THE WORLD BANK Internal Discussion Paper SOUTH ASIA REGION Report No. IDP-127 Some Lessons for South Asia from Developing Country Experience with Foreign Direct Investment Maxwell Fry June 1993 Office of the Chief Economist SASVP MICROGRAPHICS Report No: IDP- 127 Type: IDP The views presented here are those of the author, and they should not be interpreted as reflecting those of the World Bank. SOUTH ASIA REGIONAL SERIES Title AMthE Date Orinnator IDPI 11 How Composition of Public Expenditure Affects Competitiveness: The Case of Bangladesh K. Matin March 1992 P. Mitre (80419) IDP1 21 Labor Retrenchment and Redundancy Compensation in State Owned Enterprises: The Case of Sri Lanka A. Fisabein December 1992 G. Nankani (84641) IDP1 26 Some Guidelines for the Appraisal of Large Projects W. Jack February 1993 A. Estache (81442) IDP1 09 Reforming Higher Secondary Education In in South Asia: The Case of Nepal H. Abadzi May 1993 H. Abadzi (80375) IDPI 27 Some Lessons for South Asia from Developing Country Experience with Foreign Direct Investment M. Fry June 1993 A. Estache (01442) ASIA REGION DISCUSSION PAPERS I/ Title AuthorDt Originto IDP2 The Labor Force Participation of Women in the Republic of Korea: Evolution and Policy Issues C. Grootaert May 1988 F. lqbal IDP15 The Role of Exchange Rate Policy In Four East Asian Countries S.W. Nam May 1988 D. Leipzlger (813881 IDP28 The Small-Scale Enterprise Credit Program (S.S.E.P.) Under the Second and Third Calcutta Urban Development Projects (CUDP II and CUDP 111) - An Assessment F. Kahnert March 1988 F. Kahnert (81413) IDP35 Improving Tax Policy Advice: Lessons and Unresolved issues from Asia Experience H. Flelsig June 1989 H. Fielig (81413) IDP36 Direct Taxes and Fiscal Policy Issues: An illustration for East Asia A. Virmani June 1989 H. Flelsig (81413) 1/ Papers still available upon request. Asia Series discontinued following December 1. 1991 restructuring of the Region. Note: Extra copies may be obtained from the Asia Information Service Center. SOME LESSONS FOR SOUTH ASIA FROM DEVELOPING COUNTRY EXPERIENCE WITH FOREIGN DIRECT INVESTMENT Maxwell Fry June 1993 My thanks to Pedro Alba, St.jn Claessens, Tom Hutchison, Samuel Otoo, Shekhar Shah, and particularly to Antonio Estache for detailed and helpful suggestions on an earlier draft. This paper was written for the International Department and the Office of the Chief Economist, South Asia Region of the World Bank; I am most grateful to the Bank for its financial support. The views expressed in this paper do not necessarily reflect those of the World Bank, its Board of Directors, its management, or any of its member countries. MicroTSP, version 7.0, was used for all graphics and regressions presented in this paper. SOME LESSONS FOR SOUTH ASIA FROM DEVELOPING COUNTRY EXPERIENCE WITH FOREIGN DIRECT INVESTMENT Maxwell Fry Abstract In the aftermath of the debt crisis, foreign direct investment appears to be an increasingly attractive alternative to long-term bank loans as a form of capital inflow to developing countries. Over the past decade, several developing countries have taken measures to attract foreign direct investment. The effects have differed widely from one country to another. Therefore, this paper draws some lessons from the developing country experience with foreign direct investment for South Asia before that part of the developing world embarks upon any serious initiative to attract foreign direct investment. I examine the potential role of foreign direct investment in South Asia from four perspectives; increased capital formation or additional balance-of-payments support; subsequent trade balance effects; crowding out or crowding in and compositional changes to the capital stock; differential effects of FDI under varying types and degrees of economic distortions. The paper presents a five-equation macroeconomic model within a national income accounting framework to determine the effects of FDI on domestic investment, national saving, economic growth, imports, and exports. It then examines the effects of various economic distortions on the productivity of FDI. Table of Contents Summary ............................................ i 1. Introduction ............................................ 1 2. Foreign Direct Investment in the Balance-of-Payments Accounts ... ......... 6 3. The Effects of Distortions and Incentives/Disincentives on the Productivity of Foreign Direct Investment ......................... 8 4. Foreign Direct Investment in a Macroeconomic Framework ..... ......... 12 4.1 The Effect of Foreign Direct Investment on Domestic Investment .................................. 17 4.2 The Effects of Foreign Direct Investment on Saving .............. 19 4.3 The Effects of Foreign Direct Investment on Exports and Imports ....................................... 23 4.4 The Effects of Foreign Direct Investment on Economic Growth ................................... 25 5. TheEmpiricalResults ..................................... 26 5.1 A Test for Additional Financing ........................... 26 5.2 ATestforAdditionalInvestment .......................... 29 5.3 Foreign Direct Investment and National Saving ................. 33 5.4 Foreign Direct Investment and the Rate of Economic Growth ......... 35 5.5 Foreign Direct Investment and Trade Flows ........ . . . . ... . 37 6. Conclusion ........................................... 47 7. References .....................*. . . . ... ..... 0.... .... 51 8. Appendix ........................................... 56 Summary The overall conclusion of this paper is that both the nature and the effects of foreign direct investment (FDI) flows vary significantly between different regions of the developing world. Outside Asia, FDI appears to have been used in large part as a substitute for other types of foreign flows; it has not increased aggregate domestic investment. When the non-Asian countries attracted more FD inflows, national saving, domestic investment, and the rate of economic growth all declined. Hence, PDI appears to have been immiserizing in these countries. In contrast, the role of Fo in South and East Asia has been benign. In these economies, FDI financial flows are matched by increased capital formation that is just as productive as domestically financed investment. To date, inflows of FDI in South Asia have been too small to exert any noticeable macro- economic effects. However, the superior efficiency of FDI in the East Asian economies reflects not only less distorted financial conditions than in other parts of the developing world but also less distorted trading systems. The outward orientation of the East Asian economies ensures that relative prices cannot diverge too far from world market prices. Under these conditions, there are few possibilities for Fo to find high profits in protected markets. The favorable investment climates, however, ensure that FDI flows are readily available without the need for governments to discriminate in favor of this particular form of investment fi- nance. Hence, these economies have avoided the two major pitfalls of Fot, namely, low or negative productivity caused by distortions in the economy and expensive discriminatory incentives provided in the mistaken belief that roM brings positive externalities. It comes as no surprise to find a strong positive correlation between the ratio of domes- tically financed investment to G.sP and the ratio of Fo and GNP. Indeed, inflows of foreign direct and portfolio investment provide good indicators of development performance and po- tential. Policies aimed directly at stimulating these forms of capital inflows appear to be ineffective or to produce the opposite effects to those desired. The evidence suggests over- wheliningly that policies that promote domestic investment and economic growth are most likely to stimulate private sector capital inflows in all forms. By analyzin, FDI in a macroeconomic framework, this paper throws new light on various channels through which FDI can influence saving, investment, growth, and the balance of ii payments on current account. Tue first empirical finding for a sample of 16 developing coun- tries is that rof does not provide additional balance-of-payments financing for a pre-existing current account deficit. Since FD is associated with reduced domestic investment except in South and East Asia, this implies that FD is a close substitute for other capital inflows and may also crowd out domestically financed investment. In South and East Asia, however, FD raises domestic investment by the full extent of the FDI inflow. In these countries, therefore, roM has not been used as a substitute for other types of capital inflows but has increased capital formation and so worsened the current account. Furthermore, in these countries rDI has not crowded out domestically financed investment. Indeed, PD seems to have stimu- ,ated domestically financed investment in South Asia, so worsening the current account by a multiple of the FDI inflow. By increasing domestic investment in these economies, FD has increased growth rates. In examining some other effects, I find that FM has a significantly negative impact on national saving in this sample of developing countries. One possible explanation is that residents may find that terms and conditions for FDI are more favorable than they are for locally financed investment. Hence, they would have an incentive to remove capital from their country and to