Does foreign direct investment promote economic growth?

Evidence from East Asia and Latin America

Kevin Honglin Zhang . Contemporary Economic Policy 19.2 (Apr 2001): 175-185.



Although there is considerable evidence on the link between foreign direct investment (FDI) and
economic growth in developing countries, causal patterns of the two variables has not been
investigated yet with a reliable procedure. This article provides an empirical assessment of the
issue by using data for ll economies in East Asia and Latin America. Although FDI is expected
to boost host economic growth, it is shown that the extent to which FDI is growth-enhancing
appears to depend on country-specific characteristics. Particularly, FDI tends to be more likely to
promote economic growth when host countries adopt liberalized trade regime, improve education
and thereby human capital conditions, encourage export-oriented FDI, and maintain
macroeconomic stability. (JEL F21, F23, 053)
Headnote
ABBREVIATIONS
Headnote
ADF: Augmented Dickey-Fuller ECM: Error-Correction Model FDI: Foreign Direct Investment
GFCF: Gross Fixed Capital Formation IMF: International Monetary Fund NAFTA: North
American Free Trade Act
I. INTRODUCTION
The role of foreign direct investment (FDI) in developing countries has grown dramatically over
the 1980s and 1990s, from $14 billion in 1985 to $166 billion in 1998, rising nearly 12-fold
(UNCTAD, 1999). How does inward FDI interact with host economic growth? The literature on
this issue can be divided into two broad categories. One group of studies is based on the growth
theory in which FDI has been introduced as one of the factors explaining output growth (e.g.,
Borensztein et al. [1998]). The other group of studies is built on recent models of multinational
enterprises in which FDI flows are primarily associated with income levels and market sizes of
host countries. The statistical significance of the effects of income growth and market size on
FDI has been tested with cross-country data (e.g., Wheeler and Mody [1992]).
The many existing studies on the subject have made useful contributions toward an
understanding of the role of FDI in economic growth and the effect of economic performance on
inward FDI flows. The statistical approach in these studies, however, raises a critical
methodological issue. All of the investigations make an a priori presumption that FDI responds
to or causes economic growth, and few studies have considered the feedback and the long-run
equilibrium relationship between FDI and economic growth.1 Investigation of the causal link
between FDI and growth has important implications for development strategies. If there is a
unidirectional causality from FDI to income growth, it would lend credence to the FDI-led
growth hypothesis that FDI not only leads capital formation and employment augmentation but
also promotes income growth in host economies. If the causal process runs in the opposite
direction, it would imply that economic growth may be a prerequisite for developing countries to
attract FDI and the amount of FDI flows into a country depends on the country's absorptive
capacity. If the causal process is bidirectional, FDI and growth would have a reinforcing causal
relationship.
This article aims to investigate causality between FDI and economic growth for 11 developing
countries in East Asia and Latin America, based on a theoretical framework and a estimation
method that has been developed fairly recently.2 To summarize the position, estimations of the
cointegration tests show that the long-run FDI-GDP links exists for five countries. The results of
errorcorrection model estimations for the five countries then indicate that FDI and GDP in two
countries have feedback, and a unidirectional causality is found for the other three countries. For
the six countries without FDI-GDP cointegration links, the conventional Granger causality test is
conducted, suggesting that although no causal links exist in one case, unidirectional causal
effects are found for the remaining five countries.
The most significant finding thus is that there is much cross-country variation and differences
between East Asia and Latin America in the causal patterns of FDI-- growth links. Regarding to
the impact of FDI on host economies, FDI is found to boost economic growth in 5 of 11
countries, namely, Hong Kong, Indonesia, Singapore, and Taiwan in East Asia and Mexico in
Latin America. FDI seems to be more likely to promote economic growth in countries with
liberalized trade regimes, good human capital conditions, large export-oriented FDI, and
macroeconomic stability.
