to attain such
a degree of competence so as to be able to make use of the knowledge created elsewhere,
which can then be applied in ways that suit local conditions, needs, and abilities.
444 The Process of Economic Development
Questions and exercises
1 If technology is not a computer, or computer software, or a new machine, what is it? Are
people part of technology? If so, in what sense? In what ways are technological knowl-
edge and a nation‚Ä™s human capital inputs complementary to one another?
2 What does it mean to say there is not technology but only ‚Äúnational technologies?‚ÄĚ Why
doesn‚Ä™t the same manifestation of technology, say a computer, have an identical produc-
tivity effect in every economy and every setting?
3 If there were not ‚Äúnational technologies‚ÄĚ specific to each country, but rather an inter-
national supply of technology available freely and equally to all countries, would you
expect to find per capita income differences among nations persisting over time? Why,
or why not?
4 Some earlier development economists, like Alexander Gerschenkron, thought ‚Äúlate-
developers‚ÄĚ would have an advantage compared to early developers in increasing their
levels of per capita income since they would be able to use the most advanced tech-
nology without having to re-invent such knowledge themselves.
a Under what conditions would such an optimistic perspective on ‚Äúinternational tech-
nological diffusion‚ÄĚ have validity for a late-developing economy?
b What role does government policy play in creating these conditions and in
augmenting an economy‚Ä™s initial endowments that might make technological acqui-
sition and its use more productive for an economy?
5 Distinguish between an ITLC and an ITCC. What is required for an economy to have an
6 At what stage of industrialization did the East Asian economies achieve an ITLC? Why
do you think this?
7 At what stage of industrialization did the Latin American economies achieve an ITLC?
Why do you believe this?
8 Why are domestic entrepreneurs, domestic capitalists, and domestic scientists and engi-
neers so important to the development process? Why is it difficult, if not impossible,
for foreign inputs to substitute for these domestic inputs without short-circuiting the
9 Why might economically and politically powerful elites in some less-developed coun-
tries be opposed to national technological competence? In what sectors of an economy
might one expect to find supporters of an ITLC?
10 What is ‚Äúfactor displacement?‚ÄĚ How can MNCs cause ‚Äúfactor displacement‚ÄĚ? Would
this be more or less likely to occur if an economy had already created an ITLC?
11 Explain how ‚Äúopenness‚ÄĚ to the international economy, especially a growing capacity for
local firms to export manufactured goods, would be likely to contribute to helping an
economy create a national technological capacity or ITLC.
12 Over the period 1980‚Ä“2003, Lebanon averaged per capita GDP growth of 2.9 percent per
year. Over that same period, total physical capital (K) grew by 8.1 percent per year and the
employed labor force (L) by 2.4 percent. We also know that about 35 percent of Lebanon‚Ä™s
GDP in value-added terms was contributed by physical capital and 65 percent by labor.
a Calculate Lebanon‚Ä™s average TFP over this period.
b explain what the value you calculated means.
c Has Lebanon had extensive and/or intensive economic growth? If there has been
both types of growth, how much of each kind of growth has there been?
Technology and development 445
13 In Malawi, total gross domestic product has increased at an average annual rate of
2.6 percent over the last decade. The annual growth of the labor force has averaged
2.2 percent and the increase in the physical stock of capital has increased an average of
4.5 percent per year. Labor‚Ä™s contribution to total output has averaged 30 percent over
the period. Given these figures, calculate the annual rate of total factor productivity
(TFP) for the Malawi economy.
14 Choose a country in which you are interested or one you are assigned. This problem
will give you more practice in calculating TFP. Go to http://www.worldbank.org, click
on ‚ÄúData‚ÄĚ at the top of the page and then choose ‚ÄúData by Country‚ÄĚ and then choose
‚ÄúAAGs.‚ÄĚ Find your country in the drop-down menu.
a Calculate the TFP for your country for most recent year using: percentage change in
GDP per capita (next to last line in the yellow section), percentage change in gross
capital formation (this is for the K variable; it‚Ä™s the next to last line on the first page
and it is a percentage, too) and the percentage change in labor force (it won‚Ä™t be for
just one year; rather it is shown as an average); this is your L variable (you‚Ä™ll find it
near the top of page 1 of the data). Use 28 percent as the weight for the contribution
of the physical capital input to production.
b How much of economic growth has been intensive and how much extensive?
c What does the TFP value you have calculated mean?
1 In the neoclassical growth model (Chapter 4), technology is modeled as if it is a public good avail-
able to all economies freely and with the same impact on productivity. In effect, the neoclassical
model makes the not very likely assumption that all economies, rich and poor, have exactly the
same aggregate production function. They do not differ in the technology they have available to
them, since technology is a public good. Nor do they differ, apparently, in their innate capacity to
use technological knowledge. In other words, technology is defined as a ‚Äúthing‚ÄĚ that any society
can appropriate without any preconditions. economies differ as to income per capita, then, only in
their level of saving and investment and in their population growth rates, not in the availability of
technology or the specific ability of any country to make use of that technology to its fullest.
In the endogenous growth models (Chapter 8), technology is not viewed as a public good or a
‚Äúthing‚ÄĚ appropriable by any economy without preconditions. Technology (the A(K) in equation 8.3)
differs among economies, even though all potentially can tap into the same world supply of knowl-
edge. The level and pace of technology differ in diverse economies because of dissimilarities in
human capital accumulation and the stock of human capital, because of economic policies of the
state (for example, on inflation or on income distribution), because of different levels of manage-
ment and financial skills of enterprises, and because of a whole gamut of other variables that consti-
tute the ‚Äúpath‚ÄĚ a country has been following to any point in time. In this view, there are only
national technologies, and the level and effectiveness of technology in each economy is dependent
upon the resources each has devoted to technological appropriation.
2 This is not entirely true, as Fagerberg (1994: 1150 ff.) notes. Denison‚Ä™s (1967) early work had
measured a ‚Äútechnological gap‚ÄĚ that at least suggested the importance of country-specific efforts.
The theoretical work on endogenous technical progress by Kenneth Arrow, H. Uzawa, Edmund
Phelps, and Solow himself, among others, predates the current empirical work on the ‚Äúembed-
dedness‚ÄĚ of technological processes within a specific context in each economy. The major early
articles on these themes are conveniently collected in Stiglitz and Uzawa (1969), a reading of which
suggests that economists seem to rediscover old truths from time to time.
3 The pace of technological change is also related to the structural transformations discussed in
Chapters 9‚Ä“11. As a country begins to shift labor from agriculture and other rural production to
urban, industrial pursuits, this typically results in higher productivity because of the higher level
of physical capital and knowledge in use in industry. Thus the structural transformation from
446 The Process of Economic Development
agriculture to industry, often beginning with easy import substitution industrialization, is important
precisely because it sets the stage for the technological transformation that can contribute to the
transition to higher levels of development.
However, the speed at which this technological transformation will be able to progress depends
upon the attention paid to human capital accumulation, to state policies on research and development,
to the macroeconomic environment, to the skills of managers and entrepreneurs, and to a range of
institutional factors that can either support or retard the technological transformation.
4 ITLC involves both ‚Äúknow-how‚ÄĚ and increasing progress on the path of ‚Äúknow-why,‚ÄĚ or ‚Äúdeep
technological learning,‚ÄĚ to use Lall‚Ä™s distinction (Lall 1984: 116‚Ä“17) for what are here called an
ITLC and an ITCC. A schematic representation of the relations might look like the following:
know-why basic scienti c research
5 Remember that the numbers are the number of R&D researchers per 1,000,000 population for the
most recent year during the period specified. To find the total number of R&D researchers, it is
necessary to know the population of each country. For example, if we use Argentina‚Ä™s 1990 popula-
tion of 32.6 million, the total number of R&D researchers can be estimated as 23,472. China had
a smaller number of R&D researchers per million population than Argentina (663 versus 720), but
since China‚Ä™s total population in 1990 was 1,135.2 million, the total number of R&D researchers
was more than 750,000.
6 Both Level 1 and Level 2 economies were further sub-divided into three sub-categories, making for
six levels of technological capacity in the less-developed countries. This sub-division is due to the
interesting work of Weiss 1990.
7 We first encountered the concept of total factor productivity (TFP) in Chapter 8. It is the name now
given to what was called the Solow residual in earlier growth accounting exercises. Calling the
unknown factors that affect economic growth ‚ÄúTFP‚ÄĚ certainly sounds more scientific than referring
to them as a residual, but that does not change the fact that TFP is measuring all the non-identified
determinants of economic growth other than changes in the quantity of the inputs.
8 These are not the only changes that countries in these lagging regions need to undertake. We shall
see in the following chapters that there are other policy failures that can derail progress. However,
building a technological capacity via human capital accumulation and judicious R&D expendi-
tures is necessary for progress over time. Overcoming other policy lapses will be ineffective in
accelerating economic growth and contributing to human development without having put in place
the necessary augmented human capital, R&D, and scientific endowments that contribute to the
achievement of an ITLC.
9 In an interesting study, Grier (2003) argues that only Hong Kong and Taiwan are ‚Äúover-achievers‚ÄĚ
in terms of their economic growth. The other economies, at best, perform as expected if the standard
is how developed economies perform. In other words, TFP values in the region are not as large as
most studies argue. This is a controversial conclusion, but it is worth considering.
10 As pointed out in Chapter 10, there actually exists a quality-price continuum, not just one price/
quality standard for any particular good. If a firm produces a good with low quality by international
standards, it must be priced accordingly low if it is to compete in the international market. If quality
is high relative to international standards for that good, then price could be above the average price.
The point is that there exist a number of price/quality combinations that could make a product
competitive in the world market when sold against similar goods (think Kias, Hondas, BMWs, for
an example of but one price/quality continuum).
Technology and development 447
Adler, Emanuel. 1987. The Power of Ideology: The Quest for Technological Autonomy in Argentina
and Brazil. Berkeley: University of California Press.
Amsden, Alice. 1989. Asia‚Ä™s Next Giant: South Korea and Late Industrialization. New York: Oxford
Bhalla, Ajit S. and Dilmus D. James. 1986. ‚ÄúTechnological Blending: Frontier Technology in Tradi-
tional Economic Sectors,‚ÄĚ Journal of Economic Issues 20 (June): 453‚Ä“62.
