For other versions of this document, see http://wikileaks.org/wiki/CRS-97-881 ------------------------------------------------------------------------------ 97-881 F CRS Report for Congress Received through the CRS Web Multilateral Development Banks and Capital Flows to Developing Countries September 17, 1997 Jonathan E. Sanford Analyst in International Political Economy Congressional Research Service Congressional Research Service ~ The Library of Congress Multilateral Development Banks and Capital Flows to Developing Countries Summary Net capital flows to developing countries (new disbursements minus repayments or disinvestments) have increased substantially in recent years, rising from $101 billion in 1990 to over $237 billion in 1995. Private capital has been the fastest growing element. The annual net flow of foreign investment grew from $28 billion in 1990 to $128 billion in 1995, while the net flow of commercial lending went from $17 billion to $57 billion during this period. At the same time, the net flow of foreign assistance declined from $57 billion in 1990 to $53 billion in 1995. The net flow of lending from the multilateral development banks (MDBs) fell from $16 billion to $11 billion during these years. Most MDB loans are lent on near- market repayment terms. The higher interest rates and shorter repayment periods for these loans limit the net flow of resources that they provide. Many have suggested that private capital can now provide the funds that the developing world needs to finance its development. They say that MDB loans are not needed in the new international environment. A close look at the data suggests that this may not be the case, particularly for low-income countries. In 1995, middle-income developing countries received 65% of the total net flow of financial resources going to developing countries. They also received nearly 71% of all net private capital going to the developing world. Low-income countries received only 35% of the net capital flow and 29% of the net private capital in 1995. Most low-income countries also received a very small share of that total. In 1995, China received 63% of all net funds going to low-income countries and India received an additional 4% of that total. India's share would have been much higher, except that it suffered a massive outflow -- the inflow of new bilateral loans having fallen off -- from bilateral lending. They also got most of the private capital. China received 82% of the foreign investment and 91% of the commercial loans going to low-income countries in 1995. India received 6% of the net foreign investment and 12% of the commercial loans in 1995. All other low-income countries received only $5.7 billion in net foreign investment and $1.3 billion in commercial loans that year. By contrast, net MDB loans totaled more than $4.4 billion in 1995 and bilateral grant aid accounted for more than $16 billion. Most low-income countries have been unsuccessful in attracting foreign capital. Many of the activities of the MDBs and the bilateral aid agencies are focused on helping low-income countries improve their attractiveness to foreign investors and commercial lenders. Some of the activities funded by the MDBs might be of interest to foreign investors. However, most of the MDBs' loans have gone to finance activities -- education, health, social projects, national roads, reform of government institutions and economic policy reform -- that seem unlikely to be profitable to a foreign investor. Concessional MDB loans account for most of the total. Most poor countries have too little income from foreign trade and are too uncreditworthy to borrow much more than they do presently from commercial lenders. Most low-income countries will have few alternatives if the MDB concessional loan programs shrink in the future and their access to such MDB assistance declines. Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 The Flow of Resources to Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . 2 Flows to All Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 A Closer Look at the Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Middle-Income Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Flows of Private Capital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Growth and Savings Rates in These Countries . . . . . . . . . . . . . . . . . . . 4 Flows of Foreign Aid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Low-Income Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Total Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Private Capital Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 China and India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Flows of Private Funds, Excluding China and India . . . . . . . . . . . . . . . 5 Overall Flow of Funds, Excluding China and India . . . . . . . . . . . . . . . 6 Growth and Savings Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Debt Arrears . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Multilateral Banks and Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Substitute Commercial Loans for MDB Loans? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Replace Them With Foreign Investment? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Few Alternatives for Low-Income Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 List of Figures Figure 1. Net Flow of Resources to Developing Countries, 1975-95 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Figure 2. Net Flow of Resources to Middle-Income Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Figure 3. Net Flow of Resource to All Developing Countries, 1975-95 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Figure 4. Net Flow of Resources to Low-Income Countries, Excluding China and India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Figure 5. IBRD Lending by Sector, 1996 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Figure 6. IDA Lending by Sector, 1996 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Multilateral Development Banks and Capital Flows to Developing Countries Introduction The World Bank estimates that the total net flow of