II. THEORY
Hypothesis of Growth-Driven FDI FDI arises mainly from activities of the firms that operate
across countries. According to the widely accepted framework on the existence of multinational
enterprises, a firm in one country with certain ownership advantages would open a subsidiary in
another country with locational advantages, and both advantages can be best captured by
internalizing production via direct investment (Dunning, 1981).3 Studies focusing on the
location factor, given the other two factors, concern how the economic performance in a host
country along with others determines amount of FDI flows into that country.
To see how FDI flows are driven by host economic growth, it is necessary to distinguish two
types of FDI based on its motives. Market-seeking FDI, made by a multinational firm with two
or more branches in different countries, is induced by market accesses to host countries for
efficient utilization of resources and exploitation of economies of scale (Markusen et al., 1996).
Export-oriented FDI is motivated by factor-price differentials (e.g., cheap labor in host countries)
along with human capital and infrastructure conditions (Zhang and Markusen, 1999). The
growth-driven FDI hypothesis emphasizes the necessity of growing market size and improving
conditions in human capital and infrastructures for attracting FDI (Zhang, 2000). Other things
being equal, a country's market size (measured by GDP) rises with economic growth,
encouraging foreign firms to increase their investment. Rapid economic growth leads to high
level of aggregate demand that stimulates greater demand for investments including FDI.
Moreover, better economic performances in host countries provide a better infrastructural
facilities and greater opportunities for making profits, and so greater incentive for FDI.
Hypothesis of FDI-Led Growth
The impact of FDI on host economies may be analyzed in the context of its effects on growth-
driving factors, such as investment, human capital, exports, and technology. These growth-
promoting factors might be initiated and nurtured to enhance growth through FDI. In fact, FDI
has been integrated into theories of economic growth as the "gains-from-FDI" approach (Graham
and Krugman, 1995).' To the extent that FDI adds to the existing capital stock, FDI may have
growth effects that are similar to that of domestic investment, along with alleviating partly or
totally the balance-of-payments deficits in the current account. In addition to employment
augmentation, multinational firms train workers and managers whom local firms may later
employ. FDI (particularly export-oriented FDI) may promote exports by setting up assembling
plants and helping host firms access international markets for exports (Aitken et al., 1997).
Recent studies suggest that FDI might be able to enhance economic growth of host countries
through spillover efficiency and technology transfer. The spillover efficiency occurs when
advanced technologies and managerial skills embodied in FDI are transmitted to domestic plants
simply because of the presence of multinational firms. The technology and productivity of local
firms may improve as FDI creates backward and forward linkages and foreign firms provide
technical assistance to their local suppliers and customers (Rodriguez-Clare, 1996). The
competitive pressure exerted by the foreign affiliates may also force local firms to operate more
efficiently and introduce new technologies earlier than what would otherwise have been the case
(Blomstrom et al., 1992). It has been recognized, however, that the spillover efficiency and
technology transfers do not appear automatically but depends on host countries' absorptive
capability that is largely determined by human capital in host countries (Borensztein et al.,
1998).
Hypothesis of Feedback
The most interesting economic scenario suggests a two-way causal link between FDI and
economic growth. Though it is possible that the strong association between FDI and growth
could be a result of either the growthdriven FDI or FDI-led growth, it is equally likely for the
two variables move together through feedback (Caves, 1996). Countries with fast economic
growth, not only generating more demand for FDI but also providing better opportunities for
making profits, attract greater FDI. On the other hand, FDI inflows may foster economic growth
of host countries through positive direct effects and indirectly spillover effects. Both FDI and
economic growth are positively interdependent and could lead to a bidirectional causality.
111. FDI AND ECONOMIC GROWTH IN EAST ASIA AND LATIN AMERICA
General Trend of FDI
The past decades have seen profound changes in the role of East Asia and Latin America as
recipients of FDI in the world economy. Table 1 shows the trend of inward FDI over 1960-97 in
all developing countries as well as East Asia and Latin America. Two things are suggested in the
table. First, as the dominant recipient of FDI in the developing world since 1960, Latin America
and East Asia together received the lion's share of all FDI flows in developing countries. Their
share also rose over time, from 79% in 1960-69 and 80% in 1970-79, to 89% in 1986-90, and
even 92% in 1990s. Second, the relative position of the two regions changed in the mid-1980s.