bring it back again in the.form of FDI. To the extent that these individuals wish to conceal the capital outflow, they will overinvoice imports and underinvoice exports. In such case, an increase in rot would be accompanied by a reduction in recorded national saving. The ratio of rol to domestic investment has not exerted a significant impact on real GNP growth except in the non-Asian country group. In this group, a higher proportion of FDI in aggregate domestic investment is associated with a lower growth rate. Since FM has not increased capital formation in this group of countries, this negative association may be caused by debt crises that reduced output and stimulated FD in the form of debt-equity swaps. The concurrent effect of Fo on exports is negative, except in the East Asian country group. One possible explanation is that FM accommodates export declines in the non- East Asian countries. In contrast, however, F D inflows over the preceding five years are associated with higher export ratios. Given the earlier finding that rol does not increase aggregate domestic'investment outside Asia, the strong positive effect of lagged F D inflows on exports may be caused by a change in the composition of investment. In other words, Fo may crowd out domestically financed investment in countries outside Asia. Unsurprisingly, the investment financed by FD1 seems to be export-oriented. An inflow of FD1 is strongly associated with a higher import ratio. The immediate effect III of FDI, therefore, is to finance a larger import bill. However, rD inflows over the preceding five years are associated with a significant decline in the import ratio in all country groups except East Asia. In these non-East Asian countries, FDI seems to have been directed not only into export industries but also into import-substitution activities. In all country groups, except South Asia where the effects are insignificant, higher lagged FD inflows improve the current account. Outside Asia, FDI inflows over the preceding five years raise exports and reduce imports. Finally, I show that FDI raises the rate of economic growth in the absence of financial repression and trade distortions in the 16 sample developing countries taken together. How- ever, financial repression as measured by the real deposit rate of interest and trade distortions as measured by the black market exchange rate premium can both cause FDI to be immis- erizing. When the domestic economy is distorted, FD inflows are associated with a low or negative growth. When real interest rates are positive, however, FD! can accelerate the rate of economic growth more when restrictions on the sectoral location of this investment are relaxed. The policy implicatiors that might be derived for the South Asian developing countries from this analysis are: * Foreign direct investment can increase capital formation or provide additional balance- of-payments financing but cannot perform both functions at the same time. If FD is attracted for privatization or debt-equity swap programs, it may provide additional or alternative balance-of-payments support, but it will not accelerate capital formation or economic growth. * Deliberately stimulating PD through special incentive schemes may simply encourage roundtrip capital flows from the host country. In such case, measured national saving may fall. * In the presence of financial and trade distortions, FD can remove from the host country more than it contributes. In other words, it can be immiserizing. * The most efficacious way of encouraging FD is to implement policies that generally improve the investment climate. Where domestically financed investment is booming, FD! will seek to participate. Nondiscrimination discourages roundtrip capital flows and reduces the possibilities for immiserization. * Maximum benefit from PDI can be achieved in open economies that are free of domes- iv tic distortions such as financial repression and trade controls. Under such conditions, restrictions on the sectoral location of FDI reduce its growth-enhancing impact. 1 1 Introduction This paper considers the potential role of foreign direct investment (FDI) in South Asia. To date, South Asia has attracted relatively little FD. Figure 1 shows that over the period 1971-1990 net rom flows to India and Bangladesh were zero. Pakistan's FDI inflows have increased from an annual average of 0.13 per cent of GNP in the period 1971-1980 to 0.34 per cent of GNP in the period 1981-1990. Sri Lanka's rol inflows have also risen from 0.26 per cent of oNP in the 1970s to 0.70 per cent of GNP in the 1980s. Although FD rarely exceeds 5 pe. cent of GNP, it averaged 1.8 per cent in Egypt and 3.7 per cent in Malaysia over the period 1971-1990. Hence, the main issue addressed here is what lessons for South Asian can one learn froma the experience with ro1 of other developing countries. Foreign capital inflows to developing countries constitute part of the world's saving. Over the past two decades, world saving as a proportion of world income ha fallen. A comparison of the periods 1968-1981 and 1982-1988 illustrates this worldwide decline in saving ratios (Aghevli et al. 1990, pp. 9 and 36-37): saving in developed countries has fallen from 25 to 20 per cent of G NP and developing country saving has fallen from 25 to 22 per cent of GNP. One important reason for the worldwide decline in saving is rising government deficits: up from 2.9 per cent in the period 1972-1980 to 4.5 per cent in the period 1981-1988~(International Financial Statistics: 1988 and 1991 Yearbooks, p. 156). The decline in world saving implies that not every country can maintain its level of domestic investment by increasing foreign capital inflows. Overall, the decline in saving has to be matched by an equal decline in investment. In fact, saving and investment ratios have fallen in all geographical regions of the world since 1982, but least in developing countries of East and South Asia. As world saving has shrunk, so the world real interest rate has risen from 1.5 per cent during the period 1970-1980 to 4.8 per cent in the period 1981-1991, as illustrated in Figure 2. With no signs of a reversal in the declining trend in global saving and the immediate saving-reducing impacts of the war in the Gulf, reunification of Germany, reconstruction of Eastern Europe, and deliberate current account-reduction policies being implemented by Japan, Korea, and Taiwan, the costs of foreign borrowing can be expected to rise still higher in the 1990s as the saving curve in Figure 2 moves even further to the left. The decline in foreign capital inflows to developing countries has necessitated structural adjustment in the form of an increase in export earnings or a reduction in import expen- diture. The national accounting identities imply that the adjustment has to raise national saving or reduce domestic investment. To maintain or increase rates of economic growth, the 2 Pe Cent of GNP Sri Lanka 125 100 0.75 0.5. . sta. 025. *m W/ ....... om- ......-- -0251 170 172174 176 178 1980 182184 1986 1988 90 Figure 1: Foreign Direct Investment in Four South Asian Economies. adjustment must be in the form of increased exports and increased national saving. Import compression and reduced domestic investment inevitably lower growth rates. However, as Riccardo Faini and Jaime de Melo (1990, p. 492) note: " ... with the significant exception of East Asian countries, adjustment was achieved by cutting investment rather than increasing saving." The inevitable effect has beeh sharp reductions in rates of growth in all parts of the developing world, again with the exception of East and South Asia. It is against this background that For has appeared increasingly attractive to developing countries facing declining domestic investment and higher costs of foreign borrowing. Glob- ally, FDI has increased dramatically over the past decade. However, most of this increase has occurred in the industrial countries. In the developing countries, MDI has been heavily concentrated among a small number of countries. Indeed, Table 1 shows that over 90 per cent of FDI inflows to developing countries in 1990 was received by only 18 countries. Half of this total flowed to eight East Asian developing economies (Hong Kong, Indonesia, Korea, Malaysia, Philippines, Singapore, Taiwan, and Thailand). Given that neither Korea nor Tai- wan has shown strong interest in attracting ro, it may seem surprising that these economies feature in Table 1. The explanation lies in their superlative investment climates (Fry 1991). Over the past decade, many developing countries have taken a fresh look at their policies 3 Real 7 Iterest Savng1 Scmn,,.,. Rdte 2~ . *... ... 4 . 4. 1 i 10 1 20 25' 30.. 