III Market Financing of Developing Countries

International Monetary Fund
Published Date:
July 1982
  • ShareShare
Show Summary Details

Loans to developing countries account for 37 per cent of the international assets of banks and their deposits for 29 per cent of international bank liabilities. Although banks’ business with developing countries is thus considerably smaller in quantitative terms than with industrial countries, concerns about the functioning of the international capital markets tend to focus on the role of the markets in relation to developing countries. This reflects the fact that the markets play a relatively larger role for the developing countries than they do for industrial countries. The nature of their role also differs between the two types of countries.

For industrial countries, much of the flow of finance through international capital markets is a result of the growing multinationalization of their firms and integration of their financial markets. When these countries wish to attract capital, they generally can do so through appropriate adjustments in financial policies, without public sector entities directly approaching the markets. For most developing countries, on the other hand, capital inflows through the markets, or placements of financial assets in the markets, result from direct approaches to market participants by public sector entities in the countries concerned. Such transactions may be oriented toward financing imports or general balance of payments needs, on the one hand, or toward placements of international reserves, on the other. The distinction between industrial and developing countries should not, however, be exaggerated. For the more advanced developing countries a significant portion of capital market inflows is the result of transactions by the private sector, and the private sector also has significant assets in the international capital markets. Many industrial countries, moreover, supplement macroeconomic policies aimed at attracting capital with direct borrowing from the markets.

Quantitatively, the scale of the use of the markets by the non-oil developing countries is indicated by the fact that relative to exports in 1981, the stock of bank claims on them at the middle of 1981 amounted to 83 per cent. (For industrial countries the figure was 39 per cent.) Over the eight years ended in 1981, net capital market flows to the non-oil developing countries were equivalent to 53 per cent of these countries’ ex post requirements for balance of payments financing.

On the liabilities side, at the end of 1981 deposits of the oil exporting developing countries were equivalent to a full year of their imports, and over the eight years ended in 1981 one third of their cumulative cash surplus was deposited with banks.

Recent Developments

One of the most striking aspects of the development of the international capital markets in recent years has been the extent to which their role as international financial intermediaries with respect to developing countries has increased. Besides acting as a repository for the foreign exchange reserves of such countries and as an outlet for the savings of those among them with current account surpluses, the markets have provided much of the large amount of balance of payments financing which developing countries have obtained over the last few years. Between 1975 and 1981, bank liabilities to oil exporting developing countries more than tripled, reaching almost $155 billion. Over this same period, bank exposure to non-oil developing countries increased at an annual rate of about 25 per cent to $275 billion by 1981.

Between 1978 and 1980 both the aggregate current account deficit of the non-oil developing countries and net bank lending to those countries approximately doubled. In 1981, however, net bank lending to the non-oil developing countries leveled off in absolute terms, and the rate of growth of bank claims on these countries slowed to 23 per cent, compared with an average annual increase of 27 per cent in 1979–80. As their aggregate deficit again rose significantly, their net borrowing from the international capital markets was equivalent to 52 per cent of their current account deficit plus accumulation of reserves, down from close to 60 per cent in 1979 and 1980 (Chart 6). Of the World Economic Outlook’s major analytical subgroups of the non-oil developing countries, only “net oil exporters” substantially increased their borrowing in 1981; for “major exporters of manufactures” the flow increased only slightly, while for “low-income countries” and “other net oil importers” the flow declined. To some extent this pattern reflected the fact that banks were unwilling to increase their exposure to a number of countries whose creditworthiness was in doubt, but perhaps a more common factor was a perception by borrowers themselves that market conditions were not propitious for a major increase in borrowing. The statistics on bank lending to the developing countries as a whole, and to various subgroups, are dominated by the transactions of a few large borrowers. Five countries accounted for roughly two thirds of net borrowing by non-oil developing countries in 1981; fully 60 per cent of net lending went to the Latin American region.

Chart 6.Non-Oil Developing Countries: Financing Through International Capital Markets, 1973–81

1 Net of repayments; does not exclude double counting due to banks’ holdings of bonds.

New medium-term bank loan commitments to non-oil developing countries declined in 1980, as major borrowers drew down large credits which had been arranged but not drawn in 1979. By mid-1981, however, undrawn lines had fallen to a low level relative to outstanding debt to banks, and the rate of new commitments rose, reaching $43 billion for 1981 as a whole, somewhat above the record 1979 level. Net oil exporters accounted for much of the rise, Mexico alone for one third, but a broad spectrum of other countries also recorded substantial increases. India was, for the first time, a major factor in the market.