After a period of strong growth in the 1960s and 1970s, Latin America experienced a slowdown
and drastic deceleration in the growth of FDI flows until the 1990s, such that its share in both
worldwide inflows and inflows to developing countries declined in the 1980s. During the same
period, the share of East Asia in both worldwide inflows and inflows to developing countries
rose substantially. Although Latin America accounted for 46% of inflows in developing
countries in the first half of the 1980s, East Asia became the largest recipient of FDI among
developing regions in the second half of the decade, accounting for 54% of inflows.
The growing interest in East Asia as a host to FDI is a reflection both of favorable economic
performance and prospects and policy regimes that are relatively open to both international trade
and capital flows. In comparison, until the early 1990s, Latin America was lacking in both
economic and policy attributes that stimulate FDI. Moreover, the region was shackled by an
external debt burden of close to $400 billion that fostered economic instability and political
uncertainty (Braga, 1992). As a result, average real growth of GDP per capita for the region
became negative at 1.1% during the 1980s, compared to over 3% during the 1970s. Thus the
1980s have been described as the "lost decade" of economic growth and investment in the
region.
Importance of FDI
The importance of FDI in host economies, indicated by FDI stock as a percentage of GDP (Table
2) and FDI flows as a percentage of gross fixed capital formation (GFCF) (Table 3), differs
between countries as well as between East Asia and Latin America. With the notable exceptions
of Hong Kong, Singapore, Malaysia, and Indonesia, FDI stock as a percentage of GDP is mild
and small in 1980. That indicator has risen since then in all countries except Hong Kong. At the
end of 1997 the ratio in the four Latin American countries is similar, ranging from 12%-16%.
Large differences in the ratio emerge in East Asia in 1997, with high levels in four countries
(Singapore, Hong Kong, Malaysia, and Indonesia) and low levels for the remaining three (Korea,
Taiwan, and Thailand). Table 3 indicates that FDI flows as percentage of GFCF over 1987-1997
in Latin America rose significantly, while that indicator remains relatively stable in East Asia
except Indonesia with the growing ratio. Because FDI increased in both regions in the 1990s, the
pattern of the indicator seems to suggest that while Latin America might have relativelys
constant growth of domestic investment, East Asia experienced fast growth in domestic
investment than FDI.
Sectoral Pattern
The sectoral distribution of FDI and its changes over time in East Asia and Latin America are
presented in Table 4. In 1988 the two regions share a similar pattern in which majority of FDI
was concentrated in manufacturing (about 60%) and the stock of FDI in primary sector was
small (about 8%). The service sector has acquired importance in both regions since then,
particularly in Latin America, where service FDI has become dominant (the share rose from 24%
to 56% in 1997). The service FDI in East Asia rises at expense of primary FDI, while Latin
America gains service FDI at expense of manufacturing FDI (falling from 62% to 39%).
Export Performance
The role of multinational firms in exporting manufacturing has been held up as a major
contribution to host economic growth. Though data constraints prevent a full assessment of the
impact of FDI on host exports in East Asia and Latin America, export-- propensities of U.S.
majority-owned affiliates in manufacturing (Table 5) reveal some interesting points in the two
regions. The general pattern is that U.S. affiliates in East Asia tend to export much more than
those in Latin America in the past few decades. In fact, the export-propensities in most East
Asian economies are over 50%, in contrast to that indicator of less 27% in Latin America except
Mexico, in which US affiliates increase their exports due to the North American Free Trade Act
(NAFTA).
