FOREIGN CAPITAL AND ECONOMIC GROWTH OF DEVELOPING COUNTRIES: DEBT VS. EQUITY

BASU SHARMA University of New Brunswick, Canada

T he issue of the effect of foreign capital resources on economic growth of developing countries has continued to occupy a central

position in the literature dealing with development problems of our time. However, questions asked by researchers in this respect have changed along with changes in predominant forms of foreign capital resources available to developing countries.

In the 1950s and the early 1960s, the dominant form of foreign capital was foreign aid, mainly through government-to-government transfer of resources. The literature dealing with the relationship be- tween foreign capital and economic growth of developing countries dealt with this particular type (Rosenstein-Rodan, 1961; Chenery and Strout, 1966). In the late 1960s and early 1970s, foreign direct invest- ment (FDI) came into prominence. And the subsequent literature made a distinction between foreign aid and FDI while examining the relationship between foreign capital and economic growth of develop- ing countries (Papanek, 1972; Kraska and Taira, 1974; Stoneman, 1975; Bornschier, Chase-Dunn, Rubinson, 1978; Gobalet and Dia- mond, 1979; Sharma and Taira, 1982; Rothgeb, 1984).

The dominant form of foreign capital in the 1970s was the foreign private loan (FPL). Also, some of the developing countries which bor- rowed FPL heavily grew faster than others in the 1970s. Many, includ- ing some international organizations, started to assert that debt is good for economic growth. The enthusiasm about debt for economic growth however did not last long. Along with the perceived crisis in the inter- national financial market brought about by the debt problem of some developing countries, there has now emerged a new view that equity capital is better than debt capital for economic growth of developing countries. But the distinctive effect of debt capital vis-~i-vis equity cap- ital on the growth performance of developing countries has not been systematically examined yet.

This article examines growth effects of FDI, FPL, and foreign public loan (FA) by formulating a model consistent with the theory of eco- nomic growth to disentangle the issue raised above.

Sharma 61

FOREIGN CAPITAL AND ECONOMIC GROWTH

Most of the economic growth models pertaining to developing coun- tries consider capital the most important variable. These models attribute the lag in the economic progress of developing countries largely to a lack of sufficient investible capital and its productive use. The static 'vicious circle' concept of underdevelopment, the "quasi- equilibrium state" argument, the "low-level equilibrium trap" thesis, and the "stages of growth" theory indicate that the near subsistence income levels prevailing in many developing countries do not permit the mobilization of investible funds large enough to overcome the growth depressing effects which tend to be stronger at a low level of per capita income. Growth strategies such as the "big push," the "take-off," the "critical minimum effort," and the "absorption of surplus labour" which suggest for breaking out of the traditional stagnation assign cap- ital formation a central role in the process of economic growth.

Capital formation requires savings and their investment in a produc- tive way. But developing countries cannot save as much as they could usefully invest. In some cases, even if savings are enough to meet required investment, it may not be possible to carry out investment projects because of inadequate foreign exchange resources to pay for materials and components of capital goods that must be purchased abroad. The former is known as the "savings gap," and the latter a "foreign exchange gap" (Chenery and Strout, 1966). Besides, develop- ing countries are generally short in entrepreneurship and technology. Their absence or underdevelopment hinders innovations and produc- tivity growth, among other things. It is often suggested that an inflow of foreign capital helps developing countries to fill in these gaps and to overcome these constraints.

There are voluminous empirical works dealing with the relationship between foreign capital and economic growth of developing countries (Bornschier, Chase-Dunn and Rubinson, 1978). Although the literature appears to be largely controversial, at least two views concerning this relationship may be discerned (Riedel, 1979). In one view, an inflow of foreign capital is considered to constitute a net addition to resources available for investment, and hence an exogenous factor to stimulate economic growth. In another view, an inflow of foreign capital is con- sidered to have induced public and private sectors to save less and to consume more domestic resources, producing a growth-depressing effect. The basis for this claim is that some researchers consider savings in developing countries to be mainly a function of government pol-

62 Studies in Comparative International Development / Spring 1986

icies, and argue that when foreign capital is available, the governments of these countries may voluntarily relax savings efforts. Which view is correct has become an empirical issue that has been debated in the literature.