Weighted average spreads on new commitments to non-oil developing countries increased steadily from about ¾ per cent in the last half of 1979 to almost 1 per cent in the first quarter of 1981, but eased somewhat in the final half of 1981. Average maturities fell from 9½ years early in 1979 to 7½ years in the last half of 1980, and remained at about that level in 1981. Given the concentration of bank lending noted above, however, these movements in average spreads and maturities primarily reflect the terms secured by a few major borrowing countries.

The Role of Banks: Distinctions by Size of Debt to Banks

Analysis of the role of the international capital markets relative to developing countries is usually dominated by the experience of a few large debtors. The five largest debtors, for example, accounted for almost half of bank claims on developing countries at the middle of 1981. Analysis in terms of the World Economic Outlook’s major analytical subgroups of developing countries suffers from the same difficulty. The single largest debtor among the oil exporting developing countries accounts for 40 per cent of claims on such countries. For the “net oil exporter” and “major exporter of manufactures” groups among the non-oil developing countries, the single largest debtors account for, respectively, almost 75 per cent and 40 per cent of claims on the groups. While the other subgroups are not so dominated by single debtors, for both the “low-income countries” and “other net oil importers,” the four largest debtors account for half the total claims.

The present section compares the role of banks in countries where the debt to banks is relatively large with their role in countries where the debt to banks is relatively small. Table 3 sets out the 74 developing country members of the Fund whose debt to banks exceeded $100 million at the end of June 1981. For purposes of the discussion below, the 74 countries are divided into two groups of 37 countries each called “larger” and “smaller” debtors, respectively. The dividing line between the two groups falls near $1 billion in debt to banks. Some 45 developing countries with debts of less than $100 million are excluded on grounds of data unavailability, as are 13 offshore banking centers for which large volumes of interbank transactions distort the picture.

Table 3.Developing Countries Ranked by Debt to Banks, June 19811(In millions of U.S. dollars)
Larger debtors
1. Mexico46,604
2. Brazil46,382
3. Venezuela24,472
4. Argentina23,025
5. Korea17,945
6. Yugoslavia10,530
7. Philippines9,852
8. Chile8,754
9. Greece8,637
10. Algeria8,331
11. South Africa8,284
12. Portugal6,732
13. Indonesia6,348
14. Romania5,428
15. Israel5,012
16. Nigeria4,817
17. Colombia4,684
18. Thailand4,575
19. Saudi Arabia4,439
20. United Arab Emirates4,325
21. Turkey4,038
22. Ecuador4,027
23. Egypt3,984
24. Peru3,931
25. Kuwait3,767
26. Morocco3,269
27. Malaysia3,086
28. Ivory Coast2,774
29. Iran2,720
30. China2,545
31. Costa Rica1,332
32. Zaire1,159
33. India1,142
34. Pakistan1,068
35. Bolivia1,064
36. Tunisia1,064
37. Sudan949
Smaller debtors
38. Cameroon923
39. Dominican Republic908
40. Kenya855
41. Libya833
42. Gabon796
43. Uruguay745
44. Nicaragua723
45. Jamaica594
46. Qatar554
47. Honduras543
48. Zambia524
49. Trinidad and Tobago514
50. Guatemala499
51. Paraguay493
52. Jordan459
53. Cyprus456
54. Viet Nam424
55. Niger404
56. Congo397
57. Senegal340
58. El Salvador335
59. Sri Lanka335
60. Syrian Arab Republic325
61. Madagascar322
62. Oman318
63. Papua New Guinea301
64. Togo299
65. Guyana292
66. Iraq270
67. Zimbabwe269
68. Tanzania250
69. Malawi247
70. Ghana201
71. Yemen Arab Republic158
72. Mauritius150
73. Burma141
74. Guinea131
Source: Bank for International Settlements, The Maturity Distribution of International Bank Lending.

Besides the Fund member countries listed in the table, some 45 developing countries are recorded by the Bank for International Settlements as having debt to banks in amounts less than $100 million. Another 13 countries are offshore banking centers and there are also a few significant borrowers that are not Fund members.

Source: Bank for International Settlements, The Maturity Distribution of International Bank Lending.

Besides the Fund member countries listed in the table, some 45 developing countries are recorded by the Bank for International Settlements as having debt to banks in amounts less than $100 million. Another 13 countries are offshore banking centers and there are also a few significant borrowers that are not Fund members.