Economic Growth
GDP per capita and its growth rates in East Asia and Latin America over 19601992 are presented
in Table 6. Despite enormous natural resources, per-capita income in Latin America grew slowly
(0.2-2.5%) over the three decades. For instance, being one of the world's richest countries at the
turn of the last century, Argentina has become progressively poorer relative the industrial
countries. From 1960 to 1992, annual growth rate of per capita GDP is only 0.2%. The answer to
Argentina's regress from riches-- to-rags is complex as usual, but the country's inward orientation
and macroeconomic instability appear to be major culprits (Sachs and Warner, 1995).
Relative to Latin America, most East Asian countries were desperately poor in 1960, apparently
with few economic prospects. In the early 1960s, however, the countries launched a series of
economic reforms, shifting from an inward-looking, import-substitution development strategy to
one that emphasized exports, which resulted in a remarkable economic ascent. South Korea, for
example, with an income level of $641 in 1960, has grown at almost 7% per year since then and
raised its income by a factor of 8 over the next three decades (Table 6).
There are many differences among East Asian economies, but what they did have in common
were high rates of saving and investment; rapidly improving educational levels among the
workforce; stable macroeconomic conditions; and if not free trade, at least a high degree of
openness to and integration with world markets (World Bank, 1993). In 1990 the East Asian
economies saved 34% of GDP, compared with only half that in Latin America. Even in 1965,
when the East Asian countries were still quite poor, they had high enrollment rates in basic
education: Essentially, all children received basic schooling in Hong Kong, Singapore, and South
Korea; even desperately poor Indonesia had a 70% enrollment rate. By 1987, rates of enrollment
in secondary school in East Asia were well above those in Latin American nations, such as
Brazil. Relative to Latin America, East Asia also has enjoyed much more stable macroeconomic
environments since 1960, free from high inflation or major economic slums. The contrast in
openness with Latin America is even particularly clear. According to the World Bank (1993),
while the trade share in GDP for Latin America was 23% in 1988, the average share for East
Asia is over 60%, and much higher in Singapore (347%), Hong Kong (282%), Malaysia (109%),
and Taiwan (90%).
The facts portrayed so far may be summarized in the following manner: (a) Latin America has
fallen behind East Asia as a host to FDI; (b) FDI seems to play more important role in East Asian
than Latin America; (c) though a majority of FDI in Latin America is concentrated in service
sector, most FDI in East Asia is in manufacturing industries; (d) foreign branches in East Asia
tend to export more than those in Latin America; and (e) economic performance in East Asia
seems to be much more outstanding relative to Latin America.
IV EVIDENCE
To make an empirical assessment of the link between FDI and economic growth, a three-step
procedure ("unit root-- cointegration-causality tests") is adopted to investigate possible causal
relationship between the two variables. The econometric specifications of the procedure are not
described in detail because they have been standardized in the literature. The data used in this
study are annual real FDI stock and real GDP of 11 economies (i.e., Argentina, Brazil,
Colombia, and Mexico in Latin America and Hong Kong, Indonesia, Korea, Malaysia,
Singapore, Taiwan, and Thailand in Asia). The criteria for choosing these countries include
positive attitudes of host governments toward FDI, relative importance in terms of share of
inward FDI in the developing world, and availability of data for a reasonably long period of time
(30 years or so).5 Although the real GDP is used to measure economic performance, FDI stock
instead of FDI flows seems to be the more appropriate variable in relation to GDP, just like GDP
and exports that are used widely in the literature for analyzing causal links between economic
growth and exports. As is common in the literature, each variable is used in the logarithmic form.
Data for real GDP after 1968 are taken from International Financial Statistics Yearbook (various
years) in 1990 prices. The GDP data before 1968, which are in 1985 or 1980 prices, are
converted to 1990 prices on the basis of GDP deflators for each countries. Data on real FDI
stocks are generally not available for most countries. The construction of these data is primarily
based on three sources: the FDI/TNC database by United Nations (UNCTAD, 1999),
International Financial Statistics Yearbook (various years) by the International Monetary Fund
(IMF), and net FDI flows to developing countries by World Bank (1997). The UN database
contains FDI inward stock after depreciation (in current U.S. dollars), by host region and
economy, in years 1980, 1985, 1990, 1995, 1996, and 1997. The World Bank data set reports
annual net FDI flows from 1970 and 1995 in constant prices (1996 U.S. dollars). The IMF data, a
standard source for most empirical studies of FDI, are annual FDI inflows in current U.S. dollars.