FORMS OF FOREIGN CAPITAL

Foreign capital is not a homogenous category. It consists of diverse types of resources such as FDI, FPL, foreign public assistance (FA), and foreign public grants. Some of the previous studies on growth effects of foreign capital have taken an aggregative approach, lumping FDI, FPL and other forms of external resources together. That is to say, these studies have treated all these resources as one and the same (Griffin, 1970; Enos and Griffin, 1971; Weisskopf, 1972). And this method of specification implies that the various forms of foreign capital resources are substitutable. However, FDI, FPL and foreign public assistance are not the same because their sources are different. Since their sources are different, the modus operandi for their movement should also be dif- ferent.

FDI is not just capital but a bundle of capital, technology and en- trepreneurship. On the other hand, FPL does not embody these features associated with FDI. The same holds for FA. Hence the desirability and the availability of one form of foreign capital as compared to the other depends on various factors. However, the availability aspect tends to be predominant in the process of international capital movement. It may, therefore, be worth looking into the supply aspect of FDI, FPL and FA.

FDI is more a business contract between investors and host govern- ments. It moves in response to availability of profit opportunities. Profit making opportunities depend upon revenue and costs. A firm can generate a higher level of revenue if it can sell a higher level of output at a competitive price, and if it can reduce its costs by purchas- ing required inputs at a lower price. Clearly, a country which has a large domestic market, a relatively higher income level, and relatively low labour and raw materials costs will be an ideal place for foreign inves- tors.

There appear to be two major considerations as to the supply of FPL. First, foreign creditors extend loans only if the borrowing country pays a higher rate of interest relative to the rate of interest in the creditor's home market. Secondly, creditors are also risk averse. They evaluate the potential capacity of the borrower to pay interest and to repay the loans back (Stalling, 1982). Put differently, the higher the level of in-

Sharma 63

comes ~ of a borrowing country, the more accessible the FPL will be for it to borrow.

Insofar as the supply of FA and foreign public grants is concerned, they mainly appear to be political and humanitarian instruments of donor governments. These resources are said to have been frequently used in the service of international diplomacy.

Since sources, nature and logic of movement of FDI, FPL and FA are different, one should not expect a similar effect of all these resources on the economic growth of host/recipient developing countries. More specifically, the logic of profit maximization implies efficient use of FDI because, if the investment cannot be productive, opportunities for profit-making will be limited. Therefore, the effect of FDI on economic growth of host countries should be highly positive. In the case of FPL, the borrowers are largely free to use it at their discretion. How much contribution FPL can make to economic growth of debtor countries depends upon how rationally and how efficiently the borrowers can allocate this resource. Nonetheless, FPL carries with it a relatively higher rate of interest. It matures earlier than FA. These conditions may lead the borrowers to use FPL in productive economic activities as much as possible. However, there may be no such things in the case of FA. Since its movement is conditioned by political considerations, eco- nomic incentives behind FA seem to be negligible.

THE MODEL

The arguments presented above support the contention that equity capital (FDI) should be more productive than debt capital (FPL and FA). Of the latter two forms of debt capital, FPL should have contributed more to economic growth of recipient countries than FA. The dif- ferences in the efficiency of these different forms of foreign capital are assumed to arise because of the structure of the built-in incentive sys- tem. But to test these hypotheses empirically, it is necessary to incorpo- rate FDI, FPL and FA as independent variables in a model of economic growth. Therefore, we start with the familiar production function of the Cobb-Douglas type in the following form:

Y = F(K,L) (1)

where

Y = Total output, or gross domestic product

64 Studies in Comparative International Development / Spring 1986

K = Capital input L = Labour input

The production function is differentiable, and its differential gives:

dY/Y = 0Y/0K (dK/Y) 4- 0Y/0L (L/Y) (dL/L) (2)

Assuming that the marginal product of K and the elasticity of output with respect to labour are stable, one might rewrite Equation 2 as:

dY/Y = b (dK/Y) 4- h(dL/L) (3)

This establishes the central importance of capital formation and labour force growth in relation to growth of output. Explanatory variables for the empirical test of economic growth must therefore include these two variables, appropriately specified.

K in Equation 3 can be considered a composite indicator of all factors that help capital formation. In the framework of this paper, the components of K are domestic savings (DS), FDI, FPL and FA. Sub- stituting for dK in Equation 3, and allowing for an intercept term, we can write Equation 3 as:

dY/Y = a + b DS/Y + c FDI/Y + d FPL/Y + g FA/Y + h dL/L + e (4)

where

dY/Y = rate of growth of gross domestic product dL/L = rate of growth of labour force DS = domestic savings e = stochastic error term of the equation Other variables as defined above

This growth equation allows us to examine effects of foreign capital variables separately on the growth of the gross domestic product. In subsequent sectins we shall describe sources of data used to estimate Equation 4, and present empirical results.