The comparison is based on a number of indicators that provide a perspective on the relative role of banks in the two categories of debtors. Certain ratios and other statistics have been calculated for each of the countries in Table 3. Tables 47 present median values (which avoid the distorting effects of extreme cases) for the two groups and for all 74 countries.6 The corresponding figures aggregated for developing countries are presented on the right-hand side of each table. As expected, the larger debtors are those for which debt to banks is large relative to other aspects of their economies, in part reflecting the fact that many of them are perceived to be more creditworthy than most of the smaller debtors.

Table 4.Developing Countries: Scale of Bank Financing, 1978–81(Ratios, in per cent)
Median Values1Aggregate for Developing Countries
Larger debtorsSmaller debtorsAll
Bank debt/exports2
Bank debt/GDP2
Growth of bank debt320202023
Syndicated credits
Countries obtaining4958691
Relative importance51281113
Bond issues
Countries issuing676541
Relative importance7789
Memorandum items:
Number of countries373774
Of which: non-oil293362
Share of group in total developing country bank debt, mid-198190595100
Sources: Bank for International Settlements; Organization for Economic Cooperation and Development; and Fund staff estimates.

For countries listed in Table 3; larger debtors are those numbered 1–37, while smaller debtors are those numbered 38–74.

Claims at midyear relative to exports or GDP for the year; non-oil developing countries only.

End of 1978 to mid-1981, at annual rate.

Countries for which at least one credit recorded 1978–81 (per cent of countries in group).

Average annual volume of syndicated credits 1978–81, relative to stock of bank debt at mid-1981.

Countries for which at least one issue recorded 1978–81 (per cent of countries in group).

Volume of bond issues relative to volume of syndicated credits 1978–81. Median values for countries issuing bonds.

Sources: Bank for International Settlements; Organization for Economic Cooperation and Development; and Fund staff estimates.

For countries listed in Table 3; larger debtors are those numbered 1–37, while smaller debtors are those numbered 38–74.

Claims at midyear relative to exports or GDP for the year; non-oil developing countries only.

End of 1978 to mid-1981, at annual rate.

Countries for which at least one credit recorded 1978–81 (per cent of countries in group).

Average annual volume of syndicated credits 1978–81, relative to stock of bank debt at mid-1981.

Countries for which at least one issue recorded 1978–81 (per cent of countries in group).

Volume of bond issues relative to volume of syndicated credits 1978–81. Median values for countries issuing bonds.

While it would be expected that almost all countries would be able to receive some bank credit, a noteworthy feature of the data is the very large number of countries which have, or have had, significant access to bank credit; in the middle of 1981, 74 developing countries had debt to banks in excess of $100 million. There are, of course, major differences among them in the scale of bank financing. For the non-oil developing countries in Table 3, the median value for bank debt relative to exports was 78 per cent in 1981, up from 62 per cent in 1979. The median value for the larger debtors was 117 per cent, almost double the 60 per cent recorded for the smaller debtors. The fact that the aggregate bank debt relative to exports for the non-oil developing countries increased only from 72 per cent to 83 per cent reflects the fact that some of the largest developing country debtors experienced a substantial decline in the ratio of their bank debt to exports. There was less variation relative to GDP, but the median value in 1981 for the larger debtors was 17 per cent while that for the smaller debtors was 12 per cent, in both cases little changed from 1979. The two groups had similar median growth rates for bank debt in recent years, although the range of variation among the smaller debtors was much larger.

Two particular modes of access to medium-term finance from the international capital markets are syndicated credit commitments and bond issues. All but two of the countries among the larger debtors have made use of syndicated credits, and all but five of those among the smaller debtors. Of the smaller debtors, however, less than half are regular participants: commitments to the others tend to take the form of infrequent, relatively small, project financings, not directly related to general balance of payments need.

Countries that have issued bonds are much less common than those that have been involved in syndications. Generally speaking only those countries that have made the largest use of bank credit have tapped the bond markets. Over three fourths of the larger debtors issued bonds between 1978 and 1981, while only two of the smaller debtors did so. Aggregate bond issues of developing countries were equivalent to only 9 per cent of syndicated credits from 1978 to 1981.