resources to developing countries in 1996 totaled almost $285 billion, a $48 billion increase over the total for 1995. It also estimates that the net flow of foreign investment (both direct and portfolio) increased by 20% between 1995 and 1996 while the net flow of official lending (bilateral and multilateral) fell by over 50%. Lending from multilateral development banks comprised less than 5% of the total net flow in 1995. Some suggest that the MDBs need to change the focus of their operations, in order to magnify and not supplant the effect of private capital flows. Many also say that the large flow of private funds shows that these can now be relied on finance development activities and there is less need for multilateral lending. Others argue that there is still a need for traditional MDB lending and that the MDBs have already changed the focus of many of their operations. This memorandum discusses the current pattern of financial flows to developing countries. It finds that the MDBs now play an increasingly modest and specialized role in most middle-income developing countries. However, most low-income countries have difficulty attracting private capital and the MDBs still play an important role in their economies. Other factors, such as appropriate economic policies and political stability, also play a key role facilitating and promoting economic development. Many analysts argue, however, that foreign loans and grants and other kinds of foreign capital are still very important. If the flow of foreign investment remains limited and market-rate loans continue to be too expensive, most poor countries will still need continued access to MDB concessional loans. Steps might be taken to possibly increase the flow of private funds to low-income countries. Nevertheless, the poor countries would probably be unable able to substitute foreign investment and market-rate loans for their current access to MDB concessional loans if the latter became less available. This paper does not address the question whether current U.S. contribution levels are appropriate, too low, or too high or whether membership in the MDBs serves the national interests of major stockholders such as the United States. Likewise, it does not address the question whether the MDBs' loans would be more effective if the banks' organizational bureaucracy and their operational procedures were reformed. The focus of this paper is solely on the question whether it is likely that private capital flows might substitute for MDB lending if the multilateral banks' CRS-2 funding levels were reduced in size. This paper concentrates on the period through 1995, as the figures for 1996 are preliminary estimates.1 The Flow of Resources to Developing Countries Flows to All Developing Countries The flow of foreign resources to developing countries has grown substantially in recent years. According to World Bank statistics, the total net flow increased from $101 billion in 1990 to over $237 billion in 1995.2 Foreign investment is the fastest growing part of this flow. Figure 1 shows that the net annual inflow of foreign direct investment grew from $25 billion in 1990 to $96 billion in 1995. Figure 1. Net Flow of Resources to Developing Countries, 1975-95 Portfolio investment likewise increased from $3 billion to over $32 billion.3 The net flow of commercial loans grew from $17 billion in 1990 to nearly $57 billion in 1995. 1 The data shown here were derived from the 1997 issue of the World Bank's annual publication now called Global Development Finance. Through 1996, it was called the World Debt Tables. The electronic version of this document -- the Socioeconomic Timeseries Access and Retrieval System (STARS) -- was used in the preparation of this study. Upon request, a copy of the tables constructed from the STARS data can be supplied, showing the numbers that were used to generate the four area graphs and two pie charts displayed in this report. 2 Net flow is the total inflow of funds investments, grants, and loans minus the total outflow of funds from these same sources. The figures include movements of principal only. Interest payments on commercial loans and the repatriation of profits from foreign investments are not included in the totals. 3 Foreign direct investments are those where the foreign investor owns at least a majority of the stock and has operational control. Portfolio investments are those where the foreign investor owns shares of a foreign firm without having control. CRS-3 During this same period, the net flow of official assistance to developing countries declined. The flow of grant aid increased slightly, from $29 billion in 1990 to nearly $33 billion in 1995. The net flow of bilateral lending fell, however, from nearly $12 billion to $9 billion. The net flow of multilateral (mainly MDB) lending also declined from almost $16 billion in 1990 to barely $11 billion in 1995. A Closer Look at the Data The gross figures may obscure, however, more than they reveal. A very different pattern emerges if a distinction is made between low-income and middle- income developing countries. The situations facing these two groups of countries are very different. The 63 low-income countries included in this analysis have an average per capita gross national product (GNP) of less than $750 a year. The 96 middle-income countries in this study have an average per capita GNP of $751 to $8,260 a year. By comparison, the average per capita GNP of the top two countries in 1995, the United States and Switzerland, were $26,980 and $25,860, respectively. Middle-Income Countries Flows of Private Capital. Most of the money and most of the private capital flows to developing countries have gone to middle-income nations. Included in the middle-income category are such countries such as Argentina, Botswana, Costa Rica, Egypt, Hungary, Poland, Russia, and Thailand. In 1990, middle-income developing countries received about 61% of the net resources flowing into developing countries. Five years later, in 1995, they received over 65% of the funds. As Figure 2 indicates, foreign investment and commercial loans comprise a large and increasing