Denison, Edward. 1967. Why Growth Rates Differ. Postwar Experience in Nine Western Countries.
Washington, D.C.: The Brookings Institution.
Dore, Ronald. 1984. ‚ÄúTechnological Self-reliance: Sturdy Ideal or Self-serving Rhetoric,‚ÄĚ pp. 65‚Ä“80
in Martin Fransman and Kenneth King (eds.), Technological Capability in the Third World. Basing-
Easterly, William. 2001. The Elusive Quest for Growth: Economists‚Ä™ Adventures and Misadventures in
the Tropics. Cambridge, MA: MIT Press.
Edwards, Sebastian. 1992. ‚ÄúTrade Orientation, Distortions and Growth in Developing Countries,‚ÄĚ
Journal of Development Economics 39: 31‚Ä“57.
Enos, John L. and Woo-Hee Park. 1988. The Adoption and Diffusion of Technology: The Case of
Korea. London: Croom Helm.
Evenson, Robert E. and Larry E. Westphal. 1995. ‚ÄúTechnological Change and Technology Strategy,‚ÄĚ
Chapter 37 in Jere Behrman and T.N. Srinivasan (eds.), Handbook of Development Economics, vol.
3A. Amsterdam: Elsevier Science.
Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge, MA:
Harvard University Press.
Grier, Robin. 2003. ‚ÄúToothless Tigers? East Asian Economic Growth from 1960‚Ä“1990,‚ÄĚ Review of
Development Economics 7(3): 392‚Ä“405.
Grossman, Gene M. and Elhanan Helpman. 1991. Innovation and Growth in the Global Economy.
Cambridge, MA: MIT Press.
‚Ä”‚Ä”. 1994. ‚ÄúEndogenous Innovation in the Theory of Growth,‚ÄĚ Journal of Economic Perspectives 8
Harriss, John, Janet Hunter and Colin M. Lewis (eds.). 1995. The New Institutional Economics and
Third World Development. London: Routledge.
Herrera, Amilcar. 1973. ‚ÄúSocial Determinants of Science Policy in Latin America: Explicit Science
Policy and Implicit Science Policy,‚ÄĚ pp. 19‚Ä“37 in Charles Cooper (ed.), Science, Technology and
Development. London: Frank Cass.
Hirschman, Albert O. 1971. ‚ÄúHow to Divest in Latin America and Why,‚ÄĚ pp. 225‚Ä“52 in A.O. Hirschman,
A Bias for Hope. New Haven, CT: Yale University Press.
Kuznets, Simon. 1966. Modern Economic Growth: Rate, Structure, and Spread. New Haven, CT: Yale
‚Ä”‚Ä”. 1968. ‚ÄúTrends in Capital Formation,‚ÄĚ in UNESCO, Readings in the Economics of Education.
Lall, Sanjaya (ed). 1984. Special Issue on ‚ÄúExports of Technology by Newly-Industrializing Coun-
tries,‚ÄĚ World Development 12 (May‚Ä“June).
Radhakrishna, K. (ed.). 1980. Science, Technology and Global Problems: Views from the Developing
World. Oxford: Pergamon Press.
Sachs, Jeffrey D. 2005. The End of Poverty. London: Penguin Books.
Salter, W.E.G. 1969. Productivity and Technical Change. Cambridge: Cambridge University Press.
Solow, Robert. 1988. ‚ÄúGrowth Theory and After,‚ÄĚ American Economic Review 78 (June): 307‚Ä“17.
Stiglitz, Joseph E. and Hirofumi Uzawa. 1969. Readings in the Modern Theory of Economic Growth.
Cambridge, MA: MIT Press.
UNCTAD (United Nations Conference on Trade and Development). 1996. Fostering Technological
Dynamism: Technology Capacity-Building and Competitiveness. Geneva: UNCTAD.
448 The Process of Economic Development
UNDP (United Nations Development Programme). 1994. Human Development Report 1994. Oxford:
Oxford University Press.
‚Ä”‚Ä”. 1999. Human Development Report 1999. Oxford: Oxford University Press.
‚Ä”‚Ä”. 2006. Human Development Report 2006. Oxford: Oxford University Press.
Wade, Robert. 1990. Governing the Market. Princeton: Princeton University Press.
Weiss, C. 1990. ‚ÄúScientific and Technological Constraints to Economic Growth and Equity,‚ÄĚ in Robert
E. Evenson and Gustav Ranis (eds.), Science and Technology: Lessons for Development Policy.
Boulder, CO: Westview.
World Bank. 1991. World Development Report 1991. Oxford: Oxford University Press.
‚Ä”‚Ä”. 1993. The East Asian Miracle. Oxford: Oxford University Press.
‚Ä”‚Ä”. 2007. World Development Indicators. Oxford: Oxford University Press.
Problems and issues
14 Transnational corporations and
after reading and studying this chapter, you should better understand:
‚ÄĘ the variations in the types of transnational corporations (TNCs): resource-dependent
TNCs, commodity-trade controlling TNCs, ‚Äústand-alone‚ÄĚ branch plants of TNCs
operating under ISI programs, and integrated global production TNCs operating
within core-subcontracting interfirm webs and commodity chains;
‚ÄĘ the quantitative impact of TNCs on capital formation in poor nations;
‚ÄĘ the qualitative impact of TNCs on capital formation, technology spillovers and the
organization of production;
‚ÄĘ the costs of hosting TNCs in terms of transfer pricing, net long-term resource
transfers, and diversion effects of TNCs;
‚ÄĘ how ‚Äúthin‚ÄĚ globalization and weak backward linkages often result from hosting
‚ÄĘ the reasons why hosting TNCs involves poor nations in monitoring environmen-
tally risky and complex production processes;
‚ÄĘ the potential for successful bargaining with TNCs, and the reasons why most host
nations fail to reap the potential benefits of TNC investment;
‚ÄĘ the role of export processing zones (EPZs), and their limited potential for contrib-
uting to successful strategies of development;
‚ÄĘ the impact of EPZs on women workers;
‚ÄĘ why and how the impact of TNC activity has varied in Asia and Latin America.
Transnational corporations (TNCs) are companies operating in two or more nations (with
a significant equity investment of at least 10 percent in a foreign branch plant, subsidiary,
or affiliate). Transnationals are far from being a new or recent element of the structure of
economic relationships which define the less-developed world. In the early colonial period
(Chapter 3), TNCs such as the Dutch East Indies Company and the British East India
Company played a major role in the economic life of Java, India, Holland, and england.
even prior to the Industrial Revolution, these early trading corporations were determined
to reap profits from their near-monopolistic control of certain trade routes and commodi-
ties. However, most of these early TNCs were involved in trade, not in the direct production
of goods. With the onset of the second industrial revolution (1870‚Ä“1910), giant vertically
452 The Process of Economic Development
integrated corporations emerged in many branches of primary production, such as mining,
and tropical commodities, such as bananas and rubber, and oil. Many of these vertically
integrated companies established production and processing sites in the colonial areas, or in
independent but poorer nations, such as in Latin America. These resource-specific transna-
tionals often established a strong political presence, both within their nation of origin and
within the host nation or territory.1
As noted in our earlier discussion of agriculture (Chapter 11), many nations in the less-
developed world, particularly some of the poorest, remain virtual agricultural mono-exporters,
depending for the bulk of their foreign exchange earnings on one or just a few export crops.
Other nations are mono-exporters by virtue of their dependence on the marketing of one or
just a few minerals or oil. In the international market for these commodities, huge buying
TNCs, such as the grain transnational Cargill, or the oil transnationals such as Shell (with
$131 billion in foreign assets and $203 billion in foreign sales in 2004, and 53,000 foreign
employees) or manufacturing transnationals such as G.E. (with $449 billion in foreign assets
and $56 billion in foreign sales, and 142,000 foreign employees) can exert strong pressure
on the producing nations, particularly when they are among a handful of firms that dominate
the buying, transportation, and distribution of these products.
John Cavanagh and Frederick Clairmonte analyzed the state of the global commodities
markets in the early 1980s.
[I]n the last two decades the domination of primary commodity markets has passed from
single commodity traders (e.g., the former United Fruit Company) to firms paramount in
several global commodity markets. The trade in three commodities, by no means excep-
tional, illustrates the dimension of marketing leverage: the trade in bananas, where three
conglomerates dominate 70‚Ä“75 percent of global markets; the cocoa trade, of which six
corporations account for over 70 percent; and the trade in tea and tobacco, where 85‚Ä“90
percent is under the direct control of six transnational leaf buyers.
The market power of the multi-commodity traders stems from their self-reinforcing
modes of conduct that contribute to enhance their bargaining stance vis-√†-vis devel-
oping countries. Most multi-commodity traders are private and largely non-account-
able, not only in developing but also in developed market economy countries. Many
are integrated backward into plantations and forward into processing, and hence are
in an even stronger bargaining position vis-√†-vis national marketing institutions with
which they deal.
(Cavanagh and Clairmonte 1982: 16)
Table 14.1 details the pervasiveness of the near monopoly and oligopoly structures in the
global trade in primary commodities, much of which is controlled by transnationals.
Import substitution industrialization and the TNCs
After the Second World War, a third round of transnational activity began. In their search for
viable development policies, many less-developed nations adopted, as we know, easy import
substitution industrialization (ISI) strategies as the means to initiate the structural transforma-
tion (Chapter 9). The ISI approach effectively locked out of the domestic market the products
of many manufacturing companies of the advanced industrial nations, which, because ISI
relied upon protective tariffs to encourage domestic manufacturing, found it more difficult to
export to less-developed countries as a result. In response, many large manufacturing TNCs,
Transnational corporations and economic development 453
Table 14.1 Transnational control of global commodity trade, 1980
Percent marketed by top 15 TNCsa
Commodity Total exports (billions $)
Wheat 16.6 85‚Ä“90
Agricultural raw materials
Natural rubber 4.4 70‚Ä“75
Hides and skins 2.7
Ores, minerals, and metals
Source: Cavanagh and Clairmonte 1982: 17.
a In most cases, 3‚Ä“6 traders dominate the bulk of these markets.
particularly those domiciled in the United States, reacted, where they could, by ‚Äújumping
the tariff walls‚ÄĚ and setting up ‚Äústand-alone‚ÄĚ branch plants in less-developed nations with
reasonably large domestic markets.