35 40 45 SaWing, Invesment (pe cent of GNP) Figure 2: World Saving and Investment, 1970-1991. towards FDI. Since 1982, foreign capital inflows to 'eveloping countries have declined as world real interest rates have increased. In the aftermath of the debt crisis, PDI has appeared to be an increasingly attractive alternative to long-term bank loans as a form of capital inflow to developing countries. Indeed, MD has been viewed by some as a panacea for declining domestic investment and higher costs of borrowing abroad. Over the past decade. several. developing countries have taken measures to attract foreign direct investment. The effects have differed widely from one country to another. Therefore, this paper draws some lessons from the developing country experience with foreign direct investment for South Asia before that part of the developing world embarks upon any serious initiative to attract foreign direct investment. This paper examines the potential role *of ro1 in South Asia by analysing its impact ih a sample of developing countries which includes India, Pakistan, and Sri Lanka, as well as other developing countries such as Brazil, Egypt, and Malaysia that have experienced substantially greater inflows of Fi. Foreign direct investment appears attractive because it involves a risk-sharing relation- ship with investors from the home country. Such risk-sharing does not exist in the formal contractual arrangements for foreign loans. Foreign direct investment appears particularly attractive when existing stocks are low. Low stocks of foreign-owned capital imply low flows 4 Table 1: Foreign Direct Investment Inflows, 1990. Country S millions Industrial Countries 151,970 Developing Countries 32,473 Argentina 2,036 Bermuda 819 Brazil 2,118 Chile 595 China 3,489 Colombia 501 Egypt 947 Hong Kong 783 Indonesia 964 Korea 715 Malaysia 2,902 Mexico 2,632 Nigeria 588 Philippines 530 Saudi Arabia 572 Singapore 4,808 Taiwan 1,330 Thailand 2,376 Turkey 697 Venezuela 451 Source: United Nations, World lamestment Report 199-: Tronsnational Corporations as Engines of Growth (New York: United Nations, 1992), Annex Table 1, pp. 312-316. of repatriated profits. Over time, however, success in attracting FDi will increase this counter- flow, which could exceed the alternative flow of interest payments in the longer run. Clearly, therefore, the question of the cost of FDI to reduce risk must be addressed in any evaluation of the benefits to be derived from substituting FDI for foreign borrowing. The benefits to the host country will depend on both the size of the package of incentives and disincentives to FDi as well as the extent of other distortions in the economy. In the next section, I examine FD within the framework of the balance-of-payments accounts. The main point stressed in this section is that FDI can increase capital formation or provide additional balance-of-payments support. Ceteris peribus, however, it cannot do both at the same time. Given that rD1 can differ in nature from one country to another, section 3 develops an analytical framework for examining the effect of rD on economic growth in the presence of various FDI incentive-disincentive packages as well as other distortions in the economy. This provides one explanation for the widely different relationships between FDi and growth in different groups of developing countries. Section 4 examines FDI in a macroeconomic framework. If FDl influences capital forma- tion either by augmenting it or by changing its composition (by crowding out domestically financed investment), it may subsequently affect the trade balance. Indeed, FD packages of incentives and disincentives typically contain conditions regarding export targets. There- fore, section 4 develops a five-equation model containing behavioral equations for saving, investment, exports, imports, and growth. Section 5 presents and discusses the empirical results. Virtually all the empirical in- vestigations into the causes and consequences of PDi use single-equation models.' Here the five-equation macroeconomic model developed in section 3 is estimated for a sample of 16 developing countries (Argentina, Brazil, Chile, Egypt, India, Indonesia, Korea, Malaysia, Mexico, Nigeria, Pakistan, the Philippines, Sri Lanka, Thailand, Turkey, and Venezuela). I also estimate the model of FDt and economic distortions derived in section 2 using indicators of foreign trade and financial distortions as well as an indicator of FD incentive/disincentive packages in the 16 sample countries. The results provide some new information on the direct and indirect effects of PD inflows to this sample of developing countries. They also provide some strong indicators of the problems to be faced and pitfalls to be avoided should the South Asian economies decide to encourage Fol inflows in the future. 'The exceptions include the simuiarwt .o .wo-eqiaton models used by Lee, Rana, and Iwasaki (1986) and Husain and Jun (1992). 6 2 Foreign Direct Investment in the Balance-of-Payments Accounts The balance-of-payments accounts show that a current account deficit is financed by capital inflows and decreases in official reserves. Another way of presenting this identity is:2 CA+ KA a AR, (1) where CA is the current account,' KA is the capital account, and AR is the change in official reserves. As an item in the balance-of-payments accounts, foreign direct investment (Fo) is one of several capital flows. Other things equal, therefore, an increase in Fo increases capital inflows. If the change in official reserves is unaffected, the increased capital inflow is matched by a smaller current account surplus or a larger current account deficit. The current account can also be defined as the difference between national saving S and domestic investment I: CA =S - I. (2) The most obvious link between FDI and the current account in equation 2 is through do- mestic investment. If Fo finances additional.capital formation in tie host country, it raises domestic investment I. Equation 2 shows that this worsens the current account as required by equation 1. Finally, the current account can be defined as the difference between exports of goods and services X plus net factor income from abroad NFI and imports of goods and services IM: CA aX + NFI - IM. (3) If FDI results in increased capital formation in the host country, the increased investment ,ould involve increased imports of raw materials or capital equipment. Alternatively, it could reduce exports by diverting them into the additional investment. In either case, the current account must deteriorate in equation 3 to exactly the same extent as it does in equations 1 and 2. If FDI finances additional capital formation, equations 1-3 demonstrate that the current account deteriorates by the amount that rD increases capital inflows. In such case, FDI cannot provide additional foreign exchange to finance a pre-existing current account deficit. The extra foreign exchange is entirely absorbed in financing a larger current account deficit. 2This identity ignores errors and omissions 7 An increase in FD could provide additional balance-of-payments financing if it failed to result in additional capital formation in the host country. In undertaking any analysis of roi, one must recognize that DI data record financial fiows which may or may not correspond to changes in capital formation. Whether or not they do depends on the extents of crowding out of domestically financed investment and substitutability of this type of financial flow for other types of financial flows. Suppose that PDI capital inflows were used for new capital formation. If this capital formation would have taken place in any event or if this capital formation deters an equal amount of domestically financed investment in other projects, then total domestic invest- ment I remains unchanged. In such case, FDI does not affect the current account unless it changes the level of national saving. If the current account and the change in official reserves remain the same, this rDo becomes one of the sources of finance for the pre-existing current account deficit. An inflow of PD also provides balance-of-payments financing if it is not used for new cap- ital formation. For example, privatization programs in a number of countries have produced capital inflows in the form of rDo. The privatization has resulted in foreign ownership of an existing company and its capital assets, but not necessarily in any new capital formation. A capital inflow used to acquire ownership of an existing firm is recorded as FDI if it achieves ownership of 10 per cent or more in the company. Again, this PDi is not accompanied by additional capital formation and so does not increase the current account deficit. It does, therefore, provide additional or alternative balance-of-payments financing. Some of the literature on FD! Suggests that FDI can serve two purposes, namely, increase investment and relieve foreign exchange shortages. For example, Laurence Cockcroft and Roger Riddell (1991, p. 3) note: "Two of the principal factors inhibiting higher levels of economic growth in Sub-Saharan Africa in the 1990s are low levels of investment and foreign exchange shortages. The first attraction of foreign investment lies in its potential to address both these constraints." Unless FD! affects national saving, however, PDI can increase do- mestic investment or provide additional financing for a pre-existing current account deficit or achieve some combination of the two, but the linear combination of these two effects must always sum to one. In the more aggregated balance-of-payments accounts, such as those presented in Inter- national Financial Slatistics that are used in this study, FDI is recorded as a net inflow of capital. It is net in two senses: first, it nets out capital outflows; second, it nets out profit repatriation. Hence, two countries with similar FD could experience different changes in 8 foreign ownership. In one country, there might be no Fo outflow nor any repatriated profits. Hence, $1 million FDI implies a transfer of $1 million of foreign capital into ownership of companies in the host company. In another country, however, a $3 million FDo inflow and $2 million FDI outflow would imply a much larger increase in foreign ownership despite the same net flow of roM. In any event, control over a company does not require 100 per cent ownership. Further- more, 100 per cent ownership can be achieved without 100 per cent equity financing; funds can be borrowed in the host country to finance part of the acquisition or establishment of a company. Hence, the recorded net inflow of FDI may bear little or no relation to the extent of foreign control in the economy. Were one investigating the hypothesis that FDo produced more or less productive projects than domestically financed investment, one would examine data on individual companies that could be classified as foreign or domestically owned. On the other hand, if one wishes to address the question of the macroeconomic impact of rD across a number of countries, aggregate FDo flow data have to be used. 3 The Effects of Distortions and Incentives/Disincentives on the Productivity of Foreign Direct Investment The previous section stresses the point*that rDl can take different forms in different countries. In some countries, FD! may augment capital formation, while in others it may represent a substitute capital flow or crowd out domestically financed investment. These different forms of FDI are likely to have different effects on a country's rate of economic growth. Hence, this section focuses on the effect of distortions in an economy on the productivity of FoI which, in turn, affects the rate of economic growth. As the World Bank (1989, pp. 29-31) points out: "Historically, the quality of investment has been at least as important for growth as the quantity. Although the fastest-growing countries had higher rates of investment than the others, empirical studies generally find that less than half the growth in output is attributable to increases in labor and capital. Higher productivity explains the rest. ... Faster growth, more investment, and greater financial depth all come partly from higher saving. In its own right, however, greater financial depth also contributes to growth by improving the productivity of investment." An increasing body of evidence suggests that qualitative differences in investment are far more important than quantitative differences in explaining different rates of growth across countries (Fry 1988, Ch. 6, King and Levine 1993a, King and Levine 1993b, Roubini and 9 Sala-i-Martin 1992). These productivity differentials have been traced to trade distortions and financial repression imposed on the economy by government policy (Dollar 1992, Roubini and Sala-i-Martin 1991). Trade distortions manifest themselves in a set of relative prices that deviate substantially from relative prices in the world economy. Relative prices in the economies of South Asia deviate considerably from relative world market prices. As the World Bank (1991, p. 95) points out: " ... direct foreign investment in an economy with highly distorted policies is likely to generate net losses for the host country instead of welfare gains." Indeed, the theory of immiserizing growth might well apply most forcefully in the case of rol simply because PD1 that produces negative value added at world prices can be accompanied by the removal of resources in the form of repatriated profits from the country. Therefore, this section presents a framework for examining the effects of distortions in the economy, including distortions created by packages of incentives and disincentives for PDI, on the efficiency of rD. It also suggests the possibility of some interaction between the incentive-disincentive package and other distortions in the economy on the efficiency of FDI. Venkataraman Balasubramanyam (1984, pp. 732-733) concludes that incentives for FD! are generally offered to offset"... a complex web of controls and regulations." Sail Lizondo (1991, p. 79) also notes: "Incentives are seldom granted without conditions; instead, they are usually subject to the compliance of requirements that constitute disincentives to foreign.. direct investment." One might therefore consider sets of inceutive-disincentive packages of varying sizes and complexity that all attract exactly the same quantity of FDI. In a neo- classical world, any package which maintains the quantity of PD by offsetting constraints such as local content requirements or ownership limitations with incentives such as tax holidays must reduce the overall efficiency of Fo. This efficiency-reducing effect is monotonic with efficiency declining as the size of the package increases, as illustrated in Figure 3. Figure 3 provides a three-dimensional illustration of the possible interaction between package size and other distortions in the economy. With no other distortions, efficiency declines monotonically as the package size is increased. In the other plane, efficiency of FDI also declines monotonically as other distortions in the-economy increase. In the presence of other distortions, however, a package of incentives and disincentives towards FDi could be welfare improving, at least over some range. A rising segment of the plane showing efficiency. improving effects of increasing the package size when other distortions exist is also illustrated in Figure 3. There are several ways in which this interaction could occur. In a growth rate function, 10 Change in Total Factor Productivity ATFP Package Size Figure 3: Incentive-Disincentive Packages and Other Domestic Distortions. FDIY would be included by itself and also interacted with indices measuring both the package size P and the extent of other distortions D: YG = f(FDIY, P, D, FDIY *P, FDIY D, FDIY *P*D). (4) The same functional form could be used to estimate the magnitude of these interactive effects on the change in total factor productivity A TFP* ATFP = f(FDIY, P, D, FD1Y*P, FDIY.D, FDIY*P-D). (5) This is the expression illustrated in Figure 3 for a fixed value of FDIY. The empirical evidence suggests that some distortions can reduce investment efficiency quite considerably. Whether or not package size exerts effects of similar magnitudes for any given level of distortion remains to be tested. Casual observation of differing policies towards FDI in the East Asian developing economies might suggest that the distortion slope is considerably steeper than the package size slope. However, as Hal Hill (1990, p. 43) concludes: *'... the central issue of whether what may conveniently be termed the 'Singapore' or the 'Korea' strategy is preferable, from either host or home country perspective, has not -been addressed carefully in the literature." 11 This analytical framework for examining the effect of FDI on economic growth in the pres- ence of incentive-disincentive packages and other distortions in the economy is designed for empirical investigation. Unfortunately, however, the available data on incentive-disincentive packages are qualitative rather than quantitative. Nevertheless, the approach is illustrated by estimates of the differential impact of FDi on economic growth in the presence of foreign trade and financial distortions as well as changes in incentives and disincentives to PDI. For this study, I use data on seven qualitative indicators of changes in FDI incentive* disincentive packages compiled by Farok Contractor (1990, data appendix) for the United Nations Centre on Transnational Corporations (1991). Contractor's data set includes 14 of the 16 sample developing countries analyzed here.3 I also use indicators of financial and trade distortions in the form of the real deposit rate of interest and the black market foreign exchange premium. Both these variables were extracted from the World Development Report 1991: Supplementary Data diskettes. One way of analyzing the efficiency-improving role of financial intermediation starts by recognizing the fact that when real interest rates are negative there is no incentive to use capital efficiently. Excess capacity is costless, so plants are built with far more capacity than required for immediate production plans. Overtime, shift work, and other measures that increase the effective utilization of plant and machinery are not worthwhile when keeping the capital stock idle is costless. Under such circumstances, the measured capital stock exceeds the effective capital stock. For example, the effective capital stock might be equivalent to 66 per cent of the measured capital stock when the real interest rate is -15 per cent. However, the effective capital stock might equal the measured capital stock at a real interest rate of 5 per cent. In this example, therefore, the effective capital stock can be expressed as the actual capital stock times (0.915 + 1.7r), where r is the real interest rate expressed in proportional rather than percentage terms. If financial intermediaries allocate investible funds more efficiently than other allocative mechanisms, then greater financial depth caused by higher real deposit rates of interest itself improves the quality of investment. To the extent that rol is combined with national saving through joint participation or borrowing from the host country's financial institutions, the efficiency of FoI will be negatively affected in the same way as domestically financed investment by institutional interest rates that are held below their free-market equilibrium levels. 3Unfortonately, Contractor does not compile data for India or Sri Lanka. 