THE DATA

Availability of petinent data on all the variables concurrently deter- mined the number of countries included in the analysis. And we were able to compile data on pertinent variables for a sample of 62 develop-

Sharma 65

ing countries a list of which is presented in the appendix. Data related to output and domestic savings are derived from the United Nations Yearbook of National Accounts Statistics (various issues). Data on la- bour force are taken from the World Tables 1980, published by the World Bank. Statistics on foreign direct investment, foreign private loan and foreign public loan are taken from the Handbook of Interna- tional Trade and Development Statistics 1980, published by the United Nations' Conference on Trade and Development. It is important to note that foreign direct investment is net of profit repatriation, divest- ment and other outflows related to it. Foreign public loans and foreign private loans are net of debt servicing.

The input data for final analysis were prepared in the following way: the domestic savings rate, FDI/Y, FPL/Y and FA/Y are the annual aver- ages for the period 1970-77. The growth rates of the gross domestic product and the labour force are compound rates of growth for the same period.

EMPIRICAL RESULTS

Equation 4 was estimated by using the technique of Ordinary Least Squares (OLS). As OPEC (Organization of Petroleum Exporting Coun- tries) and least-developed countries tend to have rather different characteristics, a subsample of 43 countries was drawn by excluding these countries. Also, different continents may have different kinds of economic infrastructures and socio-political systems. Because of these differences, governmental elites may have set priorities differently so far as the allocation of FPL and FA is concerned. Recognizing this possibility, two other subsamples were drawn--one consisting of 16 Latin American countries an,~ *.he other consisting of 15 African coun- tries. Estimates of Equation 4 for these four panels are presented in Table 1.

The empirical results reported in Table 1 show that the coefficients of FDI/Y are positive and significant in the t test for all panels of regres- sion. On the basis of the OLS estimates obtained, one percentage point rise in the ratio of FDI to GDP leads to a rise in the growth of GDP by 0.6 percentage point in the general case. For the group without OPEC and the least-developed countries, this increases to 0.8 percentage point. The productivity of FDI in Latin American and African Coun- tries appears to be such that one percentage point rise in the ratio of FDI to GDP is associated with the growth of GDP by 2.7 and 2.1 per-

66 Studies in Comparative International Development / Spring 1986

Table 1. Regression Results for Growth of GDP, 1970-1977 (t ratios in parentheses)

constant ~s/Y ~I/~ n~/Y m/~ ~/L R 2 F

I 0.81 0.17"** 0.59** 0.13 0.05 0.27 .322 7.065 (4.56) (2.14) (0.64) (0.18) (0.8)

II -0.67 0.25*** 0.79*** 0.38* 0.10 0.30 .525 8.172 (4.8) (3.06) (1.66) (0.31) (0.92)

Ill 5.39 0.21"* 2.70*** 0.67* -0.83 0.17 .621 5.919 (2.40) (3.65) (2.07) (-1.5) (0.34)

IV-0.05 0.21"* 2.08** -0.76 1.52"* 0.02** .570 4.718 (2.83) (2.33) (1.48) (2.40) (2.35)

I =

II = III= IV =

* statistically significant at = 0.i0 ** statistically significant at = 0.05

*** statistically significant at = 0.01

62 developing countries 43 developing countries e~cluding OPEC and least-developed countries 16 Latin American Developing Countries 15 African developing countries

centage points. This may reflect the high degree ofcapital-intensivity of FDI.

The coefficients of FPL/Y are significant and positive only for two panels of regression (II and III). Countries in these panels belong to the middle-income developing countries. If the regression coefficients are any guide, we may say that middle-income countries benefitted more from the input of FPL than low-income countries insofar as their eco- nomic growth for the period of 1970-1977 is concerned.

The variable FA/Y is not significantly related to the growth of GDP in the case of panels I, II and III. However, it is significant and positive in the case of panel IV. This may provide some support to a hypothesis called "a lagged complementarity of FA and FDI" (Kraska and Taira, 1974). What this means is that countries at low incomes level benefit from FA because it can help to create much needed infrastructures. And once infrastructures are ready, those countries may be attractive to foreign investors. In this sense, FA may be complementary to FD1 in the course of economic development of developing countries.

Domestic savings are certainly the important sources of capital for- mation. Their central importance in the process of economic growth has been recognized in the literature. However, their effectiveness and

Sharma 67

productivity varies from one country to another, or from one group of countries to another. Our results also indicate this. The coefficients of DS/Y are positive and significant in all cases. The incremental capital output ratio (ICOR) appears to be in the range of 4 to 6 (the inverse of the coefficients of DS/Y). This is a credible range for developing coun- tries.