It is conventional to associate easy ISI with nationalist policies adopted by governments
in less-developed economies. And, indeed, during the peak ISI era, from roughly 1946 to
the late 1970s, many nations sought to wrest control from TNCs. Nationalizations, in which
foreign TNCs were taken over and converted to domestic ownership, were widespread from
1960 to 1980, with 587 recorded in various countries. These nationalizations were, however,
largely concentrated in the areas of ore, minerals, and metals, and in food and raw mate-
rials production, that is, they were largely directed at foreign investments which had for the
most part been established during the colonial era in resource-specific or resource-dependent
production. Furthermore, 76 percent of these nationalizations took place between 1966 and
1976, a decade when North‚Ä“South tensions reached a zenith. Not only were the bulk of
the nationalizations concentrated in time, they were also concentrated in place. A study of
seventy-nine nations during the 1960‚Ä“85 period reveals a total of over 300 political regimes
or governments, but a mere twenty-eight accounted for nearly two-thirds of all the expropria-
tions of foreign transnationals (Kennedy 1992: 68‚Ä“9).
Thus, during the time when ISI policies were being implemented, some less-developed
454 The Process of Economic Development
nations did engage in ‚Äúhostile‚ÄĚ acts of expropriation. However, at the same time, the pres-
ence of other TNCs, particularly US TNCs in the manufacturing sector of the less-developed
nations, was growing rapidly. As recent research has demonstrated, the growth of branch plant
TNCs in manufacturing and the spread of easy ISI policies coincided for good reason. The
largest US manufacturing corporations were not, in fact, in opposition to ISI, and the major
less-developed nations were not hostile to manufacturing TNCs, even as they adopted and
strengthened their ISI policies, at least partly to help create a domestic industrialist class.
During and particularly after the Second World War, the leaders of the largest corpora-
tions, and many policy-makers, were deeply focused on the difficulties of constructing a
postwar international economic environment which would not degenerate into the fractious
global struggle that had erupted prior to the Great Depression. Hanging over them was the
realization that industrial capacity had leaped forward in the United States during the war.
Policy-makers believed that a reinvigorated global system would be a necessity, and they
accepted that the larger less-developed countries, as they adopted programs aimed at rapid
industrialization, could help utilize the industrial capacity built up in the United States. As
a result of such concerns, the United States, under President Truman, had promulgated the
Point Four program (see Chapter 3) ‚Äúnot simply as an aid program but an effort to specify
planning goals for Third World development. The numerous economic missions sponsored
under the policy cost the U.S. little, but they had a substantial impact on the direction of less-
developed world economic policy‚ÄĚ (Maxfield and Nolt 1990: 58). During the Truman and
U.S. technical aid missions were sent to most underdeveloped countries to help draft
and implement ISI development plans. These plans specified the tariff, tax, and other
incentives that would channel private investment into the targeted industries. They also
became the basis for allocating U.S. and multilateral aid, and often local development
resources, as well. This effort was an autonomous U.S. initiative; it was not simply a
concession to less-developed world nationalists.
By 1960, US TNCs owned 49 percent of all of the direct foreign investment (FDI) spread
around the globe (see Table 14.2). After the Second World War, some of the new-found
strength of the United States arose as a result of its willingness to champion domestic ISI
industrialization strategies, while simultaneously obtaining a preferred niche in many less-
developed nations for US-based TNCs. Thus, even as the old-style mineral, food, and raw
material TNCs confronted nationalization in the aftermath of often difficult transitions
from colonialism to independence, US manufacturing TNCs were blazing a trail into new
Table 14.2 Share of the stock of world FDI (percentage of world total)
Country 1914 1960 1978 1992 2000 2005
12.2 6.1 3.8 8.3 8.3
10.5 1.2 7.3 9.2 7.4 9.1
0.1 6.8 13.0 4.7 3.6
United Kingdom 45.5 16.2 12.9 11.4 15.0 11.6
United States 18.5 49.5 41.4 25.3 19.2
Source: UNCTAD 1994: 131; 2001: 307; 2006b 303.
Transnational corporations and economic development 455
even though many of the new industrial plants were small, and therefore lacked econo-
mies of scale, they were profitable, because they were able to push prices upward in protected
markets. Competing foreign-made goods often could be obtained only through imports, which
frequently faced tariffs well above 100 percent, so there was plenty of room for increasing
prices. While these new TNCs were frequently the subject of bitter controversy, they were
rarely targets for nationalization. They tended to bring a new and more flexible corporate
culture, adaptable to some degree to the development aspirations of the developing nations.
In contrast, many of the older food, mineral, and raw material producers exuded an intransi-
gent attitude which had usually served them well in an earlier era, but which failed them in
the 1960s and early 1970s.
The globally integrated production system
Beginning in the late 1960s, accelerating throughout the 1990s, and continuing in the
twenty-first century, a fourth form of transnational economic activity could be noted.
Global factories began to emerge, sparked by revolutions in communications, transporta-
tion, and information-processing technologies. Here the motivation for investment was
not the domestic market, nor were the economic activities of these new TNCs resource- or
location-specific. Rather, new manufacturing activities spread in less-developed econo-
mies based upon their cheap labor, the near-absence of environmental restrictions relating
to production activities, the absence of effective unions and labor laws, or other factors
which essentially served to lower the costs of production. In the late 1960s and early
1970s, tariff barriers moved steadily downward, particularly in the developed nations. This
helped facilitate the growth of truly global factories and shifted much of the emphasis in
policy-making circles toward export-led industrialization and away from an over-emphasis
As the so-called East Asian ‚Äútigers‚ÄĚ developed during this period, and as new possibili-
ties for foreign investors and for expanding export activities opened up, ISI strategies came
under attack from economists (as we saw in Chapters 9 and 10). The institutions and tech-
nological processes needed to support the global factory system had not existed in the 1940s
and the early 1950s, nor could they be quickly created. Thus the ISI era has been eclipsed by
the era of globally integrated production as the nature of TNCs has evolved with advances
in technology, from cellular communications to far-flung computer and data entry services.
Without doubt, the age of the computer has accelerated the pace of change and the possibili-
ties at the same time. Peter evans describes the new era:
The ‚Äúnew Internationalization‚ÄĚ which has taken shape between 1973 and the present
represents a different paradigm [from ISI]. Its production strategies are defined by global
markets rather than local ones. Global production networks are typically constructed
around a series of ‚Äústrategic alliances‚ÄĚ among TNCs but occasionally include Third
World entrepreneurial groups. Manufactured exports from the Third World back to rich
country markets are central to the new paradigm, while flows from the advanced coun-
tries to the Third World increasingly take the form of services and intangibles. The new
internationalization pervades all regions of the Third World but East Asia, not Latin
America, is the archetypal site.
(Evans 1998: 197)
Why did the global factory system arise? Theorists have debated this point thoroughly.
456 The Process of Economic Development
Some have emphasized ‚Äúindivisibilities‚ÄĚ of management and technologies, patents, and
trademarks which can best be exploited by siting production and/or distribution facilities in
more than one nation. Richard Caves has suggested that the motivation to ‚Äúgo global‚ÄĚ often
arises from the fact that firms own or control specific production processes, designs, styles,
and other types of know-how. These ‚Äúintangibles‚ÄĚ are difficult to price, to divide up, and are
difficult to sell or license to other firms. Usually there is no orderly market, worldwide, for
such in-house assets. Consequently, if these assets are not fully utilized within the domestic
market, and/or if they can be adapted to higher production levels without incurring prohibi-
tively rising unit costs, these firms can often increase their profits by operating additional
production facilities abroad. Such firms might prefer to sell to foreigners these intangible
assets to earn profits, but normally they cannot. Thus, their deeper utilization via foreign
investment is one option; the alternative may be to completely forgo economic rents arising
from the exploitation of the intangible assets they have created (Caves 1991).
Other theorists have emphasized the changing global conditions external to the individual
firm, which have encouraged global production. For example, the average cost of ocean
freight plus port fees dropped by more than 50 percent between the late 1940s and 1990. The
price of an international call from New York to London fell tenfold between 1970 and 1990.
Airline fees per passenger mile decreased even faster and further than did ocean freight costs
from the late 1940s to 1990. And satellite utilization charges had declined to one-tenth their
1970 cost by 1990 (World Bank 1995: 51). Moreover, the firm-specific theory of Richard
Caves and the perspective which emphasizes what might be called the global infrastructure
approach are not mutually exclusive. Both contribute to our understanding of recent trends
in expanding TNC production.
A third group of theorists emphasize mega-changes in production systems, such as the
‚Äúfactory-of-the-future,‚ÄĚ where modern machinery can perform complex tasks previously
necessitating skilled and experienced shop-floor technicians and workers. Deskilling of
the labor force as a result of new production machinery and techniques has allowed many
firms to retain a core group of technicians and managers in the industrial countries, while
outsourcing other production to less-skilled workers in the less-developed nations.
For example, Nike, a US-based TNC with annual sales of $4 billion in 1993, directly
employed 9,000 highly skilled workers involved in product design, data processing, sales,
administration, product development, production design, marketing, and distribution. Nike
also employed 75,000 workers via independent subcontracting (offshore outsourcing)
arrangements in China, South Korea, Taiwan, Indonesia, Malaysia, and Thailand, where
labor-intensive production and assembly processes actually produce the final Nike product
(UNCTAD 1994: 193). By 2007 Nike‚Ä™s subcontractors employed more than 800,000 workers
in more than 700 plants, with nearly 60 percent of all these workers in East Asia, above all
China (see Focus 14.1).
The activities encompassing FDI are distinct from offshore outsourcing, which is growing
rapidly, particularly in the area of services. Outsourcing is also distinct from foreign trade,
yet it is not as embedded in the developing nation‚Ä™s productive structure as is FDI. It is there-
fore a separate category, yet closely tied to profit-maximizing international strategies of the
large corporations in the advanced industrial nations. Offshore outsourcing has important
impacts on developing nations, but these effects are largely defined in terms of employment
and working conditions. Outsourcing often occurs in Free Trade Zones (FTZs), alternatively
known as Export Processing Zones (EPZs). These zones are examined later in this chapter.