Table 5 sets out two aspects of the role of banks relative to other sources of international finance. In 1980 the median increase in debt to banks of the larger debtors was equivalent to 51 per cent of the amount required to finance their current account deficits plus their accumulation of bank deposits, while for the smaller debtors the proportion was only 15 per cent; in both cases the figure was down from the previous year.

Table 5.Non-Oil Developing Countries: Relative Role of Banks, 1978–80(Ratios, in per cent)
Median Values1Aggregate for All Non-Oil Developing Countries
Larger debtorsSmaller debtorsAll
Role in balance of payments financing2
Role in external indebtedness3
End of 197856324759
End of 198068404768
Sources: Bank for International Settlements; and Fund staff estimates.

For countries listed in Table 3; larger debtors are those numbered 1–37, while smaller debtors are those numbered 38–74.

Ratio of increase in bank assets to sum of current account deficit and increase in deposits with banks.

All debt to banks (including short-term) relative to total medium- and long-term external debt.

Sources: Bank for International Settlements; and Fund staff estimates.

For countries listed in Table 3; larger debtors are those numbered 1–37, while smaller debtors are those numbered 38–74.

Ratio of increase in bank assets to sum of current account deficit and increase in deposits with banks.

All debt to banks (including short-term) relative to total medium- and long-term external debt.

Another indicator of the relative role of banks is the size of debt to banks relative to total debt. While comprehensive data on total debt are not available, for the non-oil developing countries included in Table 3 estimates are available for external debt with an original maturity greater than one year. For those countries, the median ratio of debt to banks (of all maturities) to total medium- and long-term debt was 47 per cent at the end of 1980. For the larger debtors the median was 68 per cent, for the smaller debtors 40 per cent, in both cases up from the end of 1978.7

Certain aspects of the structure of bank assets and liabilities are set out in Table 6. At the end of 1978, many countries had negotiated credit commitments that had not yet been drawn down, so that undisbursed commitments, in aggregate, exceeded 30 per cent of their outstanding debt to banks. By the middle of 1981, this proportion had declined sharply, particularly for the non-oil developing countries. The decline was broadly distributed, with the median value for both larger and smaller debtors dropping sharply. Despite the fact that lending spreads were, in general, continuing to decline over the intervening period, it appears that most countries were not prepared to maintain their “pipeline” of undisbursed commitments, perhaps reflecting concern that any increase in borrowing from banks, beyond the large amounts they were already obtaining, might be available only on more onerous terms. The fact that the smaller debtors generally have higher ratios than the larger ones presumably reflects the fact that a larger share of their borrowing is in the form of trade-related or project-related credits rather than in quick disbursing financial credits.

Table 6.Developing Countries: Structure of Bank Assets and Liabilities, 1978–81(Ratios, in per cent)
Median Values1Aggregate for All Developing Countries
Larger debtorsSmaller debtorsAllOil exportingNon-oil
1. Undisbursed commitments/assets
End of 19783242363330
2. Short-term/total assets
End of 19784550494844
3. Liabilities/assets
End of 197848615215061
Source: Fund staff calculations based on BIS data.

For countries listed in Table 3; larger debtors are those numbered 1–37, while smaller debtors are those numbered 38–74.

Source: Fund staff calculations based on BIS data.

For countries listed in Table 3; larger debtors are those numbered 1–37, while smaller debtors are those numbered 38–74.

A second indicator set out in Table 6 is the proportion of bank claims accounted for by claims of short residual maturity. It might be expected that the larger debtors, with their large-scale use of medium-term financial credits, would have relatively small proportions of short-term debt, but in fact there appears to be little systematic difference between large and small debtors. Presumably, this pattern is associated with the fact that the countries whose debts are relatively small still have substantial medium-term financing associated with development projects; a large portion of this is guaranteed by export credit insurance agencies in the countries financing the projects.

In recent years, the aggregate proportion of short-term claims has increased for both oil exporting and non-oil developing countries, but this largely reflects the experience of Mexico, Venezuela, and a few other large borrowers which opted to increase their use of short-term financial credits; for most countries, the ratio declined.

A final indicator of the structure of developing countries’ relationships with banks is the ratio of bank liabilities to bank claims, i.e., the extent to which countries’ deposits with banks cover their debts. For most countries, such deposits are largely official foreign exchange reserves, but for some there are also substantial private holdings. Reflecting the fact that aggregate reserve holdings have not kept pace with the growth of debt to banks, the aggregate ratio for non-oil developing countries declined from 61 per cent at the end of 1978 to 43 per cent by the middle of 1981. The decline largely reflects the experience of the larger debtors. For the smaller debtors, median ratios in fact increased over the period.