Contrary to the theoretical expectation, the coefficients of dL/L are not significant for all panels of regressions except for the African panel. And the explanation for this is not too difficult to find. In developing countries there are two major components of the labour force: workers employing themselves and workers employed for pay. The paid employ- ment sector is, in large part, the "leading, capitalist" sector a la William Arthur Lewis. Therefore, changes in the rate of growth of employment may be more sensitively related to economic growth than changes in the rate of growth of labour force. But data on paid employment for all the countries in the sample are not available. Hence this proposition could not be tested.

The fit of the regression equations appears to be fairly good. As the multiple coefficients of determinants indicate, the fitted model ex- plains between one-third and two-thirds of the variations in the depen- dent variable. Since the estimating equation does not incorporate technical progress variable, it is quite expected that the model leaves some variations unexplained.

However, these findings need to be interpreted with some caution, keeping in mind some of the limitations the study has. Although the variable "foreign capital" has been disaggregated into FDI, FPL and FA, the debt component could not be further differentiated on the bases of maturity of loans, interest rates and other associated terms. Similarily, the study has noted that entrepreneurship and superior technology tend to be embodied in FDI. But the quantitative method employed in this work has not enabled us to measure their direct impact on eco- nomic growth.

Despite these limitations, the stability of the coefficients of FDI/Y implies that its contribution to economic growth of developing coun- tries cannot be denied. On the other hand, the instability of the coeffi- cients of FPL/Y and FA/Y indicates that no strong generalizations are possible at this stage regarding their effects on the economic growth of developing countries. This is to say, there is strong empirical support to the proposition that the quality of equity capital appears to be far superior to that of debt capital.

68 Studies in Comparative International Development / Spring 1986

SUMMARY AND CONCLUSIONS

This study has examined effects of equity capital and two main forms of debt capital, available through foreign resources, on the eco- nomic growth of 62 developing countries for the period of 1970-1977. FDI is the variable representing equity capital and FPL and FA are the ones representing debt capital for the purpose of this study. The main hypothesis tested is that equity capital is more conductive to economic growth than debt capital.

The empirical results obtained from the econometric analyses of the cross-sectional data indicate a positive and highly significant rela- tionship between equity capital and economic growth. But no such consistently positive and significant relationship between debt capital and economic growth is observed. Only for middle-income countries, the debt capital in the form of FPL is positive and significant.

Even for the middle-income developing countries the impact of eq- uity capital appears to be far greater than that of debt capital in the form of FPL. This becomes evident from a comparison of the magni- tudes of the coefficients of FDI/Y and FPL/Y. Moreover, our data refer to the period of 1970-1977. This was the period before the debt servic- ing burden started to take its toll on the debtor countries. The high growth performance of the middle-income debtor countries in the period examined may have been largely a transient phenomenon. It is because the capital raised to finance growth in this period carried too high a rate of interest. Therefore, the growth rate achieved initially was not sustainable over a long period of time as the high rate of interest charged on the loans would have soon started to eat up more than what would have come out of FPL. The current experience of the middle- income debtor countries has indicated this. For example, the Econo- mist (February 18, 1984) cites one instructive evidence: "The World Bank reckons that the South borrowed $85 billion in medium-term and long-term loans last year, but paid its creditors $96 billion of interest and capital."

In conclusion, equity appears to be better than debt for a sustainable economic growth. Even if one suspects that size and type of countries selected and the time period chosen for analysis may have produced some bias in the estimated coefficients, the stability of the coefficients of equity capital as opposed to debt capital should be reassuring regard- ing the conclusion arrived at in this paper. The findings of this study thus render a basis for making policy decisions as to obtaining foreign capital resources through debt or equity for long-run economic growth

Sharma 69

of developing countries. Put differently, if rapid economic growth is the major national objective of the governments of developing countries, it may be in their best interest to go for FDI. Needless to say, FDI may create some adverse socio-political effects in many developing so- cieties. But our understanding of this dimension of the effects of FDI is rather limited, and this is an important area for further research.