share of all the foreign funds these countries have received. In 1990, funds from private sources accounted for 52% of the net inflow of funds. In 1995, they accounted for more than 84% of the total. MDB loans accounted for less than 5% of the total net inflow to middle income countries during these five years. Private borrowers Figure 2. Net Flow of Resources to Middle-Income Developing Countries received over 48% of all long-term loans to middle-income countries. As a whole, middle-income developing countries have been the largest recipients of private funds. In 1995, nearly 71% of the net foreign investment and commercial loans going to the developing world went to middle-income countries. CRS-4 This is a decline from the 90% level seen in 1990. In recent years, China, India, and a few other countries have experienced a substantial major increase in their receipts from foreign investment and commercial loans. This is discussed below. Growth and Savings Rates in These Countries. This inflow of private funds is not necessarily related to the middle-income countries' current economic situation. As a group, between 1990 and 1995, middle-income countries had an annual gross domestic savings rate of about 25%. (By comparison, the rate for the United States was 15%.) In East Asia and the Pacific, the rate was about 38%, while in Latin America it was half that level. On the other hand, the middle-income countries grew as a whole at an average annual rate of only 0.1% between 1990 and 1995. East Asia grew by 10.3% annually, however, while Latin America increased by 3.2% and the Middle East expanded by 2.3%. Eastern Europe and the former Soviet Union, a region that has received a considerable amount of foreign investment in recent years, saw its economy shrink by 6.5% annually during these years. Flows of Foreign Aid. Before 1989, most middle-income countries received little or no grant foreign aid. Since the collapse of communism in Eastern Europe and the former Soviet Union, however, a significant flow of grant aid to those countries has occurred. However, middle-countries received few bilateral aid loans. In dollar terms, middle-income developing countries receive much more in the way of new loans from the MDBs than do low-income countries. However, most of these loans have come from the banks' near market-rate loan windows, such as the International Bank for Reconstruction and Development (IBRD). Their higher interest charges and shorter repayment periods limit the net transfer of resources to the borrowers. Between 1991 and 1995, such loans accounted for about 76% of all MDB lending to middle-income countries. The concessional loans went mainly to countries in special circumstances or to lower-middle income countries whose per capita incomes are only a little higher than those of the low-income group. Low-Income Countries Total Flows. Much less money has gone to low-income countries, even though their combined population (3.2 billion) is twice as large as that of the middle- income group. During the period 1991 to 1995, the poor countries received about 37% of all net resources flowing to all developing countries. In 1990, they got 39% of the total, while their share fell to 35% in 1995. Private Capital Flows. They also received 30% of the net private resources (foreign investment and commercial loans) going to developing countries during this five year period. In 1990, they got 10% of the total, while in 1995 their share was about 29%. As a group, low-income countries now receive a significant share of their net funds from abroad in the form of private capital. As Figure 3 indicates (next page), foreign investment and commercial loans accounted for 65% of the aggregate net inflow in 1995, while foreign aid grants represented 21% and MDB loans about 9% of the total. Bilateral aid loans accounted for only about 5% of the total. Governments have been the principal borrowers. Between 1991 and 1995, only about 6% of the net disbursements for long-term loans went to private borrowers. CRS-5 Figure 3. Net Flow of Resource to All Developing Countries, 1975-95 China and India. These overall figures on financial flows to low-income countries are significantly skewed by the preponderance of two countries, China and India.4 In 1994, China received over 56% of all net resources flowing to low-income countries. In 1995, it received 63% of the total. India received about 9% of the total flow in 1994 and 4% in 1995. India's share for 1995 would have considerably higher this last year, except that its net receipts from MDB lending fell and the net flow of bilateral lending was sharply negative ($1.25 billion, a figure equivalent to 38% of its net receipt of foreign funds). The two countries have also been relatively successful in attracting private capital. In 1994 and 1995, foreign investment accounted for 79% and 74%, respectively, of the net funds China received from abroad. Commercial loans comprised another 14% and 11%, respectively, of the total. Meanwhile, for India in 1994 and 1995, foreign investment accounted for 67% and 85%, respectively, of its net capital receipts. Commercial loans comprised another 13% and 24% of the total. The total can exceed 100% because, as noted, the net flow in 1995 for bilateral loans was large and negative. Flows of Private Funds, Excluding China and India. A very different pattern emerges if China and India are subtracted from the total. For one thing, as Figure 4 shows (below), the total net flow of resources is smaller, about $33 million in 1994 and $28 million in 1995. The population of these 61 low-income countries 4 Though India and China are both low-income countries that have had some success attracting inflows of foreign investment, there are significant differences in their relative levels of development. China's per capita income level is approximately twice that of India, for example, and industry comprises a much larger share of its economic output. Compared to its annual import bill, China's foreign exchange reserves are more than double those of India. India's infant mortality rate is twice that of China, yet India has twice