Smaller debtors generally have received less favorable terms on syndicated credits than have the larger debtors, presumably reflecting bankers’ perceptions that loans to smaller countries with less diversified economies are riskier than loans to large countries (Table 7). Over the last four years, however, there has been a considerable narrowing of the differential between the two groups. The maturities of syndicated credits to the smaller debtors have generally been shorter than those to larger debtors.

Table 7.Developing Countries: Terms on Syndicated Credits, 1978–81
Median Values1
Larger debtorsSmaller debtorsAllAverage2
(In per cent)
Spreads over LIBOR
(In years)
Source: Fund staff calculations based on World Bank data.

For countries listed in Table 3. Countries for which data on terms on syndicated credits are available vary from year to year, which possibly affects median values.

According to the Organization for Economic Cooperation and Development, weighted average terms on all syndicated credits to developing countries, excluding those which are members of the Organization.

Source: Fund staff calculations based on World Bank data.

For countries listed in Table 3. Countries for which data on terms on syndicated credits are available vary from year to year, which possibly affects median values.

According to the Organization for Economic Cooperation and Development, weighted average terms on all syndicated credits to developing countries, excluding those which are members of the Organization.

Prudential Issues in Bank Lending to Developing Countries

A major area of uncertainty for the future is the extent to which banks will be prepared to continue increasing their exposure to developing countries. Bank claims on those countries have grown at an annual average rate of 25 per cent since 1973. While a slowdown in the growth of debt to banks is desirable and inevitable—bank debt cannot continue to increase indefinitely relative to other economic aggregates—the real question is whether prudential concerns might lead banks to slow their lending to those countries so sharply as to produce unmanageable adjustment pressures.

As noted earlier, there currently appears to be some degree of strain in the capital positions of some banks, but this in itself is not likely to be a major constraint on international bank lending, particularly over the medium term. A more important issue affecting developing countries is the growing concentration in bank loan portfolios of the debts of certain of these countries. Since 1973, banks’ international claims have grown steadily relative to their overall balance sheets (Chart 7). Moreover, the proportion of international claims represented by claims on non-oil developing countries, and on the largest borrowers among them, has risen to some extent.8

Chart 7.Concentration of International Bank Claims, 1973–811

1 Excludes interbank transactions within the BIS reporting area.

2 Excludes Fund member countries except Hungary, which became a member in mid-1982.

Against this background a number of major banks have indicated that for an increasing number of developing countries their exposures have reached a point where further significant increases are felt to be imprudent, even where the borrower is considered to be creditworthy. Such countries can nonetheless expect to continue obtaining large volumes of new bank lending. In the first place, banks’ internal country lending limits tend to move upward over time as bank capital increases. In recent years, the data available suggest that aggregate bank capital in industrial countries has increased by about 10 per cent a year. At the present time, moreover, for any creditworthy country, there are large numbers of banks which still have room within their internal exposure limits to increase their lending substantially. For example, during 1981 the lending activity of banks with ownership in Arab countries and of U.S. regional banks became more important; in both cases there is room for considerable further expansion of claims on developing countries. Finally, country limits are far from immutable, and they tend to be adjusted if profitable lending opportunities arise in countries that are considered creditworthy. Some large borrowers who have encountered market resistance in obtaining the volume of funds they desired have found that at higher lending spreads funds were more readily available. All in all, exposure concentration in itself does not seem likely to prove a major impediment to the continuity of large-scale lending, even to those countries that have borrowed heavily in the past, although it could result in higher lending spreads for some such countries.

Country creditworthiness is a much more immediate concern for banks than exposure concentration. By a number of aggregate measures, the creditworthiness of developing countries as a group has recently deteriorated. For the non-oil developing countries, and each of the major analytical subgroups, the ratios to exports both of external debt and of external debt service payments rose sharply in 1981 and are projected in the World Economic Outlook to do so again in 1982. Current account deficits remain high, and there is little immediate prospect for improvement, at least in current dollar terms. Some major borrowers have benefited from the recent softness in the oil market, but others are themselves exporters of oil. The persistence of high nominal interest rates in the face of declining commodity prices is also a negative factor.