How does the globally integrated production system function? Since the early 1980s, but
with increasing emphasis, TNCs have turned to new forms of production, sometimes called
Transnational corporations and economic development 457
FOCUS 14.1 SUbCONTRACTING IN INDONESIA
Subcontracting (offshore outsourcing) has become an important strategy increasingly
adopted by transnational corporations in the 1980s and 1990s. As labor costs have risen
in Korea, Taiwan, and Singapore, many East Asian subcontractors have become small
transnationals themselves, shifting their production and assembly operations away from
nations such as Korea, to new low-wage havens, such as Indonesia.
Depending on market demand, Nike Corporation contracts with four to six Indonesia-
based Korean subcontractors who employ roughly 5,000 Indonesian workers. According
to the Nike Corporation, the advantage of locating in Indonesia is that a pair of shoes
selling for $80 in the United States will involve direct labor costs in Indonesia of only $2.60.
Thus there is greater opportunity for more profit.
Although the subcontractors who sell their output to Nike are pressured to pay a
minimum wage of $52.50 per month, they often do not. For example, one twenty-two-
year-old worker and his nineteen-year-old wife, both employees of a Nike subcontractor,
earned a total of $82 per month. They rented one room for $23 per month. This housing,
actually a six-foot-by-six-foot space, is described as follows:
A single bare bulb dangles from the ceiling, its dim glare revealing a plain bed, a single
gas burner, and a small plastic cabinet. Their room, one of a dozen in a long cement
building, is provided with one container of water daily. If they want more water, each
jug costs about 5 cents.
Attempts by the workers to raise their pay above the current average of $2.62 per day
have met with harsh treatment from the government. Independent unions and the right to
strike are not recognized by the government.
Source: Gargan 1996
flexible manufacturing, or ‚ÄúJapanese management techniques‚ÄĚ. While no two firms operate
in an identical manner, many theorists of industrial organization believe that a blending of
‚Äújust-in-time‚ÄĚ inventory controls, total quality management techniques (JIT/TQM), and
small-batch flexible production based on computer-aided designs and computer-aided manu-
facturing techniques (CAD/CAM), and, generally, ‚Äúlean‚ÄĚ production techniques capture the
recent tendencies in worldwide best-practice production technology. In a study examining
the growth of new production forms in the less-developed world, John Humphrey defined
‚Äúlean‚ÄĚ production as having the following attributes:
Lean production involves three related transformations ‚Ä“ the reorganization of produc-
tion along JIT/TQM lines, the transformation of design, and the development of new
relations with suppliers. It focuses not only on the factory, but also factors outside of
the plant such as the design function and other firms. These interrelated changes are
held to create a new production system, based on principles which contrast with [early
twentieth-century assembly-line] mass production. ‚Ä¦
The core of reorganization within the firm along the lines of JIT/TQM can be captured
in the term ‚Äúminimum factory.‚ÄĚ ‚Ä¦ The aim of manufacturing production is to produce
goods which satisfy the customer at the minimum possible cost. The ideal factory should
have every stage of production oriented toward this overall aim, and all activities in
the plant should contribute to transforming inputs into finished products which attend
customers‚Ä™ needs. Any other activities are, in principle, a waste of resources. Such waste
includes holding stocks, moving products around unnecessarily, producing items which
458 The Process of Economic Development
are defective and reworking products. The ideal factory responds rapidly to customer
demands, producing rapidly a range of products which satisfies customers‚Ä™ needs with
the minimum possible inputs of energy, materials, capital and labor.
(Humphrey 1995: 150, 152)
JIT/TQM techniques are now being adopted widely, particularly in the newly industrial-
izing countries (NICs) and above all in Asia, but also in countries such as Zimbabwe and the
Dominican Republic. Unfortunately, it presently appears that successful adaptation to the new
techniques seems to be the exception. Most countries, including some of the first-tier NICs,
such as Brazil and Mexico, are not able to compete effectively on the basis of their emerging
production systems. Clearly, cases of relatively successful adaptation are to be found, such as
the TNC automobile plants in Mexico, but the likelihood of most nations being passed over
by such technology is quite high. For, unlike the early ISI stage, the demands for JIT/TQM
global factory production systems extend much deeper into the economic fabric of a less-
developed nation. Early ISI branch plants were to a large degree ‚Äústand-alone‚ÄĚ operations,
importing some of their inputs, and relying on the parent company for design and production
techniques, but largely operating independently from the parent on day-to-day production
Under the emerging system of ‚Äúglobally integrated production,‚ÄĚ there is a greater reliance
on a web, or network, of sophisticated suppliers which must be close at hand or ‚Äújust-in-
time.‚ÄĚ This means that FDI intended to meet the standards of JIT/TQM production must be
embedded in a network of sophisticated supplier firms offering a full range of inputs, as well
as upstream services such as delivery, marketing, and transportation. As the list of items that
must be provided outside of the factory grows, the likelihood that a given less-developed
nation has the deep, complex, and quickly adjustable production support system to meet
the expectation of the TNCs declines. In fact, some case-studies in the early twenty-first
century show that TNCs are being supplied more and more by other TNCs as the demand for
sophisticated parts escalates (Ivarsson and Alvtam 2005). Thus, it is feared that these new
production systems will be adapted by a small number of NICs which are already relatively
developed, leaving the rest of the less-developed world at risk of being further marginalized
from ‚Äúdeep integration‚ÄĚ and advanced production techniques. Hence, JIT/TQM techniques
may tend to reinforce the tendency toward ‚Äúcumulative causation‚ÄĚ and ‚Äúbackwash effects‚ÄĚ
discussed in Chapter 6.
This, too, is the tentative conclusion of the United Nations, which examined TNCs as
‚Äúengines‚ÄĚ of growth. Clearly, it is too soon to draw definitive conclusions regarding the
new production systems. But just as clearly, the developing nations now face new and quite
formidable challenges if they seek to attract and manage FDI and foreign capital in an ever
more technological age. Evidence from the auto sector in Brazil and Argentina suggests that
as integrated production systems expand, the role of local suppliers contracts and foreign-
owned supplier firms arrive to service the high value-added supply needs of the TNCs:
‚Äú[D]omestic firms in developing countries supplying to affiliates that are part of integrated
production systems typically belong to a lower tier and provide relatively simple inputs ‚Ä“
cardboard focuses, plastic and foam rubber packaging materials, metal stamping, die-making
and simple assembly‚ÄĚ (UNCTAD 2001: 137).
As a result of a complex process involving either the search to maximize the return on
intangible assets, or the opportunity to take advantage of a global infrastructure, or to reap the
advantages of labor deskilling and the potential gains by offshore outsourcing of production,
by 1998, the transnationals of the advanced industrial nations were employing 19 million
Transnational corporations and economic development 459
workers (a 58 percent increase since 1992!) in the less-developed world, most of whom were
working in global factories. Overall employment by transnationals (including in the advanced
industrial nations) has grown from 19.5 million in 1982 to 62 million in 2005. From 2004 to
2005 employment jumped by 2.5 million. Employment by the transnationals in developing
nations in 2005 produced more that $500 billion in value-added (UNCTAD 2006b: 9‚Ä“10).
However, these figures do not include offshore outsourcing activities, such as those conducted
by Nike described in Focus 14.1. Including subcontracting, the World Bank estimated that the
direct employment effect of TNCs could be 100 percent higher (World Bank 1995: 62).
If these estimates are valid, it may be that in the early twenty-first century through direct
employment in transnational subsidiaries or affiliates and through offshore subcontracting
the TNCs are now employing approximately 40 million. Assuming job multiplier effects
wherein for every job created one other job is indirectly created, this would bring the total
effect to (very roughly) ¬± 80 million jobs created by the TNCs in the developing nations.
even after making allowances for the fact that many of these workers will be better paid
than the average employee in a developing nation the impact of TNCs on the employment
problem in developing nations has to be very modest. The International Labour Organiza-
tion‚Ä™s estimate of the economically active population in developing nations in 2007 was 2.53
billion people. In this context, it is clear that transnational production, whatever its collateral
effects (discussed below), is not significant in addressing employment problems in devel-
Today, transnationals are to be found in most less-developed nations, their presence being
felt under the four broad categories of transnational activity:
1 trading companies controlling the marketing process;
2 resource-intensive vertically integrated transnationals;
3 branch manufacturing plants; and
4 global factory production sites.
In any given nation, it is likely that more than one of the above forms of transnational activity
will be present.
Foreign direct investment
Foreign direct investment (FDI) entails the ownership of productive assets by a parent
corporation in another nation. Such ownership should be distinguished from the purchase
of foreign stock or the lending of funds to foreign companies and governments. These latter
forms of investment are known as portfolio investments.
The World Investment Report 2006 recorded that the advanced industrial nations oper-
ated some 55,500 transnational corporations and there were approximately 773,000 foreign
affiliates spread throughout the world, almost half of them in developing nations (UNCTAD
2006b: 271). Within these same industrial nations there was an accumulated stock of
$7.1 trillion of foreign capital in 2005. (The stock is a measure of the current market value of
all previous foreign direct investments.) The developing nations had accumulated a stock of
$2.8 trillion of FDI: $1.55 trillion in East and South Asia, $937 billion in Latin America, and
$151 billion in Africa. Overall, the stock of FDI has grown at an impressive rate: in 1990 the
FDI stock/World GDP ratio was 8.5, leaping to 22.7 by 2005 (UNCTAD 2006b: 307).
While nearly 28 percent of the total stock of FDI of the developed nations was invested
in the less-developed world, in recent years the flow, that is, the annual change in the stock,
460 The Process of Economic Development
of FDI has increasingly been directed to the less-developed world. In the three-year periods
1978‚Ä“80, 1988‚Ä“90, and 1998‚Ä“2000 nearly one of every five dollars of FDI went to devel-
oping nations. In the 2003‚Ä“6 period, marked by unprecedented cross-border mergers and
acquisitions among the developed nations, an estimated 34 percent of these flows went to the
developing nations (UNCTAD 2006b: 7).
A much more meaningful method to measure the trend of FDI flows is the ratio of FDI
flows to GDP in the developing nations: this ratio has risen from 0.79 percent of GDP in
1975 to 2.34 percent in 2000 for a broad sample of fifty-nine representative developing
nations (Ghose 2004: 38). For all developing nations in 1990 the stock of FDI equaled 9.8
percent of GDP, while in 2006 the stock of FDI had risen to 27 percent of GDP (UNCTAD
2006b: 308). In short, FDI is a leading variable and has been for more than one-quarter
of a century.