The data of inward FDI stock are calculated in a two-step procedure. First, annual FDI flows by
IMF are converted to constant price by U.S. GDP deflator (1990 price), with some adjustments
based on the World Bank data. Second, the data of FDI stock are obtained by accumulating over
years, with adjustments based on UN data.
A. Unit-Root Tests for Stationarity and Cointegration Analysis
Table 7 presents results of unit-root tests and cointegration tests, which are required for further
causal analyses. For eight countries (Argentina, Colombia, Indonesia, Korea, Malaysia, Mexico,
Taiwan, and Thailand), the augmented Dickey-Fuller (ADF) tests for both series of FDI and
GDP reject the null hypothesis of nonstationarity in first-differences at the conventional
significance levels, suggesting that both FDI and GDP are integrated of order of one, that is, I(1).
For the remaining three countries (Brazil, Hong Kong, and Singapore), both FDI and GDP are
integrated of order two (1(2)). As the two variables are nonstationary with time-dependent means
and variances, the tests of cointegration are necessary to establish long-run relations between the
two series (Engle and Granger, 1987).
As discussed in section II, the theory of FDI generally suggests a positive link between FDI and
GDP in developing countries. I want to know if such a positive link may sustain in the long run
or is just a short-run phenomenon. The long-run relationship might be investigated by the
Johansen procedure to see whether the series of FDI and GDP are cointegrated. The last column
of Table 7 contains likelihood ratios for maximum eigenvalues, which test the hypothesis of no
cointegration; that is, there are no any equilibrium forces that keep FDI and GDP together in the
long run. There are five countries (Colombia, Mexico, Hong Kong, Indonesia, and Taiwan) for
which FDI and GDP are cointegrated, and thereby a long-run equilibrium relationship between
the two variables is suggested. For the other six countries, the estimations fail to reject the
hypothesis of no integration.
B. Causality in the Short Run and Long Run
Thus far I have established a long-run positive relationship between FDI and GDP for five
countries and found no long-run relationships for six countries based on the cointegration tests.
The next question is the patterns of Granger causality across the variables. For those countries
whose FDI and GDP do not move together in the long run, it is possible for FDI and GDP affect
each other in the short-run. The conventional Granger causality tests will thus be applied for
these cases. For the countries with cointegration between FDI and GDP, the appropriate format
to investigate the long-run causality is the errorcorrection model (ECM).
Using the selected optimal lags determined by Akaike's minimum final prediction error (Hsiao,
1981), the Granger model is estimated for the short-run causality for the countries where FDI and
GDP are not cointegrated. The causal directions are detected by F-statistics, and the signs of the
causal effects are determined by adding the coefficients on lagged independent variables (Ram,
1988). The causal inference of the Granger tests is summarized in the first three columns of
Table 8. Four countries exhibit unidirectional and positive short-run causal effects from GDP to
FDI, namely, Brazil, Korea, Malaysia, and Thailand. For Singapore, the estimates suggest a
unidirectional positive causality running from FDI to GDP. No causal links between GDP and
FDI are found in Argentina. There is no bidirectional causality in any case.
The long-run causality between FDI GDP is investigated with the ECM for the five countries in
which FDI and GDP are cointegrated. The lag length is set at one so that the two sources of
causation (lagged terms and error terms) for FDI and GDP might be exposed, in addition to the
consideration of relatively small size of the sample.6 Causal inferences for the ECM estimates
are reported in the second half of Table 8. As shown, the long-run causality between FDI and
GDP runs in both directions for Indonesia and Mexico, and for three other countries (Colombia,
Hong Kong, and Taiwan) there is unidirectional causality.