The outlook for the medium term is somewhat more encouraging, with debt and debt service ratios and real current account deficits projected in the World Economic Outlook’s central scenario to fall from present levels by the mid-1980s. In any case, lenders focus on individual countries not on broad aggregates, and the important factor here is that many countries for which economic indicators such as debt ratios look high do manage their economies well. Adverse external developments, moreover, are often dealt with successfully through programs of adjustment. Nonetheless, banks indicate that the number of countries which they view as “problem” countries has been growing. One objective indicator of problems is the existence of payments arrears: some 30 Fund member countries are known to have had payments arrears at the end of 1981.9 Even when arrears have been eliminated through, for example, adoption of a debt rescheduling agreement, banks have shown reluctance to increase their exposure in the countries concerned.

Changes in perceptions of country risk have sometimes displayed a sort of “bandwagon” effect, waves of enthusiasm for particular countries being followed by waves of pessimism. Recently, such a process seems to have emerged not only regarding individual countries but on a regional basis. Thus, some countries in Eastern Europe whose economies and payments prospects are not themselves the focus of major questions have found that their access to bank lending has been sharply limited by the financial problems of neighboring countries, against a background of international political uncertainty. Similar, although much more moderate, difficulties reportedly have been encountered by some countries in Latin America in response to problems in other countries in the region.

Restructuring of Debt to Commercial Banks

The role of the international banks in financing the payments deficits of nonindustrial countries increased dramatically during the 1970s, especially in the period immediately after the first round of oil price increases. As a result, both of the much greater volume of bank lending and of the fact that bank credits carried shorter maturities and higher interest costs than the official loans and guaranteed suppliers credits, which previously had been the predominant source of finance for the developing world, the share of amounts owed to banks in total debt service of developing countries escalated sharply during the latter part of the decade and the restructuring of bank debt emerged as an increasingly important element in the efforts of countries experiencing payments difficulties to restructure their foreign obligations.

In a study prepared late in 1980, the Fund staff analyzed the experience of six countries which had sought multilateral reschedulings from the banks during the last half of the 1970s. Of these six countries, Peru was the first to reach a formal rescheduling agreement with the banks in 1978, Jamaica and Turkey followed in 1979, and the agreements for Nicaragua and Zaire were concluded in 1980. Although discussions between Sudan and the banks began in mid-1979, an agreement was not signed until the end of 1981.

Since that staff study was issued, a number of new bank debt restructuring cases have emerged. Both Bolivia and Poland reached agreement with the banks during 1981, although the Polish agreement was not signed until April 1982. In addition, four other countries—Costa Rica, Liberia, Senegal, and Romania—are in the process of negotiating multilateral reschedulings of their debts to the banks. There are similarities, but also some important differences, between the experiences of those countries covered by the staff study and the more recent bank debt rescheduling cases.

One of the major findings of the 1980 study was that, in every case, bank lending had expanded very rapidly over a short period and then declined precipitously when the possibility of debt servicing difficulties became apparent. This description is equally valid for the more recent cases.

At some time over the three years through December 1980, Bolivia, Costa Rica, Romania, and Senegal each more than doubled their debt to the banks within 24 to 30 months. Poland’s debt to the banks increased by about 50 per cent between June 1978 and December 1980. For all these countries new bank lending came to a standstill in 1981, and in some bank exposure actually declined. (As the statistics on bank claims on Liberia include large claims on borrowers who are not integrated into the Liberian economy, the statistics do not permit an isolation of the debt of the Liberian economy itself.) Romania was the only one of these six countries to obtain new medium-term bank credit commitments in 1981, and almost all Romania’s commitments were secured in the early part of the year. While the ultimate responsibility for appropriate debt management rests, of course, with the borrowing country, the banks’ willingness to lend substantial sums over short periods, followed by an abrupt closing of the window when problems emerged, was in itself a contributing factor to the difficulties a number of these countries faced in managing their domestic economies and their international payments position.

Although the percentage expansion in bank claims was lowest in Poland, the absolute amounts were by far the largest. At the beginning of 1981, Poland’s debt to the banks was $16.2 billion, more than four times the total bank exposure to any of the countries that had previously sought a debt rescheduling. Poland is, in fact, the sixth largest debtor to the banks among the nonindustrial countries. Of the remaining countries that have sought a rescheduling of their bank debts, none ranks among the 10 largest developing country debtors to the banks, only Romania ranks among the 20 largest, and only Turkey and Peru are added if the list is extended to the 30 largest. The banks’ exposure to the remaining 7 countries (excluding Liberia) ranges from $350 million for Senegal to $1.3 billion for Costa Rica, and taken together these 7 countries account for less than 2 per cent of the banks’ claims on the nonindustrial world. While total bank exposure to countries experiencing payments difficulties has not been large when measured against the banks’ international claims, at least until Poland developed debt problems, the amounts involved have generally been quite large compared with the dimensions of the borrowers’ economies.