In any given year, a relative handful of the developing nations receive the vast bulk of the
FDI. In the period 1982‚Ä“92, for example, the top ten nations in a given year received over
70 percent of all the direct capital flows to the less-developed world. While there was some
variation in the top ten nations over the ten-year period, eighteen nations have accounted
for over 80 percent of such investment. With few exceptions, these eighteen nations were
not the poorest; in Africa, where per capita income is extremely low, only Egypt, Nigeria,
and Tunisia were among the top eighteen recipients of FDI. In 2000 the top ten recip-
ient nations ‚Ä“ four in Asia ‚Ä“ received 76.7 percent of all FDI going to developing nations
(UNCTAD 2001: 52). In 2005 the top five (nearly a constant group since 1996) received 48
percent of total flows. International investments tended to flow to countries where economic
growth already was taking place.2
Of the 55,500 TNCs with parent companies based in the developed nations, the top 100
firms accounted in any given year for roughly two-thirds of new FDI. In terms of asset
size, the most important foreign productive activity of the top 100 firms is in the electronics
sector, followed by mining and petroleum, motor vehicles, and chemicals and pharmaceuti-
cals (UNCTAD 1994: 10).
While US-based TNCs continue to dominate global production patterns, their relative
significance has changed rapidly in recent years. Table 14.2 shows that while in 1960 US
TNCs accounted for 49 percent of the total stock of FDI, by 1992, this share had shrunk to
just 25 percent, falling yet again by 2005 to 19 percent. For a while, most of the relative gain
was made by Japan, whose share increased from merely 0.7 percent in 1960 to 13 percent
in 1992 ‚Ä“ but then stagnation pushed Japan by 2005 to a relative level not seen since the
The European Union in 2005 held 51.3 percent of the accumulated stock of FDI (up from
45.2 percent in 2000). This increasing national diversity of FDI sources tends to strengthen
the relative bargaining position of host nations vis-√†-vis the TNCs, since they are likely to
have to deal with more than one potential group of national investors for a given project and
that is, as we shall see, critical to the host country if it is to reap the potential benefits of TNC
While trade between countries has historically received a great deal of attention in devel-
opment economics (see Chapter 16), it is important to keep in mind the fact that the sales of
foreign affiliates of the TNCs now exceed the total value of all exports of goods and services
for the entire world (UNCTAD 1994: 20). In 2005, TNCs had sales from their foreign affili-
ates of $22.2 trillion, while the total value of world exports of goods and services stood at
$12.6 trillion (UCTAD 2006b: 9). Considering that global exports have grown rapidly in
recent years, it is remarkable that while in 1982 sales of foreign affiliates were roughly equal
Transnational corporations and economic development 461
to exports of goods and services and now the relationship is approaching a 2:1 ratio. About
one-third of all TNC sales were intra-firm trades, a fact of some importance in our discussion
of transfer pricing later in this chapter (UNCTAD 1994: xxi).
In the view of many specialists, the role of the TNC has now eclipsed that of foreign
trade as a factor determining the overall evolution of the global economic system. When UN
researchers combined the domestic production of TNCs with their international production
and sales, they arrived at the estimate that one-third of all global output is now under the
direct governance of the TNCs (UNCTAD 1994: 135). Never, they stated, had the influence
of the TNCs been greater. Surely, it is much more so today.
Who in the less-developed countries gains from FDI?
In neoclassical economic thinking, as characterized by the Solow, savings-centered theory,
developing nations are viewed as deficient in physical capital investment. Consequently, it
would appear that inflows of FDI could only have a positive effect on the growth rate of a
poorer nation. On average, however, new FDI amounted to less than 3 percent of total invest-
ment throughout the less-developed world in the period 1980‚Ä“92.
There are important exceptions, of course; Hong Kong received more than 10 percent
of its investment from foreign sources during this period, Singapore over 20 percent, and
Malaysia over 6 percent. From 1993‚Ä“9 reliance on FDI as a source of capital formation
increased: In Asia it averaged 8.5 percent, in Africa 7.5 percent, and in Latin America 14.1
percent (UNCTAD 2001: 19, 24, 29).
The higher level achieved in the 1990s suggests that FDI can make a contribution to the
overall investment level in many developing nations. For the years 2003‚Ä“5 the average level
of FDI expressed as a percentage of new capital formation (machinery and equipment, resi-
dential and non-residential construction) was 10.9 percent. Still, since this investment flow
is concentrated among a few nations, most developing nations will receive only modest
economic stimulus from FDI.
Furthermore, even when a nation is receiving large amounts of FDI, such flows do not
always result in new capital formation. Existing plant and equipment are quite often the
target of TNC investments through mergers and acquisitions, reducing the ownership and
control of domestic capitalists. This effect has been referred to as denationalization, that is,
the transfer of ownership of local capital to foreign capital owners (Gereffi and Evans 1981).
While Mexico was considered one of the top receivers of FDI in 2001 71 percent of such
investment was devoted to the purchase of already existing Mexican companies (Gazc√≥n
2002: 15). More generally, according to United Nations data, 20 percent of the FDI flows to
developing nations were used to buy out existing companies in 2004.
Even if a so-called ‚ÄúGreenfield‚ÄĚ facility, that is, a new plant, is built by an TNC, this
need not necessarily increase the total level of investment of a less-developed nation on
a one-to-one basis. Ajit Ghose‚Ä™s study of fifty-nine representative developing nations in
the 1975‚Ä“2000 period demonstrated significant ‚Äúcrowding out‚ÄĚ as foreign investors raised
the price for inputs (and/or caused currencies to appreciate) or rapidly captured significant
market share. This adversely affected nationally owned businesses that then cut back on their
investments (Ghose 2004: 22)
Still, Ghose found that FDI did raise investment levels, but that it would be wrong to
assume that this occurred on a one-to-one basis. That is, the negative effect on domestic
producers (crowding out) was not so large as to negate all the impact of FDI, so that net
investment went up moderately. In such cases, foreign FDI does not completely act as a
462 The Process of Economic Development
complement to local investment, promising to increase the rate of growth, but rather, in part,
as a substitute for local capital ownership, local control, and perhaps for local learning.
In a large study over the 1971‚Ä“2000 period, Manuel Agosin and Roberto Machado found
evidence of crowding out in some sub-periods in some developing regions (Africa and Latin
America) and a ‚Äúneutral‚ÄĚ effect in Asia. FDI failed to stimulate ‚Äúcrowding in‚ÄĚ of domestic
investment, but then again it did not crowd out investment in Asia (Agosin and Machado
2005). The authors speculate that the reason why crowding out did not occur was that the
Asian nations as a group were more aggressive in selecting foreign investors and more prone
to impose conditions on these investors that would guard against adverse affects.
How important is it, then, if a developing nation receives an average injection of invest-
ment (equal to 10 percent of its investment level) from foreign investors? In the best-case
situation, where the impact of FDI is neutral, it is likely that the overall impact of FDI on
GDP growth will be small, as can be demonstrated using some concepts of early growth
In Chapter 4, the ideas of Roy Harrod and Evsey Domar were introduced. Harrod and
Domar‚Ä™s simple Keynesian growth model can be shown to arrive at the following set of rela-
tionships: a nation‚Ä™s growth rate is the result of two key ratios ‚Ä“ the investment/GDP ratio
(I/Q) and the capital‚Ä“output ratio (K/Q). K/Q shows how much output is generated in a year
from a given stock of capital (K).
K/Q varies from country to country and over time, but a workable ratio for many devel-
oping nations would be within the range of 3 to 4. K/Q measures the ‚Äúproductivity‚ÄĚ of capital
and the ‚Äúcapital-intensity‚ÄĚ of production. Ghose‚Ä™s data for the fifty-nine representative
developing nations suggests that the I/Q ratio averages approximately 22 percent, but varies
widely from year to year.
Combining all of the above with the Harrod-Domar formula for economic growth we have
‚ąÜ Q/Q = I/Q √— Q/K (14.1)
This simple formulation assumes that the higher the rate of investment and the lower capital‚Ä“
output ratio, the higher the growth rate (‚ąÜ Q/Q). An investment rate of 22 percent and a
capital‚Ä“output ratio of 3.5 will, according to the H-D equilibrium conditions, generate a
growth rate of (.22 √— 1/3.5 = 6.3 percent growth in GDP).
But how much difference does FDI make? FDI adds perhaps 10 percent to the level of
funds invested, but then ‚Äúcrowding out‚ÄĚ perhaps subtracts away one-half of this. In other
words, had there been no FDI the investment rate would have stood at 95 percent of what it
actually turned out to be. Or (.22 √— .95 = .209 √— 1/3.5 = 5.97 percent growth in GDP).
From this simple, but not distorted, example it can be readily granted that FDI, taking
into account a reasonable ‚Äúcrowding out‚ÄĚ effect, will have a modest positive quantitative
effect on growth ‚Ä“ but it is hardly the panacea that it is sometimes represented as being. Two
final points must be taken into account: First is the fact that a significant portion of FDI will
likely not add to the growth-generating capabilities of a developing nation, because luxury
housing, the construction of sumptuous hotels and apartment buildings and palatial shopping
centers all can be registered as FDI if they are financed abroad. A second twist is that all
retained earnings by foreign corporations (profits not shipped abroad to the parent corpora-
tion) are registered as FDI. Yet it is common for TNCs to place these funds in the financial
markets of the developing nation, where it is virtually certain such funds will not be used for
Transnational corporations and economic development 463
If, with few exceptions, new FDI constitutes only a modest portion of total physical capital
formation in any given year in less-developed economies, why has the role of the TNCs been
so central to much of the recent development literature? The answer, clearly, must turn on
considerations of the qualitative effects of FDI, rather than on considerations of quantitative
Collateral effects of TNCs: the modernization perspective
It has been suggested that TNCs, at least those associated with the ISI era and those that
form part of the globally integrated system of production, can generate potential resource
transfer effects through their activities in the forms of new capital, new product and process
technologies, and management/labor-training and other organizational innovations. Propo-
nents of this modernization perspective believe it is largely a fruitless exercise to attempt to
‚Äúunbundle‚ÄĚ the multiple interactive stimuli that accompany FDI.