The main results might be summarized as follows:
1. In Latin America, we found that FDI and GDP in Argentina do not move together in the long
run, nor impact on each other in the short run. A positive causality running from GDP to FDI is
found for Brazil in the short run and for Colombia in the long run. FDI and GDP in Mexico
exhibit a positive long-run equilibrium relationship in which such co-movement is caused by the
bidirectional causality across the two variables.
2. In East Asia, the short-run causal link from GDP to FDI is found for Korea, Malaysia, and
Thailand. Although the causality from FDI to GDP does exist in Hong Kong and Taiwan in the
long run, the causality is suggested in Singapore in the short run. A positive long-run feedback
between FDI and GDP is indicated in Indonesia.
How do we interpret the empirical results of links between FDI and GDP, particularly effects of
FDI on host economic growth? Evidence in favor of the FDI-led growth hypothesis is mixed in
the sense that among the 11 countries included in the study are 5 cases where economic growth
is found to be enhanced by FDI. The role of FDI in host economies seems country-specific and
sensitive to host economic conditions and FDI policies. Moreover, for the five countries with
positive effects of FDI on economic growth, four (Hong Kong, Indonesia, Singapore, and
Taiwan) are located in East Asia (the fifth being Mexico in Latin America). Does this suggest
that there are different patterns of FDI-growth links in the two regions due to different country
characteristics in the past decades?
The impact of FDI on economic growth may depend on, along with others, hostcountry
characteristics, such as trade strategy, human capital, and export propensities of FDI. FDI might
be more likely to be growthenhancing in countries that pursue export promotion than in those
following import substitution strategy (Balasubramanyam et al., 1996). A key benefit created by
FDI to host countries is technology transfer and spillover efficiency, as mentioned in section II.
The benefit, however, does not automatically occur but depends on host countries' absorptive
capability, which is positively related to human capital (measured by school enrollment ratios) in
host countries. Exports have been recognized for long time as a growth-enhancing factor. It is
natural to believe that host economies would be more likely to benefit from export-oriented FDI
than market-seeking FDI (Caves, 1996).
The success of utilizing FDI to enhance economic growth in the four East Asian economies thus
may be related to their export-promotion strategy, improving education and thereby human
capital conditions, and policies that encourages export-oriented FDI, along with macroeconomic
stability. The failure of three Latin American countries seems to be a result of what they did
oppositely in these aspects. These necessary conditions, however, are not sufficient to guarantee
the positive effects of FDI on growth. The insignificant effects of FDI on growth in Korea, for
instance, may be explained by its small FDI stock relative to GDP and low share of FDI inflows
in domestic capital formation (Tables 2 and 3). The positive effects of FDI in Mexico may be a
result of its many favorable changes and reforms since the 1980s. With the growing economic
integration with the United States under NAFTA, Mexico have received large U.S. direct
investment flows, particularly export-oriented FDI. Export propensities of U.S. affiliates in
Mexico rose sharply, from 20% in 1983 to 32% in 1993 and then to 53% in 1996 (Table 5). In
addition, Mexico introduced a broad stabilization and reform program in the 1980s, combining a
significant commitment to free trade by joining the General Agreement on Tariffs and Trade and
an aggressive reduction in publicsector deficit and foreign debt. The cases of Malaysia and
Thailand are little more complex, reflecting the impact of other forces that affects economic
growth and FDI flows.
In sum, the results of this study suggest that it is clear that FDI may potentially be a growth-
enhancing factor to host economies, but its actual impact on host economic growth depends on
many factors, particularly host-country characteristics.