as many doctors and twice as many hospital beds per capita. See: Comparing Countries' Levels of Development, CRS Report 97-759 F, August 7, 1997. CRS-6 (most of them located in sub-Saharan Africa or South Asia) is about equal to that of the middle-income group. For another thing, these countries had proportionally much less access to private sector funds. In 1995, they received a net $5.7 billion in foreign direct and portfolio investment, about 4% of the net foreign investment inflow to developing countries. They also received a net $1.3 billion in commercial loans, about 2% of the overall total. The inflow of major amounts of portfolio investment and commercial lending began in 1994. Figure 4. Net Flow of Resources to Low-Income Countries, Excluding China and India Overall Flow of Funds, Excluding China and India. Foreign aid grants are the most important source of foreign funds for this group of low-income countries. In 1995, grants accounted for more than 57% of the total net flow of resources they received. Net MDB loans comprised another 16%, while bilateral aid loans were about 2% of the total. Foreign direct investment and portfolio investment accounted for about 20% of the net financial receipts of these countries, while commercial loans comprised a little less than 5% of the total. Between 1991 and 1995, over 93% of the net disbursements of new MDB aid to this group of low-income countries was provided on concessional terms. The World Bank's International Development Association (IDA) accounted for most of the concessional loans. Most donor countries now provide their bilateral foreign aid to low-income countries in the form of grants rather than loans. Growth and Savings Rates. If China and India are included in the total, the growth rate for the low-income countries has been impressive, averaging 6.8% a year between 1990 and 1995. If they are removed from the group, however, the growth rate for the low-income countries was about 1.8% annually for this period. Growth rates in South Asia (including India) averaged about 4.6% annually, while those for sub-Saharan Africa were 1.4%. The gross domestic savings rate for the low-income countries without China and India was about 10% annually. CRS-7 Debt Arrears. Many low-income countries have had difficulty keeping up their foreign debt payments. Between 1991 and 1995, the total arrears for principal and interest on their long-term debt increased by $46.6 billion, more than doubling the total. Three-quarters of their arrears are owed to official creditors, mainly bilateral lenders. Concern about repayment or blocked repatriation of profits is almost certainly a reason why the private sector has been reluctant to put much money into these low-income countries. China and India have had no arrears for many years. Between 1991 and 1995, middle-income countries reduced their arrears by $5.4 billion. More than half (52%) of their arrears are owed to private lenders. Arrears due to the multilateral banks from all borrowers totaled $2.6 billion in mid- 1996. Most of this was attributable to a few countries. Low-income countries accounted for about $833 million. Yugoslavia accounted for nearly half the total arrears due to MDBs. Multilateral Banks and Developing Countries There have been discussions, in Congress, in other countries, and elsewhere about possible reductions in the size of the donor countries' contributions to MDB concessional loan programs, such as the IDA and the Asian Development Fund. This is feasible if the goal is principally a reduction in government expenditures. It may not be effective, however, if the goal is one of encouraging the MDBs to adopt reforms that they have avoided previously.5 Similarly, it seems unlikely that cuts in donor country payments to the multilateral banks will lead to more growth and larger flows of private capital to developing countries. The MDB concessional loan programs must receive annual contributions from their donor country governments in order to continue operations. The multilateral banks' near market-rate loan programs are mainly funded with money the banks borrow in world capital markets. In most cases, their capital base is sufficient for them to continue their current levels of lending indefinitely. The question arises whether developing countries have realistic alternatives if the size of the MDB loan programs diminishes in future years. Two alternative sources of capital have been proposed -- more commercial lending and expanded flows of foreign private investment. 5 Many observers believe the MDBs have not moved fast enough to institute reforms in their organizational structure and their operational procedures. Some argue that cuts in U.S. contribution levels will catch the banks' attention and encourage them to be more attentive to the need for reform. About 75% of the MDBs' annual lending is provided by their near market-rate loan facilities, using funds the MDBs borrow themselves, while about 75% of their member countries' annual contributions to fund the banks' concessional loan programs. Some analysts believe that cutbacks in the donors contributions to the MDB concessional loan facilities may not be painful enough to induce the banks to adopt reforms. In addition, funds for aid to poor countries would be cut but no reduction would likely occur in the amounts of money available for loans to middle-income countries. CRS-8 Substitute Commercial Loans for MDB Loans? Foreign aid accounts for about 16% of the net financial resources that middle- income countries receive from abroad. MDB loans account for between a quarter and a third of this aid. As a whole, middle-income countries would not be injured if the MDB concessional loan programs were terminated or reduced in size. Most of them borrow little or nothing from those facilities. Many of the activities currently funded with MDB near market-rate loans could be financed with commercial loans, though the higher cost of these loans may limit the amount countries can afford to borrow. MDB near market-rate loans typically cost one-half percent or so more than the average rate the MDBs pay to borrow funds in