But the modernization perspective rarely focuses on the important linkages between FDI
flows and flows from stock and bond purchases (portfolio investment), bank and corporate
lending (private loans), bilateral aid institutions (such as the United States Agency for Inter-
national Development), and multilateral institutions (such as the World Bank or the regional
development banks) which make official loans and grants. It is important to be aware of
such linkages, however, since nations that maintain a so-called good investment climate
toward FDI, that is, a pro-business environment, will often reap substantial indirect resource
transfers via such inflows of capital. In this section, we explore some of the research find-
ings of those who analyze FDI from the modernization perspective, a view which focuses
on the potential growth-stimulating role which TNCs may play through capital formation,
technology transfers, and superior organizational production structures.
To begin, Table 14.3 summarizes the distribution of capital and other financial flows in the
1991‚Ä“2005 period. In observing this data it should be stressed that FDI amounted to only
26 percent of the net long-term flows to developing nations in the 1986‚Ä“90 period. Note
that, in the last row of the table, FDI‚Ä™s share of total flows jumped upward all through the
1990s, reaching a truly remarkable level of 83 percent of total flows in 2001, after which as
a result of surging portfolio investments it fell back to 53.5 percent of all flows. Heretofore
Table 14.3 Net long-term resource flows into developing regionsa (billions of dollars)
Type of flow 2005b
1991 1997 1999 2001 2003
Foreign Direct Investment 35.7 168.7 183.3 176.9 161.6 237.5
21.5 71.7 41.7 11.1 45.7 116.4
5.0 44.0 ‚ą’7.1 ‚ą’10.8 9.8 67.4
Official flows 62.6 34.9 40.2 35.8 29.0 22.3
124.2 319.3 258.1 213.0 246.1 443.6
FDI/total (%) 28.7 52.8 71.0 83.1 65.7 53.5
Source: World Bank, Global Development Finance, 2002: 3 (www.worldbank.org/prospects/gdf2002/; 2006).
a Excludes International Monetary Fund financing on account of its short-term nature.
b World Bank estimate.
c Includes net purchases of securities by foreigners in domestic stock markets and bond financing as well as ‚Äúother‚ÄĚ
464 The Process of Economic Development
developing nations relied heavily on (a) ‚Äúofficial loans‚ÄĚ from the World Bank and the
regional development banks (such as the Asian Development Bank), (b) grants, and (c) bank
lending. But in the 1990s these sources of funding atrophied and have fallen ever since
(‚Äúofficial‚ÄĚ forms of lending are analyzed in Chapter 17).
Portfolio investment soared until the Asia crisis of 1997, while FDI increased more than
fourfold in the period 1991‚Ä“9, and has increased 30 percent in the period 1999‚Ä“2005. Nearly
70 percent of FDI went to either East Asia (32 percent) or Latin America (35 percent ‚Ä“
largely thanks to massive privatizations), with the Middle East attracting only 3 percent and
Sub-Saharan Africa receiving only 6 percent in the 1990s.
Table 14.3 demonstrates the rising importance of private sector flows, particularly those
of transnational corporations, from an upward spiral in investment projects driven by their
new enthusiasm for globally integrated production systems. Yet the poorest nations rarely
received any FDI ‚Ä“ in 2000 the forty-nine least-developed nations obtained a mere 0.3
percent of total FDI (UNCTAD 2001: 1).
While the absolute amount of capital formation provided by TNCs, in relation to total invest-
ment in a developing nation, is likely to be small, probably below 10 percent, the quali-
tative significance of such investment can be much higher than one might assume. Such
investment is often directed to a narrow range of industries that are important for economic
growth, precisely because they are concentrated in manufacturing and services where new
investment is associated with significant increases in productivity and production. Aggregate
data can distort or even hide the role played by TNCs, a point well illustrated through the
following summary of research on TNCs in India.
In India, a study of 28 manufacturing industries in 1977‚Ä“78 found that in nine indus-
tries, including motor vehicles, electrical machinery, metal products, plastics, chemicals
and pharmaceutical, the foreign ownership share was greater than 20 percent. A second
study of TNCs in India found that foreign-owned firms accounted for more than 30
percent of sales in manufacturing in 1975‚Ä“76 and 1980‚Ä“81. Foreign direct investment as
a share of gross domestic investment has been very small in India, at 0.1 percent in the
period 1976‚Ä“80 and only slightly higher ‚Ä“ 0.2 percent ‚Ä“ in the late 1980s.
(UN Transnational Corporations and Management Division 1992: 119)
Furthermore, aggregate FDI data fail to reveal non-equity arrangements within the host
nation which can have a substantial impact on productivity and output. Among such deals,
often numerous, one finds franchising agreements, licensing, long-term subcontracting, and
non-equity joint ventures with local capital. Any of these production linkages with TNCs
may form a conduit for the diffusion of product and/or process technologies and manage-
ment/labor-skills, as well as providing opportunities for learning about more advanced
organizational structures in marketing, advertising, finance, or research and development. It
is often the case that indigenous business owners lack, as a group characteristic, an export
culture, whereas manufacturing TNCs are likely to be quite proficient and dynamic in foreign
marketing, almost by definition. To the degree that such skills are ‚Äúspun off‚ÄĚ to the host nation
through joint ventures and/or the turnover of management personnel working for TNCs or
other transmission mechanisms, TNCs can be an important mechanism for augmenting and
enhancing the proficiency of domestic managers, professionals, and perhaps skilled workers.
Transnational corporations and economic development 465
Since the scarcity of foreign exchange is likely to form a crucial bottleneck for many
less-developed nations, curbing their development potential, the ability to expand into foreign
markets can be of paramount importance to the nation. The domestic market is too small in
many developing nations as a result both of the highly unequal distribution of income and
the size of the population. Thus, foreign sales can be important in ‚Äúwidening‚ÄĚ the market
and in allowing firms to realize greater economies of scale in production. Such effects, when
they do occur, can further assist the development process by lowering production costs,
possibly further expanding the market as lower-income consumers could be brought into
the market if lower costs translate into lower prices. Thus, to the extent that TNCs transmit
export skills to domestic producers, this is another potential benefit to the host country of
Table 14.4 reflects the new emphasis on global production, as well as the success of some
Asian nations in building their own complex production facilities. Developing nations‚Ä™
exports as a share of world exports have grown at a tremendous pace, reaching 39.8 percent
in 2006 (UNCTAD 2006‚Ä“7: 10). Notice the dramatic drop in primary commodities exports,
from over one-half of the non-fuel exports in 1980 to a mere 19 percent in 1998. Since then
primary commodities have risen rapidly, reflecting the ‚Äúresource boom‚ÄĚ that accelerated in
the 2003‚Ä“7 period, partly because of high commodities demand by China and India.
The generalization that many developing nations find much of their economy dependent
upon commodities exports should not be abandoned. We also see that relative to the overall
level of world trade in 1998 (19 versus 14.8 percent) developing nations as a whole are
much more reliant on commodity exports than were developed nations in 1998. Low-end
labor-intensive production processes (assembly operations occurring commonly in Export
Processing Zones) and resource-based manufacturing processes account for slightly less of
total developing nations‚Ä™ exports in 2003 than in 1998 and 1980.
Note the dramatic rise in exports for medium- and high-skill production processes.
Medium-skill exports more than doubled their share of total developing nations‚Ä™ exports
in only eighteen years (1980‚Ä“98), while a surprising 34 percent of all developing nations‚Ä™
Table 14.4 Developing nations‚Ä™ exports: skill level, capital intensity (percentage share)
Share of exports from Share of world exports
1980 1998 2003 1980 1998 2006
‚Ä“ ‚Ä“ ‚Ä“ 15.4 24.3 39.8
Developing nations‚Ä™ exportsa
50.8 19.0 23.9 25.7 14.8 ‚Ä“
21.8 23.2 14.7 15.0 ‚Ä“
Resource-based and labor-intensive 20.7
5.8 7.3 5.6 10.1 7.6 ‚Ä“
Low-skill and technology-intensive
16.8 15.6 26.4 29.6 ‚Ä“
Medium-skill and technology-intensive 8.2
11.6 31.0 34.2 30.2 ‚Ä“
High-skill and technology-intensiveb 20.2
Source: UNCTAD 2002: 55, 68; 2005: 150.
a Included are 225 production categories including raw materials, excluding fuels.
b Included are electronics products, parts and components ‚Ä“ 80 percent of this category in 2003.
466 The Process of Economic Development
exports were accounted for by complex production systems that produce semiconductors,
telecommunications and electronic equipment, and so on in 2003. And it is in these areas,
typically not associated with developing nations, where the TNC-driven integrated produc-
tion process is making its greatest gains in terms of the share of world exports. Yet it should
be emphasized that much of the technologically intensive production, and the underlying
R&D, occur in the advanced industrial nations, with developing nations often adding little
more than cheap labor to the globally integrated process.
Are there spillover effects from FDI?
As we have seen in the above sections, when developing nations host transnational corpora-
tions they cannot reasonably anticipate that direct foreign investment will substantially help
resolve the massive problems of underemployment, disguised unemployment, and unem-
ployment. Nor can a nation expect that, except in the rarest of short-term circumstances, the
rate of investment will significantly rise, thereby boosting the growth rate of the economy.
Greatly debated in research sources is the role played by FDI in spawning technological
learning (know-how) and innovative capacity (know-why). Backward linkages to national
suppliers, sometimes promoted through local content requirements either directly or indi-
rectly (e.g. granting a subsidy or tax break to a company that buys locally) can promote
learning as national firms rise to meet the standards of a top international manufacturer who
shares some production know-how.
Many times parent TNCs operate within relatively tight ‚Äúvertically integrated‚ÄĚ systems
where the backward linkages are performed ‚Äúin-house‚ÄĚ by the firm or its subsidiary. Here
there is likely to be little call for local content, and broad possibilities for the sharing of
some ‚Äúknow-how‚ÄĚ are few. But some learning can occur as the TNC firm advances the
skills of the national workforce to meet the high ‚Äúsupplier‚ÄĚ quality demands of the core
firm. In this ‚Äúsecond-best‚ÄĚ situation, some workers with enhanced skills will eventually
leave the firm and the new knowledge may be diffused to important portions of the national
However, this process can be negated if the parent company does not use much local
production, but rather imports a high portion of its inputs. Hence, few or no backward link-
ages of importance are created, a situation common when a developing nation hosts a Japa-
nese TNC (Belderbos, Capannelli, and Fukao 2001). If spillover effects are to occur a host
nation must have a strong, deep, and modern infrastructure and a base of capable national
suppliers, otherwise parent firms will prefer imported inputs or the setting up of vertically
Surprisingly, a large number of the empirical studies that attempt to find and calibrate
possible spillover effects combine advanced industrial nations and developing nations in
their sample of host nations, making their results difficult for understanding how FDI impacts
developing nations. Yet one interesting result of such a combined study showed that when
the parent company was primarily targeting the national market, instead of setting up an
export platform, backward linkages were substantially higher (ibid.: 202). Meeting the needs
of the local market tended to make the parent company more prone to using local knowledge
adapted to existing consumer product expectations.