V. CONCLUDING REMARKS The purpose of this study is to address the link between FDI and
economic growth in Latin American and East Asia, which have received a dominant share of
total FDI into the developing world. The hypotheses of growth-driven FDI and FDI-led growth
are first developed on the basis of recent studies on decisions of multinational firms and
economic effects of FDI in developing countries. Some stylized facts of patterns of FDI and
economic growth are then presented in comparison with the two regions. Empirical analyses of
FDI-growth link, based on a three-step procedure, show that FDI tends to be more likely to
promote economic growth in East Asia than Latin America. In particular, FDI is found to
enhance host economic growth in 5 cases (Hong Kong, Indonesia, Singapore, Taiwan, and
Mexico) out of 11 countries.
The major point that emerges from this work is that patterns of FDI-growth links display
significant differences between East Asia and Latin America, and the differences probably
reflect the enormous cross-national diversity in economic structures. The impact of FDI on host
economy is country-specific, but FDI tends to be more likely to promote economic growth when
host countries adopt liberalized trade regime, improve education and thereby human capital
conditions, encourage export-oriented FDI, and maintain macroeconomic stability.
There several directions in which further work can be done. An increase in the number of
countries, particularly including African countries, should lead to a helpful broadening of the
cross-countries perspective. Extension of time-series data back to the early 1950s and up to 1999
for inclusion of the Asian financial crisis would also be useful. Last, studies conducted at
industrial or firm levels appear worthwhile.
Footnote
1. Another problem in empirical specification is related to cross-country data. Estimates obtained
from cross-section data are useful in many ways, especially when the number of observations for
individual countries is small. However. there is evidence of tremendous parametric variation
across countries in regard to estimates of the growth equations and FDI equations typically used
in such contexts. Although imposition of a common structure in the form of cross-section models
can be a drastic simplification, the important parametric differences could be masked in cross-
section estimates even when the sample chosen looks fairly homogeneous with reference to
certain prior criteria. It seems important,
Footnote
therefore, to investigate the FDI-growth nexus for individual countries on the basis of time-series
data.
2. Although the multivariate causality framework might be able to provide more information
about the relationship between FDI and economic growth, an appropriate theoretical structure for
that is not obvious. Moreover, limited data availability for the additional variables for most less
developed countries lowers the feasibility of a multivariate approach.
Footnote
3. Three sets of advantages for a firm to go multinational are identified in Dunning's "OLI"
paradigm of eclectic theory (Dunning, 1981): Ownership advantage includes factors within the
firm that enable it to compete across borders (e.g., proprietary technology in products, process
designs, or marketing networks). Location advantages are factors in a host country that
determine the country as the best site for the firm's subsidiary (e.g., cheap labor and growing
market). Internalization factors associated with the firm's trade-off between FDI and exporting or
licensing (e.g., transaction costs and barriers to trade).
Footnote
4. The role of FDI in host economies, however, is still subject to considerable disputes. A good
survey on the issue can be found in Helleiner (1989). Some studies argue that FDI does not
improve and may even reduces the welfare of a recipient country when multinationals create
enclave economies within host countries and when market distortions exist in the host economy
due to tariffs or taxes. According to the dependency theory, FDI might actually lower domestic
savings and investment, lead to the shrinking of indigenous industries, widen the income gap,
and bias the economy toward an inappropriate technology and product mix. Some even believe
that multinational firms may engage in a series of tactics that enhance the welfare of
industrialized countries at the expense of host countries.
Footnote
5. In fact, under the criteria all major FDI-host developing countries in past four decades are
included in the sample except China. Starting to be open to FDI in 1979, China has become the
largest recipient of FDI in the developing world since 1993 due to its liberalized FDI regime,
huge market size, and rapid economic growth. The results of causality tests for China conducted
with the same procedure mentioned here suggest that there is a positive feedback between FDI
and economic growth over 1979-1999.
Footnote
6. The analysis also is repeated with greater lag lengths until 3, in which the patterns of causality
in the ECM are determined by F-tests. The results with longer lag length are identical
qualitatively with those in Table 8.
References
REFERENCES
References
Aitken, B., G. Hanson, and A. Harrison. "Spillovers, Foreign Investment, and Export Behavior."
Journal of International Economics, 43, 1997, 103-32.