world commercial capital markets. Low-income countries, on the other hand , would probably be unable to replace their current flow of MDB concessional loans with higher cost loans. In 1995, foreign aid to the group excluding China and India accounted for about 75% of the net financial assets they received from abroad. Almost all of it was on concessional terms. MDB loans accounted for about a fifth of that total. Most of these countries are too poor and their creditworthiness is too weak for them to borrow much from the MDB near market-rate loan facilities or from commercial lenders. Replace Them With Foreign Investment? Figures 5 and 6 show the sectoral distribution of IBRD and IDA lending in 1996. Altogether, loans in the areas of finance and industry and the construction of economic infrastructure accounted for about 47.7% of IBRD lending and 18.3% of IDA lending in 1996. Some of these projects might be of interest to foreign investors, particularly some of the projects designed to create new electric power capacity. A closer look suggests, however, that many will not be attractive. Most MDB finance projects seek to help countries develop new budget or pension procedures or to reorganize their domestic financial systems. Improvements in the relevant CRS-9 agencies and the adoption of more effective policies and procedures were generally the main focus of the loans. Institution building and policy reform were likewise the main focus of MDB loans in the areas of industry and mining. Some MDB transportation loans may fund bridges or highways where the revenue yield might be sufficient to interest a foreign investor. Most, however, are designed to rehabilitate or build national road networks or to reorganize and strengthen the agencies responsible for the maintenance and management of national transport systems. Many of the other activities financed by the MDBs are not likely to be of much interest to foreign investors. Projects in the social sectors, for example basic education and basic health, may offer substantial long-term benefits for society, but it is difficult to see how they could generate much revenue for a private investor. Likewise, projects in the area of public sector management -- administrative and civil service reform, reorganization of government agencies, and the adoption of new policies and procedures -- may improve the efficiency and effectiveness of government. This may help encourage more privatization and it may help improve the overall productivity of the economy. It is difficult to imagine how a foreign investor could generate a profit, however, through investment in public sector reform. The same point could be made regarding MDB loans for urban development, for the construction of water and sewerage systems in impoverished areas, for structural adjustment and fundamental economic policy reform, or for agricultural projects targeted to help low-income rural residents. Few Alternatives for Low-Income Countries Worldwide, the total level of bilateral development aid has been declining. Nevertheless, if donor countries were willing to expand their bilateral aid programs, they might pick up the cost of some of the programs the MDBs have been funding. In that case, however, the budgetary savings from reductions in the size of the MDB program would be limited. Most bilateral foreign aid agencies do not have the capacity or experience to undertake these programs, and some will not be willing to invest the time and money needed for their acquisition. Many low-income countries will be unable to borrow more from commercial lenders and they may have trouble finding organizations with the requisite skills to help them if they want to continue funding these programs on their own. Most likely, particularly for low-income countries, many of the activities now financed by the MDBs would not be sustained if the multilateral banks reduce the future size of their operations. Low-income countries might be able to attract more foreign investment if they adopt economic policy reforms and they create an environment more attractive to foreign investors. The public mood in many countries may not be such that a policy of open acquiescence to the concerns of foreign investors can be adopted. Most low- income countries have small domestic markets and are not likely to attract many investors interested in procuring local sales. Similarly, in many countries the roads, railways, airport, and port facilities may not be sufficient for investors oriented towards production for export. The local telecommunications system may be inadequate for the needs of firms that require close continuous contact with other countries. The local workforce may not be sufficiently productive due to lack of CRS-10 education and poor health. Many countries also lack the legal and policy framework necessary for the operation of a modern market system. They may not have adequate rules and institutions to both protect and regulate private commerce. In many cases, the multilateral banks' current loan programs in poor countries aim at remedying exactly these situations. The issues are often complex and time- consuming. The United States and other donors might want assurances that the resolution of these problems is among the banks' top priorities for low-income countries. However, a reduction in the net flow of MDB assistance to these countries is not likely to encourage the countries or the MDBs to accelerate their activity on this front. Often, a political consensus must be achieved to overcome domestic opposition in the recipient country to desired changes. Without affordable assistance from the MDBs or comparable assistance from other foreign aid agencies, the governments in many low-income countries may have difficulty building domestic support for economic reform and market-oriented economic policies. In the absence of such reforms, it seems unlikely that major inflows of foreign investment or commercial credit would be available to them. ------------------------------------------------------------------------------ For other versions of this document, see http://wikileaks.org/wiki/CRS-97-881