As noted in the 1980 Fund staff study, the absence of established procedures and generally accepted precedents led to protracted discussions in some of the bank debt rescheduling cases that emerged during the latter part of the 1970s. The more recent discussions have, it appears, been facilitated by the experience gained in the earlier cases. In the mid-1970s both the banks and the countries concerned were extremely reluctant to consider a formal rescheduling, even when it was apparent to all concerned that a serious problem existed. Efforts were made instead, generally with little success, to arrange new loans or refinancing packages. More recently, however, all parties concerned appear more willing to entertain the possibility of a formal rescheduling from the outset and to recognize that it is in the interest of both the banks and the debtor that discussions not be prolonged unduly.

The scope and terms of most (but not all) bank debt rescheduling agreements have fallen within certain broad outlines, and proposals made by both sides now tend to reflect these precedents. Agreements have typically covered between 80 and 95 per cent of principal due to the banks over a period of 12 to 24 months. Implementation of the agreement has usually been linked to the existence or entry into force of an operative program with the Fund during that time. Principal has generally been rescheduled with a grace period of two to four years and a total maturity of between five and seven years, with interest on the rescheduled amounts being charged at spreads ranging from 1 ¾ to 2¼ per cent above LIBOR.

The banks are extremely reluctant to reschedule interest. Only in Nicaragua has future interest due been incorporated in the rescheduling agreement, and the banks have usually insisted that any interest in arrears be made current before an agreement is signed. Partial deferral of repayment of interest arrears has been provided for in a few cases, but only for relatively short periods. In addition to interest in arrears, the banks also generally require that amounts of principal in arrears and not rescheduled, i.e., between 5 and 20 per cent of the amount in arrears, be paid before the agreement is signed or shortly thereafter. In some cases, especially where the discussions have been protracted, the amount of arrears and, hence, the required payment at signing can be large. Such upfront payments have been a major issue in a number of recent cases and the difficulties faced by the debtor in raising the required amounts have threatened the conclusion or implementation of some agreements.

Most of the bank debt reschedulings are too recent to allow an assessment of their impact on the country’s ability to regularize its payments situation and restore its credit standing. Some tentative conclusions can, however, be drawn for Jamaica, Peru, and Turkey, the three countries which concluded their initial agreement with the banks in 1978 or 1979. Of these three, Peru was the quickest to regain access to the markets, with the banks providing major new medium-term commitments within a year of the conclusion of the restructuring agreement. Peru’s credit standing was, however, buoyed during this time by very favorable price developments for its exports. The banks were rather more cautious in renewing their lending to Jamaica and Turkey, but by 1981 normality in commercial financing relations was being approached for both of these countries. In 1981, the banks also arranged a major medium-term credit for the Bank of Jamaica, and by early 1982 Turkey had re-entered the syndicated credit market. In June 1981 Jamaica concluded a further rescheduling agreement with the banks covering amounts due over the two years through April 1983, and in early 1982 the banks signed an agreement extending by two years the grace period and maturity on the original rescheduling package for Turkey that had been agreed in 1979. All three of these countries were, at the time of their re-entry into the markets, successfully implementing adjustment programs supported by the use of Fund resources.

Developing Country Access to International Bond Markets

Recent Experience

International capital market lending to developing countries has been almost entirely channeled through banks. Bond market lending only once reached 15 per cent of total market flows and has usually been less than 10 per cent. Despite these low proportions, about 20 to 25 developing countries have made at least one issue or placement each year since 1975, and a few countries have tapped these markets for from $500 million to $2 billion in some years.

In the last decade, developing country bond issues reached a peak of $5.8 billion in 1978, but then declined to $2.5 billion in 1980 before recovering somewhat to $3.5 billion in 1981. Although part of the decline after 1978 reflected a general deterioration in bond market activity owing to the effects of rising inflation and greater variability of interest and exchange rates, bonds marketed by developing countries fell even more rapidly than those issued by other borrowers. The proportion of total international bond issues accounted for by developing countries fell from about 15 per cent in 1978 to only 6 per cent in 1980 before recovering slightly to 7 per cent in 1981 and to 8 per cent in the first quarter of 1982. This decline in the relative position of the developing countries was partly related to the sharp increases in their current account deficits in the 1979–81 period, which adversely affected the perceptions of bond investors regarding the creditworthiness of non-oil developing countries. It also reflects the fact that in times of difficult market conditions underwriters consider the risk of bringing other than the “best names” to the market to be unacceptably high.