Balasubramanyam, V N., M. Salisu, and D. Sapsford. "Foreign Direct Investment and Growth in
EP and IS Countries." Economic Journal, 106, 1996, 92-105.
References
Blomstrom, M., A. Kokko, and M. Zejan. "Host Country Competition and Technology Transfer
by Multinationals." NBER Working Paper No. 4131, 1992.
Borensztein, E., J. De Gregorio, and J.-W Lee. "How Does Foreign Direct Investment Affect
Economic Growth?" Journal of International Economics, 45, 1998, 115-35.
References
Braga, C. "Foreign Direct Investment in Latin America and East Asia: A Comparative
Assessment." World Bank Discussion Paper No. 3, November 1992.
References
Caves, R. Multinational Enterprises and Economic Analysis, 2nd ed. Cambridge, MA:
Cambridge University Press, 1996.
Dunning, J. International Production and the Multinational Enterprises, London: George Allen
and Unwin, 1981.
Engle, R. F, and C. W. J. Granger. "Cointegration and Error Correction: Representation,
Estimation and Testing." Econometrica, 55, 1987, 251-76.
Graham, E., and P. Krugman. Foreign Direct Investment in the United States, 3rd ed.
Washington D.C.: Institute for International Economics, 1995.
Helleiner, G. "Transnational Corporations and Direct Foreign Investment," in Handbook of
Development Economics, edited by H. Chenery and T N. Srinivasan. New York: Elsevier
Science Publishers BX, 1989, 1441-80.
Heston, A., and R. Summers. The Penn World Table (Mark 5.6). http://cansim.epas.utoronto.ca:
5680/pwt/pwt.html
Hsiao, C. "Autoregressive Modeling and Money-Income Causality Detection." Journal of
Monetary Economics, 7, 1981, 85-106.
International Monetary Fund (IMF). International Financial Statistics Yearbook. Washington,
DC: IMF, Various years.
References
Markusen, J., A. Venables, D. Konan, and K. Zhang. "A Unified Treatment of Horizontal Direct
Investment, Vertical Direct Investment, and the Pattern of Trade in Goods and Services." NBER
Working Paper No. 5696, 1996.
Ram, R. "Additional Evidence on Causality between Government Revenue and Government
Expenditure." Southern Economic Journal, 54(3), 1988, 763-69.
References
Rodriguez-Clare, A. "Multinationals, Linkages and Economic Development." American
Economic Review, 86(4), 1996, 852-73.
Sachs, J., and A. Warner. "Economic Convergence and Economic Policies." NBER Working
Paper No. 5039, 1995.
United Nations Conference on Trade and Development (UNCTAD). World Investment Report
1999. New York: United Nations, 1999.
Wheeler, D., and A. Mody. "International Investment Location Decisions: The Case of U.S.
Firms." Journal of International Economics, 33, 1992, 57-76.
References
World Bank. The East Asian Miracle. New York: Oxford University Press, 1993.
Foreign Direct Investment. Washington, D.C.: International Financial Corporation (IFC) of the
World Bank, 1997.
Zhang, K. H. "Human Capital, Country Size, and NorthSouth Manufacturing Multinational
Enterprises." Economic Internazionale/International Economics, 53(2), 2000, 237-60.
Zhang, K., and J. Markusen. "Vertical Multinational and Host-Country Characteristics." Journal
of Development Economics, 59, 1999, 233-52.
AuthorAffiliation
KEVIN HONGLIN ZHANG*
AuthorAffiliation
*This is a revised version of the paper presented at the Western Economic Association
International annual conference, San Diego, 1999. The three referees' comments are extremely
helpful. The usual caveat applies.
AuthorAffiliation
Zhang: Assistant Professor, Department of Economics, Illinois State University, 422E Stevenson
Hall, Campus Box 4200, Normal, IL 61790-4200. Phone 1-309-438-8928, Fax 1-309-438-5228,
E-mail khzhang@ilstu.edu
Copyright Western Economic Association Apr 2001