12 4 Foreign Direct Investment in a Macroeconomic Framework Table 2 presents the macroeconomic model derived in this section; hats denote the endogenous variables. The extent and financing of a current account deficit depend both on a country's desire to spend more than its income and on the willingness of the rest of the world to inance the deficit from its saving. In other words, a current account deficit is determined simultaneously by both the demand for and the supply of foreign saving. My current account model attempts to capture.the essential determinants of this interactive process. It also permits the ratio of foreign debt to GNP to converge to a constant and hence sustainable steady state. A steady state exists if a higher level of foreign indebtedness improves the current account." If foreign indebtedness reduces investment by more than it reduces saving, or raises investment by less than it raises saving, rising foreign indebtedness improves the current account and so slows down the buildup of foreign debt. The key components of the model are represented in Figure 4. This figure echoes Lloyd Metzler (1968) in viewing the current account deficit as the difference between domestic investment and national saving. It shows the planned levels of national saving, foreign saving and domestic investment at different levels of inflation-adjusted or real interest rates. The domestic investment function I slopes downwards indicating that there is more investment at lower interest rates. The national saving function Sn is nearly vertical indicating that national saving does not vary greatly with changes in the domestic real interest rate. Most developing countries face an upward-sloping supply curve of foreign saving Sfo. However, the effective cost at which foreign saving begins to be supplied in any particular year depends on the country's debt position inherited from past borrowing. The effective cost of foreign borrowing is also the effective domestic real interest rate. At an effective interest rate of ro, domestic investment Io exceeds national saving Sno. Hence, the inflow of foreign saving is positive and the country runs a current account deficit on its balance of payments equal to lo-Sno. The accumulation of debt resulting from the current account deficit in year -0 raises the foreign saving curve to SA. This change produces an effective cost of foreign borrowing of ri in year 1. In this case, foreign debt accumulation reduces domestic investment and rises national saving through a higher domestic real interest rate. As this process continues in subsequent years, the current account deficit declines until it reaches a steady-state equilib- 4'This constitutes an informal error-correction process. 13 Table 2: A Macroeconomic Model of Foreign Direct Investment. 0/+ ly = b1o + bu fDIY + 612FLYt-1 + t3DDCY + 614RW + bsREXLt-j + LeY + b7YG,-i + lY-i. (6) 0/- ? ? SNY = b +b2l FD1Y +bnFLY-, +62aRW + Y24 . 0/- , + b25 (YG Ff]Y) + 126(YG* FLY-1) + ?27(Yd RW) + 12sSY-1. (7) oi+ + 0 V 0/+ IMY bo+ b3 FDY + b32REXL +sTTL + 1a4AY + basXY+ basOAY 0/+ 0+ + + bv7FD1Y-1 + bUsOAY-1 + balMAY-1-. (8) 0/++ + XY = 6a+ b4'FD' + 42 RM&c + b43oTy 0i+ ..0/+ + + b44 FDIY-i + 64s OAY-i + 64slMAYt-. (9) YG = 6so+6s FDII + bs2AY + bwDDCY + 6A5XG. (10) rium in which the debt/GNP ratio is constant. Here then is an explanation for the positive effect of foreign debt on the U.S. current account detected by Paul Masson, Jeroen Kremers, and Jocelyn Horne (1994) and Michael Wickens and Merih Uctum (1993); higher foreigh indebtedness eventually improves the current account and so reverses or reduces the debt buildup. This is the stabilizing financial effect of foreign debt accumulation. Much foreign debt in developing countries takes the form of government and government- guaranteed foreign debt. The level of this type of foreign debt accumulated from past current account deficits may itself alfect the position of the saving function in Figure 4. Presumably, the modern Ricardian equivalence view would hold that if households expect the existence of 14 Table 2: A Macroeconomic Model of Foreign Direct Investment (Concluded). Eadogenoues Variables 1Y - Domestic investment/oNP (current prices). FDIY - Net inflow of foreign direct investment/oNP (dollar values converted to domestic currency, current prices). DDCY - Change in domestic credit/oNP (current prices). YG - Rate of growth in oNP (constant prices, continuously compounded). SNY - National saving/oNP (current prices). IMKY - Imports/GNP (constant prices). REXL - Real exchange rate [(domestic ON? deflator/U.S. wholesale price index)/domestic currency per U.S. dollar]. IKY - Domestic investment/ONP (constant prices). XY - Exports/oNP (current prices). OKY - Other capital flows/oNP (dollar values converted to domestic currency, current prices). XKY - Exports/ONP (constant prices). FDII - Net inflow of foreign direct investment/domestic investment (dollar values con- verted to domestic currency, current prices). XKG - Rate of growth in exports (constant prices, continuously compounded). Exogenous or Predetermined Variables TTL - Terms of trade (export price index/import price index). RW - World real interest rate (6-month LIBot deposit rate minus U.S. inflation, contin- uously compounded). FLY - Cumulated net foreign liabilities/oNP (dollar values converted to domestic cur- rency, current prices). government-guaranteed foreign loans to necessitate government expenditure and hence higher taxation in the future, private saving would rise as more guarantees are extended. Hence, the Ricardian equivalence hypothesis suggests that more foreign debt could actually raise the national saving ratio, since this future contingent government liability does not reduce the current level of government saving. Alternatively, however, as households notice that government and government-guaranteed foreign debt is rising and anticipate higher future tax burdens for its servicing, they will have 15 Reel Sf, Sn Sn + Sf, Rate / ftSo / Sn + Sfo I l.00 r .1 I0 -.i . . . . . . . . . . . . Sn Sni It lo SWng. Ivetment Figure 4: Domestic investment, Notional Soving and the Current Account Deficit an increasing incentive to transfer assets abroad, as foreseen by David Ricardo (1817, pp. 338): A cotiitry which has accumulated a large debt is placed in a most artificial situ- ation; and although the amount of taxes, and the increased price of labour, may not, and I believe does not, place it under any other disadvantage with respect to foreign countries, except the unavoidable one of paying those taxes, yet it becomes the interest of every contributor to withdraw his shoulder from the burthen, and to shift this payment from himself to another; and the temptation to remove himself and his capital to another country, where he will be exempted from such burthens, becomes at last irresistible, and overcomes the natural reluctance which every man feels to quit the place of his birth, and the scene of his early associations. A country which has involved itself in the difficulties attending this artificial system, would act wisely by ransoming itself from them, at the sacrifice of any portion of its property which might be necessary to redeem its debt. Savers could also perceive that a high and rising foreign debt ratio may goad the government into stimulating exports, which would involve a devaluation in the real exchange rate. In this case, the real returns on assets held abroad could be higher than the real returns on domestic assets. Hence, one might expect a higher value of foreign debt to reduce measured national 16 saving, implying a leftward shift in the saving function in Figure 4. The magnitude of capital flight caused by a ,- Idup of foreign debt can be, and in several developing countries has been, destabilizing. Instead of an increase in foreign debt reducing domestic investment and increasing national saving, the foreign debt buildup shifts the na- tional saving function to the left; hence, the current account deficit increases. Real interest rates can reach, and in several developing countries have reached, astronomical levels with- *out reducing the current account deficit. The end result is financial and economic paralysis. This then is the explanation of a negative effect of foreign debt on the current account; an increase in foreign debt can worsen the current account. This is the destabilizing fiscal effect of foreign debt accumulation which may or may not outweigh the stabilizing financial effect of foreign debt accumulation discussed above. A country's foreign indebtedness can be estimated by cumulating the current account deficit over time; here one must use the balance-of-payments definition of the current account, which subtracts unrequited transfers from the measured deficit. The dollar stock of net foreign liabilities derived in this way can then be converted into domestic currency and expressed as a ratio of GNP. The variable representing these net foreign liabilities FLYt-1 is last year's stock of cumulated net foreign liabilities converted into domestic currency and divided by last year's GNP. The main problem in estimating the model sketched in Figure 4 is that effective costs of foreign borrowing or domestic shadow interest rates, ro and r, in Figure 4, are unobservable. As Vassilis Hajivassiliou (1987, p. 205) points out: "The spread over the London interbank offer rate (LIBOR) does not perform the key role of clearing the market for international loans. Instead allocation of s 1 1 - - n.s. n.s. National Saving - n.s. n.s. - n.s. - n.s. Economic Growth n.s. n.s. n.s. n.s. - n.s. n.s. Exports - n.s. n.s. - - - n.s. Imports 1 n.s. a 1 + + + + Exports (lagged FDI) + n.s. + n.s. + + + Imports (lagged FDI) - n.s. + n.s. - - n.s. Note: u.s.: Not significantly different from zero at the 95 per cent confidence level. channels through which rDT can influence saving, investment, growth, and the balance of payments on current account. Table 10 sets out the direct effects of larger FDI inflows in the seven groups of developing economies examined in this paper. The first empirical finding for a sample of 16 developing countries is that FDI does not provide additional balance-of- payments financing for a pre-existing current account deficit. Since rD! is associated with reduced domestic investment except in South and East Asia, this implies that FDI is a close substitute for other capital inflows and may also crowd out domestically financed investment. In South and East Asia, however, rDo raises domestic investment by the full extent of the rD inflow. In these countries, therefore, Mol has not been used as a substitute for other types of capital inflows but has increased capital formation and so worsened the current account. Furthermore, in these countries rol has not crowded out domestically financed investment. Indeed, FDI seems to have stimulated domestically financed investment in South Asia, so worsening the current account by a multiple of the PDI inflow. By increasing domestic investment in these economies, FDI has increased growth rates. In examining some other effects, I find that PDI has a significantly negative impact on national saving in this sample of developing countries. One possible explanation is that residents may find that terms and conditions for FD are more favorable than they are for locally financed investment. Hence, they would have an incentive to remove capital from their country and to bring it back again in the form of FDI. To the extent that these individuals wish to conceal the capital outflow, they will overinvoice imports and underinvoice exports. In such case, an increase in FDI would be accompanied by a reduction in recorded national saving. 49 The ratio of F DI to domestic investment has not exerted a significant impact on real GN P growth except in the non-Asian country group. In this group, a higher proportion of FDI in aggregate domestic investment is associated with a lower growth rate. Since roM has not increased capital formation in this group of countries, this negative association may be caused by debt crises that reduced output and stimulated FD1 in the form of debt-equity swaps. The concurrent effect of FDI on exports is negative, except in the East Asian country group. One possible explanation is that roM accommodates export declines in the non- East Asian countries. In contrast, however, FDI inflows over the preceding five years are associated with higher export ratios. Given the earlier finding that FD does not increase aggregate domestic investment outside Asia, the strong positive effect of lagged FD! inflows on exports may be caused by a change in the composition of investment. In other words, FD may crowd out domestically financed investment in countries outside Asia. Unsurprisingly, the investment financed by FDI seems to be export-oriented. An inflow of FDI is strongly associated with a higher import ratio. The immediate effect of FDI, therefore, is to finance a larger import bill. However, FD inflows over the preceding five years are associated with a significant decline in the import ratio in all country groups except East Asia. In these non-East Asian countries, FD seems to have been directed not only into export industries but also into import-substitution activities. In all country groups, except South Asia where the effects are insignificant, higher lagged FDI inflows improve the current account. Outside Asia, FDI inflows over the preceding five years raise exports and reduce imports. Finally, I show that FD! raises the rate of economic growth in the absence of financial repression and trade distortions in the 16 sample developing countries taken together. How- ever, financial repression as measured by the real deposit rate of interest and trade distortions as measured by the black market exchange rate premium can both cause FD to be immis- erizing. When the domestic economy is distorted, PD inflows are associated with a low or negative growth. When real interest rates are positive, however, Mol can accelerate the rate of economic growth more when restrictions on the sectoral location of this investment are relaxed. The policy implications that might be derived for the South Asian developing countries from this analysis are: * Foreign direct investment can increase capital formation or provide additional balance- of-payments financing but cannot perform both functions at the same time. If FDI is attracted for privatization or debt-equity swap programs, it may provide additional or 50 alternative balance-of-payments support, but it will not accelerate capital formation or economic growth. * Deliberately stimulating Fo1 through special incentive schemes may simply encourage roundtrip capital flows from the host country. In such case, measured national saving may fall. * In the presence of financial and trade distortions, FDl can remove from the host country more than it contributes. In other words, it can be immiserizing. * The most efficacious way of encouraging rD is to implement policies that generally improve the investment climate. Where domestically financed investment is booming, pDi will seek to participate. Nondiscrimination discourages roundtrip capital flows and reduces the possibilities for immiserization. * Maximum benefit from FD can be achieved in open economies that are free of domes- tic distortions such as financial repression and trade controls. Under such conditions, restrictions on the sectoral location of Mor reduce its growth-enhancing impact. 51 References Agarwal, Jamuna P., Andrea Gubits, and Peter Nunnenkamp (1991), Foreign Direct Invest- ment in Developing Countries: The Case of Germany (Tubingen: J.C.B. Mohr (Paul Siebeck), Kieler Studien 238, 1991). Aghevli, Bijan B., James M. Boughton, Peter J. Montiel, Delano Villanueva, and Geoffrey Woglom (1990), The Role of National Saving in the World Economy: Recent Trends and Prospects (Washington, D.C.: International Monetary Fund, Occasional Paper 67, March 1990). Altken, Brian and Ann Harrison (1992), "Does Proximity to Foreign Firms Induce Technology Spillovers? Evidence from Panel Data" (Cambridge, Mass.: Massachusetts Institute of Technology, July 1992). Balasubramanyam, Venkataraman N. (1984), "Incentives and Disincentives for Foreign Direct Investment in Less Developed Countries," Weltwirtschaftliches Archiv, 1984, pp. 720-735. Blejer, Mario I. and Mohsin S. Khan (1984), "Government Policy and Private Investment in Developing Countries," International Monetary Fund Staff Papers, 31(2), June 1984, pp. 379-403. Blinder, Alan S. and Joseph E. Stiglits (1983), "Money, Credit Constraints, and Economic Activity," American Economic Review, 73(2), May 1983, pp. 297-302. Calvo, Guillermo A. and Fabrizio Coricelli (1991), "Stagflationary Effects of Stabilization Programs in Reforming Socialist Countries: Enterprise-Side vs. Ho.,sehold-Side Factors" (Washington, D.C.: International Monetary Fund, 1991). Cockcroft, Laurence and Roger C. Riddell (1991), "Foreign Direct Investment in Sub-Saharan Africa" (Washington, D.C., International Economics Department, WPS 619: World Bank, March 1991). Contractor, Farok J. (1990), "Do Government Policies towards Foreign Investment Matter? An Empirical Investigation of the Link between National Policies and FDI Flows" (New Brunswick, NJ: Rutgers University, Graduate School of Management, 1990). Cuddington, John T. (1986), "Capital Flight: Estimates, Issues, and Explanations," Prince- ton Studies in International Finance, (58), December 1986. De Gregorio, Jose and Pablo E. Guidotti (1992), "Financial Development and Economic Growth" (Washington, D.C.: International Monetary Fund, WP/92/101, December 1992). 52 Dollar, David (1992), "Outward-oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs, 1976-1985," Economic Development and Cultural Change, 40(3), April 1992, pp. 523-544. Dooley, Michael P. (1986), "Country-Specific Risk Premiums, Capital Flight and Net Invest- ment Income Payments in Selected Developing Countries" (Washington, D.C.: Interna- tional Monetary Fund, DM/86/17, March 1986). Dooley, Michael P. (1990), "Comment" in Taxation in the Global Economy edited by Assaf Razin and Joel Slemrod (Chicago: University of Chicago Press for the National Bureau of Economic Research, 1990), pp. 74-78. Edwards, Sebastign (1986), "The Pricing of Bonds and Bank Loans in International Markets: An Empirical Analysis of Developing Countries' Foreign Borrowing," European Economic Review, 30(3), 1986, pp. 565-589. Faini, Riccardo and Jaime de Melo (1990), "Adjustment, Investment and the Real Exchange Rate in Developing Countries," Economic Policy: A European Forum, (11), October 1990, pp. 491-519. Feder, Gershon (1982), "On Exports and Economic Growth," Journal of Development Eco- nomics, 12(i), February-April 1982, pp. 59-73. Fry, Maxwell J. (1980), "Money, Interest, Inflation and Growth in Turkey," Journal of Mon- etary Economics, 6(4), October 1980, pp. 535-545. Fry, Maxwell J. (1988), Money, Interest, and Banking in Economic Development (Baltimore: Johns Hopkins University Press, 1988). Fry, Maxwell J. (1989), "Foreign Debt Instability: An Analysis of National Saving and Do- mestic Investment Responses to Foreign Debt Accumulation in 28 Developing Countries," Journal of International Money and Finance, 8(3), September 1989, pp. 315-344. Fry, Maxwell J. (1991), "Domestic Resource Mobilization in Developing Asia: Four Policy Issues," Asian Development Review, 9(1), 1991, pp. 15-39. Fry, Maxwell J. (1993), "Foreign Direct Investment in a Macroeconomic Framework: Some Further Findings" (Birmingham: University of Birmingham, International Finance Group, IFGWP-93-03, June 1993). Fry, Maxwell J. and David M. Lilien (1986),"Monetary Policy Responses to Exogenous Shocks," American Economic Review, Papers and Proceedings, 76(2), May 1986, pp. 79- 83. 