Most developing country bond issuance in the international bond markets has primarily represented the activities of only a few major issuers and the share of the bonds issued by these major issuers has been increasing. In 1978, for example, 53 per cent of all developing country bonds were issued by the four largest developing country bond issuers. By 1981, the share of the four largest borrowers had increased to 78 per cent and, during the first quarter of 1982, increased to 89 per cent. While the composition of the top four issuers has varied over time, Israel and Mexico have generally been among the largest issuers.

While Eurobonds of developing countries are marketed in many markets, foreign bond issues have been much more focused on a few major financial markets. The relative share of the financial markets in Switzerland, Japan, the Federal Republic of Germany, and the United States has varied over time. In 1978, for example, almost 18 per cent of developing country public issues of foreign bonds were sold in Germany; in 1980 only 6 per cent were sold in Germany and in 1981 there were virtually no such sales. The share of the Japanese and Swiss markets each remained between 20 per cent and 40 per cent throughout the 1978–81 period.

There have not been any major institutional changes in recent years affecting developing country access to international bond markets. In some major industrial countries, policies designed to limit outflows of capital have at times restrained external bond issues in their currencies. While the policies were not discriminatory in principle, in effect they sometimes had a disproportionate impact on developing country issuers. The policies, which were initiated as exchange rates came under pressure, have been or are now being relaxed. A number of countries are also taking steps to reduce registration costs and increase competition in the bond market. For example, the new shelf registration procedure in the U.S. bond market will help to reduce the costs of issuance for both domestic and foreign borrowers who regularly issue in that market.

Prospects for Developing Country Bond Issues

The short-term outlook for international bond issues by developing countries depends on the general conditions that are likely to prevail in international financial markets (discussed in the next section), but more importantly on country-specific factors influencing borrowers’ perceived creditworthiness. As discussed in Appendix II, the last decade has witnessed a decline in issuance activity in real terms, a shortening of bond maturities, the emergence of higher real yields, and a redistribution of a portion of the risk of bond price variability from the investor to the issuer. These changes reflected the responses of bond market participants to higher levels of inflation and interest rates, greater variability of exchange rates and interest rates, and large capital losses experienced by bondholders in the late 1970s. A significant recovery in the real size of the international bond markets during the 1980s will take place only if there are fundamental improvements in the expectations of investors and borrowers regarding the levels of inflation and nominal interest rates and the general stability of financial and exchange markets. Even if these general macroeconomic and financial market conditions are fulfilled, developing countries will be able to have easy access to the international bond markets only if they are perceived to have crossed the threshold of “market credibility,” have a record of effective economic management, and if international conditions are conducive to a continued improvement in their economic positions.

The discussion is based largely on data reported semiannually by the Bank for International Settlements (BIS) on certain aspects of bank exposure to countries outside the BIS reporting area. Compared with the quarterly BIS data, which underlie most of the analysis of this paper, the semiannual series are more comprehensive and provide a more accurate classification by country. On the other hand, they extend back only to 1978 and are currently available only up to June 1981.

As bank debt includes short-term debt while the other debt figure does not, the ratios cited here are larger than the true share of bank debt in total debt.

The most relevant measure of exposure concentration is exposure relative to bank capital. Such a calculation should be cast in terms of particular banks’ exposure to particular countries, but a broad indicator of recent changes can be calculated on the basis of aggregate capital of banks. For banks in the BIS reporting area, the general order of magnitude of total capital appears to have risen from $240 billion at the end of 1978 to $330 billion at the end of 1981. On this basis, claims on the 20 largest borrowers among the non-oil developing countries relative to bank capital rose from about 50 per cent to over 70 per cent between the end of 1978 and the end of 1981.

Most of these are not large borrowers from banks; bank claims on such countries amounted to only 7 per cent of total bank claims on non-oil developing countries at the end of June 1981. On top of the $20 billion in bank debt of such countries, one nonmember country with substantial arrears—Poland—accounted for a further $15 billion in bank claims.

    Other Resources Citing This Publication