This runs counter to the hypothesis that as a result of export-oriented FDI significant
learning will take place, but not if production is geared to the national (home) market. This
same study of Japanese TNCs in twenty-four nations found that the higher the degree of
research and development embodied in the product the less likely that local firms would
Transnational corporations and economic development 467
be brought into the production chain. Hosting high-tech operations, other things remaining
the same, would mean that a developing nation would find very few spillovers. Of course,
this would tend not to hold true if a nation, on its own, had achieved substantial advances in
R&D-related production ‚Ä“ an exceptional situation that would tend to prove the rule, such as
in some sectors of India‚Ä™s economy and in Korea and especially Taiwan.
China is the largest single recipient of FDI as a result of a policy initiative premised
on opening the economy to acquire spillover effects. In a study of over 2,500 large and
medium-sized textile and electronics firms in China in the 1995‚Ä“9 period, Albert Hu and
Gary Jefferson found that in the short run the impact on domestic electronics firms was nega-
tive ‚Ä“ they were crowded out, losing market share and experiencing decline in productivity.
However, national firm survivors did show some capacity to close the gap with foreign firms
over a longer period, and the authors suggested that learning had occurred. It is interesting
to note that the national firms were more R&D-intensive than were the TNCs, which may
explain how some were able to survive and flourish.
This, however, raises important issues: the survivor national firms could have later
improved their productivity in many ways that would not arise from direct or indirect impacts
stemming from TNCs. This underlines a broader problem, which is that most of the attempts
to find spillovers use productivity of domestic firms as a proxy for spillovers in econometric
tests. If they find correlation with FDI and productivity it is assumed that the cause of the
productivity increase was FDI. But the assumed causal relationship needs to be proven by
direct case-studies that econometricians rarely undertake. In the case of China‚Ä™s national
textile producers FDI did impose a negative impact through the entry of foreign firms, in
that market share for national firms declined but productivity for the national firms did not
decline (Hu and Jefferson 2002: 1073).
A major 1999 study by Brian Aitken and Ann Harrison that showed a negative relationship
between an increase in FDI and productivity in national firms in Venezuela was the catalyst
for a flurry of studies that attempted to find positive spillover effects for FDI (Aitken and
Harrison 1999). Since this critical study appeared a deluge of attempted refutations have
appeared in print, but these studies have not produced vigorous results. In other words, while
results have varied to some degree, overall subsequent studies have revealed that econo-
mists‚Ä™ research on the relationship is inconclusive.
In a recent important book devoted to the study of the effects of externalities and spillovers
from FDI it was once again confirmed that economists are unable to show definitive evidence
of spillover effects. Summing up five new studies in this book, Gordon Hanson states:
As international economists, what can we tell policymakers in developing countries
about how they should treat multinational firms? Based on empirical work to date,
the answer, unfortunately, is ‚Äúnot much.‚ÄĚ The literature is just beginning to seriously
consider empirical evidence issues about FDI‚Ä™s effect on domestic firms ‚Ä¦ Given the
developing state of the field, it is important that policymakers realize that we do not
know how [transnational] firms affect their economies. ‚Ä¦ an abundance of evidence
that FDI generates positive spillover effects does not exist. So far, researchers have yet
to uncover robust empirical support for the kinds of subsidies that many countries have
begun to offer multinational enterprises.
(Hanson 2005: 178)
In the same collection of research, Maria Carkovic and Ross Levine studied a sample
of seventy-two countries in the 1960‚Ä“95 period to see if there was evidence that nations
468 The Process of Economic Development
that receive FDI experience general macroeconomic improvements in their overall level of
productivity. Thus their study looked at economies in their entirety instead of ‚Äúfirm-level‚ÄĚ
studies. While the authors note that some macroeconomic studies of FDI have shown posi-
tive effects regarding growth, these studies should be viewed with skepticism, they argue,
because they have not adequately controlled for a number of factors that suggest flawed
methods of investigation (Carkovic and Levine 2005: 196). In contrast, in their search for a
qualitative impact from FDI, their study concludes that:
After resolving many of the statistical problems plaguing past macroeconomic
studies ‚Ä¦ we find that FDI inflows do not exert an independent influence on
economic growth. ‚Ä¦ [Our] results are inconsistent with the view that FDI exerts a
positive impact on growth.
(Carkovic and Levine 2005: 219)
While not uniform in their results, it does seem that the bulk of the studies conducted on
the spillover dispute have come to the conclusion that FDI has either a negative impact on
domestic productivity, or ‚Ä“ in the most positive versions ‚Ä“ a very slight positive effect. This
takes us far from the optimistic scenario projected in Chapter 8, where endogenous growth
would occur as externality and spillover effects proliferate. It is worth stepping back for a
moment from the complexities and vagaries of the many studies conducted on spillovers to
remember some basic propositions that seem to have been lost in the search for spillovers.
First, transnational corporations are the premier private sector owners and creators of
technological learning and capacity. This is their strongest asset. As such, they are going to
combat wherever and whenever they can the loss of their competitive edge, guarding against
‚Äúreverse engineering,‚ÄĚ imitation of their products, and the possibility that subcontractors can
abscond with their hard-won technical knowledge (Nambiar 2001).
Second, when such firms decide to part with some of their power base, they will do so only
grudgingly, when they find that the benefit of some loss of this knowledge base is outweighed
by some cost-cutting advantage through the use of domestic suppliers. At this stage sharing
of some capability is normally partial and therefore does not threaten the dominance of the
transnational. What firms tend to part with is technical capacity that is fairly common, rather
than forms of knowledge and technical capacity that are key to the firm‚Ä™s ability to exercise
monopoly power. So it is helpful to envisage a spectrum of capacities used by the multi-
nationals, divided into low, intermediate, and high. TNCs will part with the low-level capa-
bilities if they find a net advantage in doing so, but this will not normally give national
supplier firms a sizeable boost in their level of efficiency, because the TNC is able to combine
the low-level capabilities with intermediate and highly valued, unique, and very costly forms
of specialized know-how and know-why.
Third, in rare instances, TNCs can find it advantageous to part with some of their interme-
diate capacities. Under these conditions significant spillover effects can occur. But they will
occur only if developing nations have absorptive capacity. That is, the developing nation has
to be in a position to absorb transfers of knowledge, and this entails reasonably deep know-
how capacities that are the outcome of development policies in education, science, manage-
ment, labor skills training, and engineering.
If we take these considerations into account it is not surprising that development econo-
mists have not been able to easily confirm the idea that TNC investments create spillover
effects. The fact that some studies show modest positive effects, some no effect, and some
negative effects would seem to be a logical outcome of the fact that TNCs will, in the first
Transnational corporations and economic development 469
instance, be unwilling to part with any of their knowledge and capability. No passive
knowledge transfers will occur. Only when nations are active in selecting foreign firms, only
when nations have built the capacity to absorb lower levels of knowledge transfers, and only
when such transfers appear to be advantageous to the TNCs ‚Ä“ a fairly demanding string of
conditions ‚Ä“ will we be able to locate spillover effects from FDI.
Potential costs of TNCs to a host country
While there is evidence of situations where TNCs have enhanced productivity and economic
growth in developing nations, the role of the TNCs has created a storm of controversy
precisely because other research highlights the possible costs of TNCs to the host nations.
In this section, we briefly consider how TNCs and FDI might deepen underdevelopment or
facilitate a process of biased economic growth wherein the bulk of the economic benefits are
retained by the TNCs.
TNCs buy many of their inputs from and sell much of their output to other branches or
affiliates of the same TNC, though these often are located in different countries. The extent
of such transactions varies from firm to firm and from industry to industry. Nonetheless,
virtually all the research conducted on such TNC intra-firm transactions indicates that they
are significant. Furthermore, since such transactions are not of an ‚Äúarm‚Ä™s-length‚ÄĚ character
between two independent economic agents who have agreed on the terms at which to buy
and sell, the TNCs are in a position to set favorable intra-firm transfer prices when it is to
their advantage to do so.
A transfer price is simply an accounting price that all firms use for intra-firm transactions
of inputs and semi-finished final goods, particularly as these are shipped between different
branches of the same firms or different divisions for either reprocessing or sale. In the normal
course of business, transfer prices are necessary to allow firms to keep track of costs within
divisions or branches of the firm and to measure the profitability and productivity of different
divisions or branches.
However, there exists the possibility, particularly in the intra-firm movement of inputs and
semi-processed goods between countries, for TNCs to use transfer prices as a mechanism
for avoiding taxes in one country or another and for avoiding any profit repatriation or other
restrictions placed by any country within the company‚Ä™s global operations. Or a firm might
simply want to disguise the extent of its profits as a matter of public relations or in order to
avoid adverse repercussions, which could range from labor tensions to nationalization. Then
it is likely the company will have two sets of books; one for ‚Äúreal‚ÄĚ transfer prices so the TNC
can monitor the effectiveness of production globally, and another to ‚Äúcook the books‚ÄĚ so as to
increase and move global profits of the TNC to the most advantageous location. In the age of
computer operations, the movement of such financial capital is only a keystroke away.