53 Fry, Maxwell J. and Andrew Mason (1982), "The Variable Rate-of-Growth Effect in the Life Cycle Saving Model: Children, Capital Inflows, Interest and Growth in a New Specifica- tion of the Life-Cycle Model Applied to Seven Asian Developing Countries," Economic Inquiry, 20(3), July 1982, pp. 426-442. Goldstein, Morris, Donald J. Mathieson, and Timothy Lane (1991), "Determinants and Sys- temic Consequences of International Capital Flows" in Determinants and Systemic Con- sequences of International Capital Flows edited by Research Department of the Inter- national Monetary Fund (Washington, D.C.: International Monetary Fund, Occasional Paper No. 77, March 1991), pp. 1-45. Haddad, Mona and Ann Harrison (1992), "Are there Positive Spillovers from Direct Foreign Investment? Evidence from Panel Data for Morocco" (Washington, D.C.: World Bank, May 1992). Hajivassiliou, Vassilis A. (1987), "The External Debt Repayments Problems of LDCs: An Econometric Model Based on Panel Data," Journal of Econometrics, 36(1-2), September- October 1987, pp. 205-230. Hill, Hal (1990), "Foreign Investment and East Asian Economic Development," Asian-Pacific Economic Literature, 4(2), September 1990, pp. 21-58. Husain, Ishrat and Kwang W. Jun (1992), "Capital Flows to South Asian and ASEAN Coun- tries" (Washington, D.C.: World Bank, International Economics Department, WPS 842, January 1992). Johnston, John (1984), Econometric Methods, third edition (New York: McGraw-Hill, 1984). Keller, Peter M. (1980), "Implications of Credit Policies for Output and the Balance of Payments," International Monetary Fund Staff Papers, 27(3), September 1980, pp. 451- 477. Khan, Mohsin S. and Nadeem Ul Haque (1985), "Foreign Borrowing and Capital Flight: A Formal Analysis," International Monetary Fund Staff Papers, 32(4), December 1985, pp. 606-628. King, Robert G. and Ross Levine (1993a), "Finance and Growth: Schumpeter Might Be Right," Quarterly Journal of Economics, 108(3), July 1993, forthcoming. King, Robert G. and Ross Levine (1993b), "Finance, Entrepreneurship, and Growth: Theory and Evidence," Journal of Monetary Economics, 32, 1993, forthcoming. 54 Lee, Jungsoo, Pradumna B. Rana, and Yoshihiro Iwasaki (1986), "Effects of Foreign Capital Inflows on Developing Countries In Asia" (Manila: Asian Development Bank, Economic Staff Paper No. 30, April 1986). Lizondo, J. Sadil (1991), "Foreign Direct Investment" in Determinants and Systemic Con- sequences of International Capital Flows edited by Research Department of the Inter- national Monetary Fund (Washington, D.C.: International Monetary Fund, Occasional Paper No. 77, March 1991), pp. 68-82. Mason, Andrew (1987), "Nationil Saving Rates and Population Growth: A New Model and New Evidence" in Population Gowth and Economic Development: Issues and Evidence edited by D. Gale Johnson and Ronald D. Lee (Madison: University of Wisconsin Press for the National Academy of Sciences, 1987), pp. 523-560. Masson, Paul, Jeroen Kremers, and Jocelyn Home (1994), "Net Foreign Assets and Interna- tional Adjustment: The United States, Japan, and Germany," Journal of International Money and Finance, 13(1), February 1994, forthcoming. Maxwell Stamp Ltd. (1991), Rejional Study of Resource Flows and Financial Intermediation (London: Maxwell Stamp, October 1991). Meade, James E. (1951), The Theory of International Economic Policy. Volume 1: The Balance of Payments (Oxford: Oxford University Press, 1951). Metzler, Lloyd A. (1968), "The Process of International Adjustment under Conditions of Full Employment: A Keynesian View" in Readings in International Economics edited by Richard E. Caves and Harry G. Johnson (Homewood, Illinois: Richard D. Irwin for the American Economic Association, 1968), pp. 465-486. Molho, Lazaros E. (1986), "Interest Rates, Saving, and Investment in Developing Countries: A Re-examination of the McKinnon-Shaw Hypothesis," International Monetary Fund Staff Papers, 33(1), March 1986, pp. 90-116. Naya, Seiji (1990), "Direct Foreign Investment and Trade in East and Southeast Asia" in The Political Economy of International Trade: Essays in Honor of Robert E. Baldwin edited by Ronald W. Jones and Anne 0. Krueger (Cambridge, Mass.: Basil Blackwell, 1990), pp. 288-12. Peng, WenSheng (1993), "The Effects of Inflation on Natural Output Growth in Six Pacific Basin and Latin American Developing Economies" (Birmingham: University of Birming- ham, International Finance Group Working Paper, IFGWP-93-02, April 1993). 55. Polak, Jacques J. (1989), Financial Policies and Development (Paris: Development Centre of the Organisation for Economic Co-operation and Development, 1989). Rana, Pradumna B. and J. Malcolm Dowling (1990), "Foreign Capital and Asian Economic Growth," Asian Development Review, 8(2), 1990, pp. 77-102. Ricardo, David (1817), On the Principles of Political Economy, and Taxation (London: John Murray, 1817). Roubini, Nouriel and Xavier Sala-i-Martin (1991), "Financial Development, the Trade Regime, and Economic Growth" (Cambridge, MA: National Bureau of Economic Re- search, NBER Working Paper No. 3876, October 1991). Roubini, Nouriel and Xavier Sala-i-Martin (1992), "Financial Repression and Economic Growth," Journal of Development Economics, 39(1), July 1992, pp. 5-30.. Turner, Philip (1991), Capital Flows in the 1980s: A Survey of Major Trends (Basel: Bank for International Settlements, Monetary and Economic Department, BIS Economic Papers No. 30, 1991). United Nations Centre on Transnational Corporations (1991), Government Policies and For- eign Direct Investment (New York: United Nations, UNCTC Current Studies, Series A, No. 17, November 1991). United Nations (1992), World Investment Report 199*: Transnational Corporations as En- gines of Growth (New York: United Nations, 1992). Watson, C. Maxwell, G. Russell Kincaid, Caroline Atkinson, Eliot Kalter and David Folkerts- Landau (1986), International Capital Markets: Developments and Prospects (Washing- ton, D.C.: International Monetary Fund, World Economic and Financial Surveys, De- cember 1986). Wickens, Michael and Merih Uctum (1993), "The Sustainability of Current Account Deficits: A Test of the US Intertemporal Budget Constraint," Journal of Economic Dynamics and Control, 17, 1993, forthcoming. World Bank (1989), World Development Report 1989 (New York: Oxford University Press for the World Bank, 1989). World Bank (1991), World Development Report 1991 (New York: Oxford University Press for the World Bank, 1991). 56 Appendix Tables 1 and 2 show the investment and saving estimates when the rate of economic growth is omitted. Tables 3 and 4 show the export and import estimates with the real exchange rate excluded. Tables 5 and 6 show the real exchange rate REXL estimates.1 In Table 5, all the coefficients in the estimate for the complete sample take the opposite signs to those in the import demand function. Since the real exchange rate equation is simply the import demand equation rearranged in terms of price rather than quantity, except that lagged price rather than lagged quantity is included, this is the result one expects. For completeness, Table 6 shows the real exchange rate estimates with 5-year lagged Fo. Tables 7 and 8 show estimated current account ratios CAY. Since FDI affects the rate of economic growth and growth affects both domestic investment and national saving, FDI exerts both direct and indirect effects on the current account as a ratio of GNP CAY. The simplest way of determining the overall effect is to estimate a quasi-reduced form current account equation for the sample countries. Table 7 shows that the coefficient of FDIY is not significantly different from -1 in any of the estimates except that for South Asia. Since the direct negative saving effect is offset to some extent by a direct negative investment effect, FDIY appears to exert a negative indirect effect through the rate of economic growth. Table 8 shows the current account CAY estimates with 5-year lagged FDL. 'Although REXL is divided by 10 for all other estimates, it is not adjusted in equatios that use it as the dependent variable. 57 Table 1: Domestic Investment Estimates IY without Growth Rate Variable All S. Asia E. Asia L.A. Non-Asia Open Closed FRY 0.322 3.284 0.981 -1.118 -0.494 0.181 -0.099 (2.360) (2.963) (3.211) (-2.949) (-1.851) (0.670) (-0.113) FLY-1 -0.132 -0.054 -0.067 -0.014 -0.063 -0.105 -0.035 (-15.567) (-2.098) (-3.861) (-0.524) (-2.899) (-5.473) (-1.780) DDY 0.101 0.338 0.314 -0.055 -0.033 0.197 -0.020 (6.902) (4.913) (4.238) (-2.554) (-1.614) (4.062) (-0.743) REXLt-1 -0.249 -0.333 0.198 !.009 -0.247 -0.399 -0.251 (-S.773) (-3.109) (1.294) kJ.062) (-2.463) (-3.126) (-3.196) 1yt-1 0.572 0.481 0.602 0.703 0.776 0.754 0.757 (21.962) (7.389) (12.222) (10.248) (16.442) (18.774) (13.663) 1? 