A P P E N D I X List of Sample Countries

1. Algeria 32. Madagascar 2. Argentina 33. Malawi 3. Barbados 34. Malaysia 4. Benin 35. Mali 5. Bolivia 36. Mauritius 6. Brazil 37. Mexico 7. Burma 38. Morocco 8. Cameroon 39. Nicaragua 9. Central African Republic 40. Nigeria

10. Chad 41. Pakistan 11. Chile 42. Panama 12. Columbia 43. Paraguay 13. Congo 44. Peru 14. Costa Rica 45. Philippines 15. Dominican Republic 46. Portugal 16. Ecuador 47. Rwanda 17. E1 Salvador 48. Senegal 18. Ethiopia 49. Sierra Leone 19. Fiji 50. Singapore 20. Ghana 51. Somalia 21. Greece 52. South Korea 22. Guatemala 53. Spain 23. Guyana 54. Sri Lanka 24. Haiti 55. Thailand 25. Honduras 56. Togo 26. India 57. Tunisia 27. Indonesia 58. Turkey 28. Iran 59. Urganda 29. Ivory Coast 60. Upper Volta 30. Jamaica 61. Venezuela 31. Kenya 62. Zambia

N O T E S

The author is grateful to Professors Koji Taira of the University of Illinois at Urbana-Cham- paign, Lorne Switzer of the Concordia University, and Basil Kalymon of the University of Toronto for their generous comments. Two anonymous referees' helpful comments are also gratefully acknowledged. But none of them should be implicated for any remaining errors and deficiencies.

1. The level of incomes of a borrowing nation may be determined either in terms of its natural resources or in terms of market potential.

70 Studies in Comparative International Development / Spring 1986

REFERENCES

BORNSCHIER, VOLKER, CHRISTOPHER CHASE-DUNN, and RICHARD RUBINSON 1978 "Cross-National Evidence of the Effects of Foreign Investment and Aid on Eco-

nomic Growth and Inequality: A Survey of Findings and a Reanalysis?' American Journal of Sociology 84:651-683.

CHENERY, HOLLIS B. and ALAN STROUT 1966 "Foreign Assistance and Economic Development?' American Economic Review

56: 679-733. The Economist

1984 "Beggaring the Poor." The Economist, February 18, 1984, 15-16. GOBALET, J. and L.J. DIAMOND

1979 "Effects of Investment Dependence on Economic Growth." International Studies Quarterly 23:412-444.

GRIFFIN, KEITH 1968 "Coffee and the Economic Development of Columbia." Oxford Bulletin of Eco-

nomics and Statistics 30: 105-127. GRIFFIN, KEITH and J.L. ENOS

1970 "Foreign Assistance: Objectives and Consequences?' Economic Development and Cultural Change 18:313-327.

KRASKA, MICHAEL and KOJI TAIRA 1974 "Foreign Capital, Aid and Growth in Latin America." The Developing Economies

XII: 214-228. PAPANEK, GUSTAV E

1973 "Aid, Foreign Private Investment, Savings, and Growth in Less Developed Coun- tries." Journal of Political Economy 81 : 120-130.

RIEDEL, JAMES 1979 "Economic Dependence and Entrepreneurial Opportunities in Host Country-

MNC Relationship." In Robert G. Hawkins, (ed.), The Economic Effects of Multi- national Corporations. Connecticut: JAI Press, Inc.

ROSENSTEIN-RODAN, EN. 1961 "International Aid for Underdeveloped Countries." Review of Economics and

Statistics 43: 107-138. ROTHGEB, JOHN M.

1984 "The Effects of Foreign Investment on Overall and Sectoral Growth in Third World States." Journal of Peace Research 21: 5-15.

1984-85 The Contribution of Foreign Investment to Growth in Third World States." Stud- ies in Comparative International Development X1X: 3-37.

SHARMA, BASU and KOJI TAIRA 1982 "Foreign Direct Investment, Foreign Aid, Economic Growth and Employment in

Selected Asian Countries?' Asian Economies 42: 5-17. STALLINGS, BARBARA

1982 "Euromarkets, Third World Countries and the International Political Economy" In Harry Makler, et. al., (eds.), The New International Economy. Beverly Hills, California: SAGE Publications.

STONEMAN, COLIN 1975 "Foreign Capital and Economic Growth" World Development 3:11-26.

United Nations Yearbook of National Accounts Statistics, New York: United Nations.

UNCTAD 1980 Handbook of International Trade and Development Statistics. New York: United

Nations Conference on Trade and Development. WEISSKOPE T.

1972 "The Impact of Foreign Capital Inflows on Domestic Savings in Underdeveloped Countries." Journal of International Economics 2: 25-38.

World Bank 1980 World Tables. Baltimore: Johns Hopkins University Press.