Not all TNCs are engaged in devious transfer pricing schemes, of course. But in one
important study of the pharmaceutical industry in Colombia, Constantine Vaistos found that
82 percent of the companies‚Ä™ actual profits were hidden in transfer pricing schemes. Another
study of such practices in Kenya by the International Labour Organization found that the
actual outflow of profits and dividends was understated by perhaps 100 percent as a result of
the over-pricing of inputs (these studies were from the 1970s, as cited by Crow and Thorpe
470 The Process of Economic Development
In the latter case, it is easy to see how such over-pricing makes TNC profitability in Kenya
seem lower than it really is; the TNC, however, increases its global profitability. Usually,
such strategies are utilized to extract profits from a country where removing profits now or
later might be difficult or subject to restrictions. With such transfer over-pricing, however,
it is easy to see how profits and income actually created in Kenya are withdrawn in the
form of higher costs. Since most less-developed nations have extremely weak tax-collecting
agencies, porous tax laws, and lax enforcement practices anyway, any loss of tax revenues
from the TNCs through such transfer pricing arrangements can be extremely serious. (For a
detailed treatment of the issue see Plasschaert 1994.) Further, such practices tend to reduce
the pool of potential domestic investment funds.
Income transfers via TNCs
While TNCs can contribute to net resource inflows to a developing nation, they can also
contribute to net resource outflows in other ways beside transfer pricing. First, parent corpo-
rations in the developed countries commonly make loans to their subsidiaries in less-devel-
oped economies. In time, such loans will be repaid as interest and amortization, constituting
a potential drain on the balance of payments and foreign exchange earnings of the less-
Unless the subsidiary is earning foreign exchange via exports or saving foreign exchange
by contributing to import substitution in the host country, the outflows of interest and prin-
cipal can exceed the original inflow of financial capital, thus creating a net outflow over time.
Likewise, declared profits often are repatriated to the parent corporation, though profits from
the subsidiary may remain within the host nation either to be lent out or reinvested in the
operations of the subsidiary.
TNCs often find that their relative strength is in the mastery and control of intangibles,
such as organization and information technologies and product technologies. As unique
owners of such technologies, they are in a position, should they choose to do so, to sell or
lease such technologies and other intangibles via joint ventures, franchises, and other interac-
tions where there is no market for the product sold or leased, it being unique. It is presumed,
therefore, that the price charged in such transactions does not represent only the cost to the
TNCs of creating such unique information, but that the TNC also derives some degree of
monopoly rent from such transactions. This, too, contributes to an outflow of income earned
in the less-developed nation flowing to the TNC, presumably headquartered in a developed
Some evidence suggests that when TNCs enter an economy, indigenous research and devel-
opment is curbed and redirected toward adaptive inquiry which merely follows the lead of the
TNCs, rather than participating in the creation of knowledge. When such effects are present,
there is a presumption that the indigenous technological base is narrowed and weakened.
Thus the total effect of a strong presence of TNCs in an economy may be either to make no
net addition to the R&D process or to divert that process away from appropriate domestic
technologies, with an adverse impact on future growth possibilities (UN 1992: 148).
It has long been the claim of the structuralist economists that the TNCs employ capital-
intensive production systems which are inappropriate in poorer nations where labor is both
abundant and relatively cheap, and where the real rate of unemployment and underemployment
Transnational corporations and economic development 471
may be alarmingly high, even if disguised by informal sector activities. Thus with more
capital-intensive production techniques in use, TNCs contribute to urban unemployment
and underemployment. Another diversion effect of the TNC may be an internal brain drain,
whereby some of the top managers and best university graduates seek employment in the
TNCs, leaving the indigenous industrial and agricultural firms with a relatively narrow cadre
of managerial talent, and perhaps not always the best trained.
Increasing industrial concentration
The presence of TNCs is generally associated with increasing industrial concentration. Thus,
according to standard static microeconomic analysis, resources are more likely to be used
sub-optimally as monopoly power is enhanced and the degree of control of oligopolies is
expanded. As economic concentration increases, the distribution of income is further tilted
toward those at the top. Enhanced TNC activity can thus be associated with a greater polariza-
tion of incomes and a tendency to divorce economic growth from enhancement of economic
well-being for the vast majority of the population (Newfarmer and Frischtak 1994). In a
test of the hypothesis that FDI tends to worsen the distribution of income, Pan-Long Tsai
reviewed the available research and concluded that virtually every study conducted on the
subject had reached the conclusion that higher levels of FDI were associated with a wors-
ening of the income distribution.
The most striking result ‚Ä¦ is the unambiguous positive partial correlation between (the
stock of direct foreign investment) and the Gini coefficient.
‚Ä¦ Our results thus confirm previous findings and support the assertion of the depend-
ency proponents. That is, continuing inflows of foreign capital into the LDCs is very
likely to be harmful to the income distributions of the host economies.
(Tsai 1995: 475)
Weak linkages, thin globalization
Much recent data, some of it recorded in Table 14.4 above, suggests that developing nations
are making great strides in terms of rapidly attaining efficient production systems that allow
for a major surge in exports from developing nations. This perception, however, needs to be
understood in a broader context that includes the new systems of global production financed
For national development the most telling indicator of successful incorporation of skill
and technology transfers relates to the magnitude of the national linkages between the TNCs
and the host economy. Since Mexico has regularly been one of the top ten nations in terms of
FDI inflows in recent years it is instructive to note that current research does not suggest that
Mexico‚Ä™s national industrial base has either grown, or diversified, or deepened its capital and
knowledge/skill levels to any serious extent (Cypher 2001). Rather than articulating a new
dynamized industrial sector to the national economy, Mexico exhibits the characteristics of a
‚Äúdisarticulated‚ÄĚ economy with a dynamic export sector, overwhelmingly dominated by TNCs
and largely unlinked to the broader domestic economy (Delgado Wise and Cypher 2007).
Kathy Kopinak refers to the huge maquiladora sector (employing well over a million
workers in assembly operations) as an example of thin globalization (Kopinak 2003). By this
it is meant that, while foreign capital has increasingly moved across borders and into Mexico,
the degree of connectiveness between the export sector and the national economy is very low.
472 The Process of Economic Development
This is an important generalization because it allows us to move beyond the stale ‚Äúyes or no‚ÄĚ
debate as to whether ‚Äúglobalization‚ÄĚ has been achieved. The modifier ‚Äúthin‚ÄĚ permits a critical
understanding of the shallow level of spin-off and assimilation of technology transfers that
dominates the processes entailed in the integrated global production system.
Thus TNCs and their joint-venture or strategic alliance partners may well achieve ‚Äúthick‚ÄĚ
integration of their proprietary processes across national borders, while only loosely linking
to subcontractors and domestic suppliers. The end result of the weak linkages scenario is to
retain most of the value-added in manufacturing within the structure of the TNCs. (Value-
added refers to the difference between the cost of all inputs into the manufacturing process,
and the value of total output. High value-added activities normally are those where high
skills and complex technologies are utilized in production.)
As we have seen in our discussion of Hirschman‚Ä™s concept of ‚Äúbackward linkages‚ÄĚ
(Chapter 5), major investments can create complex, interactive, virtuous circles of forces
that will push an economy to a higher level of development. Recent research has centered
on Hirschman‚Ä™s concept in great detail. The 2001 World Investment Report encapsulated the
meaning of Hirschman‚Ä™s concept in the following quotation:
[Backward linkages] are defined as transactions that go beyond arm‚Ä™s length, one-off
relations and involve long-term relations between firms. In fact, a very large proportion
of intra-industry transactions in every country involves linkages in this sense, marked
by sustained exchanges of information, technology, skills and other assets. Linkages
are of particular significance to developing host economies, because they provide a
means of diffusing valuable knowledge throughout the economy ‚Ä“ through direct flows
to the linked firms as well as spillovers to and from the latter. The benefits provided
through linkages with foreign affiliates tend to be of greater competitive significance
than those among domestic firms because of the stronger knowledge and skills base of
many foreign affiliates.
(UNCTAD 2001: 127)
In the instances where these researchers find integral connections to the local production
base two possibilities arise:
1 Core suppliers to the large transnationals are increasingly foreign-owned. These firms
often exhibit the characteristics of cutting-edge suppliers sought by the TNCs ‚Ä“ mastery
of quality control, capacity for flexible ‚Äújust in time‚ÄĚ delivery, ability to independently
design components and supplies at the level of original equipment standards, and,
perhaps most important, the capacity to jointly address production problems with the
contracting TNCs. Clearly, in this instance neither learning nor technology transfers
occur with heightened FDI, and the domestic economy remains disarticulated from the
accumulation process driven by the TNCs.
2 Deep linkages and dynamic technology and learning transfers occur normally when host
nations intervene and set up their own ‚Äúlinkages promotion programs.‚ÄĚ As discussed
in Chapter 7, the state can play a crucial role in tying together the forces of the TNCs
and the national needs for rapid development. Not surprisingly, most successful linkage
promotion programs have occurred in East Asia. Rather than passively receiving FDI,
nations in East Asia have struggled to upgrade existing linkages, create new domestic
sourcing possibilities, and force TNCs to reorient their production toward linkages to
higher value-added activities. Six key processes have been encouraged by the state:
Transnational corporations and economic development 473
Create public/private sector forums to open a dialogue between the TNCs and
unions, regional planners, national development agencies, business associations,
supplier industry associations, and financial sector firms.
b Disseminate to all parties information regarding successful examples of linkages.
Limit and target specific sectors or industries, bypassing areas where internation-
ally integrated production systems are already dominant.
Choose to host foreign affiliates on the basis of their commitment to interact with
and their potential to spin off crucial learning/technology to local suppliers.
e Select suppliers based on their capability of meeting production standards, quality
requirements, workforce skill requirements, and the commitment of local entrepre-
neurs to restructuring their operations to meet continually evolving standards set by
the contracting TNC.
f Monitor and evaluate local suppliers, rewarding those that meet the above goals.
Nations with developmental states, such as many East Asian nations, have been able to
build strong backward linkages. Elsewhere, the lack of such states ‚Ä“ particularly in Latin
America ‚Ä“ has led to a passive approach to hosting TNCs, and the results have been that these
nations are now enmeshed in a process of ‚Äúthin‚ÄĚ globalization. In this situation the national
economy is largely disarticulated and a new ‚Äúglobal‚ÄĚ dualism has emerged ‚Ä“ ultramodern
TNCs operating in a nearly autonomous transnationalized sphere and a domestic economy
mired in low productivity, poverty, social decomposition, capital flight, and massive unem-
ployment exhibited by a burgeoning ‚Äúinformal‚ÄĚ sector of ‚Äúself-employed‚ÄĚ day laborers who
are, in reality the ‚Äúdisguised unemployed.‚ÄĚ
Table 14.5 documents the trends to be noted in the new integrated global production
system. Many nations have dramatically increased their manufacturing exports, but most