0.768 0.890 0.853 0.740 0.757 0.788 0.718 No. obs. 358 67 114 108 177 133 133 Note: Hats denote endogenous variables, 9 statistics are given in parentheses. Table 2: National Saving Estimates SNY without Growth Rate. Variable All S. Asia E. Asia L.A. Non-Asia Open Closed FDIY -0.585 0.668 0.238 -1.134 -0.225 -0.213 1.733 (-4.727) (0.444) (0.841) (-2.370) (-1.074) (-0.910) (1.950) FLYt-1 -0.001 -0.019 0.003 0.045 0.024 0.008 0.038 (-0.193) (-0.615) (0.152) (2.547) (1.997) (0.440) (2.383) RW 0.077 -0.005 -0.109 -0.270 0.028 -0.207 0.003 (1.671) (-0.059) (-1.027) (-2.507) (0.389) (-1.830) (0.041) SNY-1 0.719 0.684 0.839 0.799 0.757 0.828 0.848 (23.317) (7.506) (20.489) (14.878) (15.156) (21.758) (16.279) R 0.853 0.815 0.785 0.860 0.833 0.858 0.827 No. obs. 364 67 114 114 183 137 135 Note: Hats denote endogenous variables, I statistics are given in parentheses. 58 Table 3: Export Estimates XKY without Real Exchange Rate. Variable All S. Asia E. Asia L.A. Non-Asia Open Closed FDI_ -0.307 -0.518 0.578 -0.408 -0.348 -0.376 -0.155 (-2.724) (-0.407) (1.713) (-2.058) (-2.024) (-1.949) (-0.583) OKY -0.127 -0.080 -0.260 -0.153 -0.134 -0.157 -0.154 (-10.920) (-1.800) (-6.127) (-6.756) (-6.981) (-4.217) (-7.010) XEY-I 0.960 0.873 0.994 0.973 0.939 0.990 0.976 (67.894) (16.089) (52.860) (25.465) (28.415) (48.708) (26.470) R2 0.979 0.983 0.985 0.969 0.944 0.973 0.936 No. obs. 367 68 118 109 181 141 141 Note: Hats denote endogenous variables, t statistics are given in parentheses. .59 Table 4: Import Estimates LVKY without Real Exchange Rate. Variable All S. Asia E. Asia L.A. Non-Asia Open Closed fDlY 0.426 -1.552 0.808 0.763 0.236 0.085 1.370 (4.030) (-0.709) (2.751) (3.862) (1.570) (2.333) (4.455) MY 0.158 0.145 0.227 0.125 0.119 0.218 0.164 (12.588) (1.576) (4.685) (7.214) (7.051) (4.240) (7.248) TTL -0.022 0.305 0.026 -0.063 -0.020 0.135 -0.002 (-0.897)- (1.996) (0.296) (-2.210) (-0.649) (1.678) (-0.057) MY 0.129 0.248 0.088 0.139 0.118 0.016 0.160 (8.188) (1.641) (1.922) (6.679) (5.394) (0.348) (6.370) fy- 0.139 0.093 0.250 -0.047 0.046 0.209 0.021 (9.537) (0.657) (5.635) (-1.762) (2.114) (4.622) (0.722) 3fMli 0.730 0.787 0.674 0.418 0.653 0.753 0.649 (38398) (13.427) (14.648) (6.782) (14.697) (19.770) (13.836) 2 0.970 0.973 0.966 0.946 0.956 0.958 0.930 No. obs. 366 68 118 108 180 141 140 Note: Hass denote eadogenous variables. stsatistics are given is pareatbes. 60 Table 5: Real Exchange Rate Estimates REXL. Variable All S. Asia - E. Asia L.A. Non-Asia Open Closed FDJY -1.418 6.986 1.058 2.032 2.345 0.104 -5.496 (-4.212) (0.879) (1.231) (1.420) (2.988) (0.162) (-1.665) Y -0.321 -0.081 -0.502 0.020 -0.388 -0.389 -0.008 (-6.083) (-0.342) (-2.996) (0.105) (-3.281) (-4.175) (-0.047) IMAY 1.039 0.887 -0.164 0.943 1.060 0.211 1.526 (11.832) (3.438) (-0.787) (1.478) (5.050) (1.416) (5.294) TTL 0.427 0.154 0.965 -0.378 0.937 0.197 0.560 (3.152) (0.202) (2.759) (-0.943) (3.490) (0.763) (1.752) INY -0.330 -1.663 -0.190 1.195 -0.408 -0.076 1.104 (-4.781) (-3.074) (-0.950) (3.034) (-2.354) (-0.511) (3.903) -0.745 -1.387 0.203 -2.003 -1.759 -0.396 -2.012 (-11.935) (-3.239) (1.242) (-7.081) (-8.755) (-3.288) (-8.083) REXLt-1 0.670 0.618 0.864 0.139 0.547 0.887 0.306 (25.430) (7.713) (13.470) (2.067) (10.246) (19.847) (5.441) R2 0.850 0.926 0.798 0.697 0.787 0.922 0.805 No. obs. 367 68 118 109 181 141 140 Variable All S. Asia E. Asia L.A. Non-Asia Open Closed FDY -0.739 17.197 0.869 3.140 1.612. 0.187 -5.323 (-2.049) (2.365) (1.025) (2.296) (1.930) (0.290) (-1.489) -0.263 -0.003 -0.526 0.132 -0.250 -0.372 0.171 (-4.546) (-0.014) (-3.164) (0.693) (-2.128) (-4.050) (0.870) TTL 1.107 1.248 0.915 -0.541 0.851 0.216 0.893 (6.709) (1.681) (2.720) (-1.341) (2.934) (0.832) (2.685) IAY 0.257 -1.354 -0.254 1.571 0.083 0.020 1.796 (4.313) (-2.410) (-1.398) (6.288) (0.544) (0.155) (6.289) A -0.524 -1.017 0.1'31 -1.949 -1.376 -0.299 -1.707 (-7.265) (-2.518) (1.062) (-6.548) (-7.175) (-2.841) (-6.156) REXLt-I 0.752 0.796 0.844 0.168 0.688 0.927 0.382 (25.419) (12.030) (13.841) (2.402) (14.769) (25.811) (6.434) R2 0.839 0.906 0.802 0.689 0.791 0.919 0.781 No. obs. 367 68 118 109 181 141 140 Note: Hats denote endogenous variables, t statistics are given in parentheses. 61 Table 6: Real Exchange Rate Estimates REXL with Lagged Foreign Direct Investment. Variable All S. Asia E. Asia L.A. Non-Asia Open Closed FDIYL 0.047 -0.143 0.073 0.364 -0.369 0.019 0.468 (1.328) (-0.809) (0.870) (2.369) (-4.607) (0.360) (3.076) Y -0.158 0.003 -0.529 -0.004 0.037 -0.337 0.219 (-2.689) (0.012) (-3.217) (-0.024) (0.320) (-3.633) (1.156) TTL 1.385 1.118 1.053 -0.558 -0.058 0.269 0.860 (7.521) (1.409) (2.647) (-1.404) (-0.194) (0.934) (2.426) INY 0.155 -0.592 0.019 1.801 0.227 0.008 1.504 (2.572) (-1.467) (0.097) (7.398) (1.754) (0.071) (5.993) rXY -0.630 -1.234 0.020 -2.543 -1.144 -0.277 -2.343 (-7.691) (-2.120) (0.135) (-7.269) (-6.599) (-2.517) (-6.800) REXLt-i 0.692 0.713 0.812 0.148 0.536 0.929 0.366 (21.720) (8.169) (14.009) (2.224) (9.289) (22.886) (6.218) R2 0.829 0.902 0.809 0.676 0.782 0.920 0.773 No. obs. 362 64 117 109 181 140 136 Note.: Hat denote endogenous variables, t statistics are given in parentheses. .62 Table 7: Current Account Estimates CA Y. Variable. All S. Asia E. Asia L.A. Non-Asia Open Closed FDY -1.064 -4.239 -1-139 -0.445 -0.717 -1.108 -0.204 (-15.287) (-4.878) (-6.122) (-1.680) (-3.739) (-5.835) (-0.531) MY -0.476 -0.508 -0.501 -0.425 -0.446 -0.529 -0.387 (-46.271) (-12.234) (-13.956) (-17.551) (-25.094) (-20.600) (-13.388) FLY-1 0.009 0.058 -0.003 0.031 0.016 0.048 0.037 (1.672) (3.336) (-0.239) (2.289) (1.496) (3.155) (3.224) TTL 0.016 0.039 -0.121 0.066 -0.022 -0.017 0.144 (0.809) (0.405) (-1.480) (1.611) (-0.726) (-0.220) (2.596) DfcY -0.010 0.010 -0.168 -0.002 -0.006 -0.055 0.007 (-1.338) (0.178) (-3.087) (-0.155) (-0.550) (-1.572) (0.562) CAY-1 0.856 0.807 0.798 0.790 0.846 0.907 0.735 (62.570) (13.689) (22.361) (19.018) (30.038) (31.575) (18.195) R2 0.903 0.909 0.904 0.851 0.898 0.919 0.873 No. obs. 358 67 114 108 177 133 133 Note: Hats denote endogenous variables, S statistics are given in parentheses. Table 8: Current Account Estimates CAY with Lagged Foreign Direct Investment. Variable All S. Asia E. Asia L.A. Non-Asia Open Closed FDYL 0.029 -0.042 0.026 0.105 0.050 0.010 0.044 (1.997) (-0.969) (0.710) (2.874) (1.912) (0.320) (1.755) I -0.502 -0.457 -0.506 -0.442 -0.464 -0.538 -0.388 (-42.533) (-8.152) (-12.366) (-16.960) (-22.024) (-17.770) (-14.082) FY-1 -0.027 0.063 -0.010 -0.040 -0.030 0.075 0.014 (-2.361) (1.358) (-0.334) (-1.591) (-1.509) (2.843) (0.768) TTL -0.029 0.075 -0.024 0.108 -0.003 0.185 0.179 (-1.092) (0.673) (-0.239) (2.188) (-0.083) (2.221) (3.555) DDCY 0.015 -0.044 -0.208 0.017 0.015 -0.076 0.014 (1.978) (-0.654) (-3.389) (1.338) (1.301) (-1.966) (1.109) CAY 0.900 0.887 0.770 0.721 0.837 0.951 0.686 (56.321) (10.349) (14.846) (14.960) (25.036) (26.295) (16.706) 0.876 0.909 0.880 0.831 0.873 0.882 0.81 No. obs. 354 63 114 108 177 133 129 Note: Hats denote endogenous variables, i statistics are given in parentheses. ASIA REGION DISCUSSION PAPERS I/ (Continued) 1itl1 Auh .iolnator 1DP37 Commodity Taxation in Selected Countries In South East and East Asia 2. Shali June 1989 H. Flelsig (81413) IDP38 Tax Analysis In Developing Country Settings R. Musgrave June 1989 H. Fielsig (81413) IDP39 Indonesia: External Shooks, Policy Response and Adjustment Performance S. Ahmed June 1989 S. Ahmed (82467) IDP42 An Analysis of the Nature of Unemployment W.T. Dickens In Sri Lanka and K. Lang July 1989 R. Zagha (80433) IDP44 Asslating Poor Rural Areas Through Groundwater Irrigation F. Kahnert August 1989 C. Chamberlin (81409) IDP61 Educational Development in Asia: A Comparative Study Focussing on Cost J-P. Tan and and Financing Issues A. Mingat October 1989 J-P. Tan (81408) IDP62 Chinese Reforms, inflation and the Allocation of Investment in a Socialist Economy 0. Yenal October 1989 0. Yenal (81415) IDP63 Public Policy to Promote Industrializa- tion: The Experience of the East Asian NICB and Lessons for Thailand D. Dollar May 1990 D. Dollar (50518) tDP66 A Study of the Poor in Sri Lanka C. Rouse June 1990 Y. Huang (80434) IDP68 Health Sector Financing in Asia C. Griffin August 1990 J-P. Tan (81408) IDP74 A Case Study of a Gradual Approach to Economic Reform: The Viet Nam Experience of 198588 Z. Drabek September 1990 Z. Drabek (80504) IDP85 On Estimating Inadequacy of Energy Intakes: Revealed Food Consumption Behavior versus Nutritional Norms B.S. Minhas September 1990 S. Jayanthl (81419) IDP88 Asia Region Seminar on Policy Challenges In India October 1990 C. Chamberlin (81409) IDP93 Parametric Population Projection and Its Usefulness for Policy Analysis W. Sanderson April 1991 J-P. Ten (81408) IOP96 A Review of Korea's Trade Pattern D.M. Lelpziger and S.Y. Song March 1991 D. LeIpAger (81388) IDP97 Sustainability as Intergenerational Equity: The Challenge to Economic Thought and Practice R.B. Norgaard June 1991 F. Johansen (81410) IDP99 Housing Prices, Affordability, and Government Policy in Korea K-H. Kim July 1991 D. Lelp*iger (81388) IDP101 Lessons from Experience with Wage Flexibility in Asia L Nabi August 1991 I. Nabt (80514) j1 Papers sti avaiable upon request. Asia Sais. discontinued following December 1. 1991 restructuring of the Region. Note: Extra copies may be obtained from the Asia Information Service Center.