William R. Cline
Institute for International Economics, Washington, D.C.
Nearly six years after the outbreak of the debt crisis, policymakers and the private financial sector can claim some success in managing the problem. There is no longer imminent danger to the international financial system, as the large banks have increased their capital relative to claims on developing countries. The widespread fears at the outset that there would be a wave of defaults similar to those of the 1930s have so far proven unfounded. Despite the proliferation of proposals for concessional debt relief, key debtor countries such as Brazil, Mexico, Venezuela, and Chile have shown important progress in recovering from the debt crisis without such relief. Even the widespread expectation that banks would completely cut off new lending has been contradicted repeatedly, as banks have provided large new money packages each time a major debtor country has requested support and entered into an IMF-supported adjustment program. Despite these and other gains, it seems increasingly likely that in some low-income countries, particularly in Africa, concessional relief may be necessary. Even for the largest debtor countries, market-oriented “voluntary debt relief holds promise as a source of increased flexibility in management of the debt problem.
Latin American debt
The prevailing perception of the debt problem, in general, and for Latin America in particular, is unduly negative. Thus, even the World Bank staff recently concluded that after five years the debtor countries’ creditworthiness has “stubbornly failed to improve” because “no country involved in rescheduling its debts has significantly reduced its debt ratios.” Curiously, the report looks only at quantity, not price. Just as the number of barrels of oil would convey a false impression about the burden of oil imports because oil prices have fallen by half, the ratio of debt to exports fails to consider the sharp decline in the price of debt. In 1982 the average cost of bank loans was 17 percent; in 1987 it was only 7½ percent.
The proper measure of the debt burden is the ratio of interest payments to exports of goods and services, which incorporates the joint effect of debt and average interest cost. By this measure, there has been clear progress toward restored creditworthiness in Latin America. The region as a whole has reduced its interest/export ratio from 41 percent to 30 percent since 1982; Brazil, from 57 percent to 34 percent; and Mexico, from 41 percent to 29 percent. And although interest rates will be up somewhat in 1988, exports should rise to offset that increase.
Even more important, Latin America has moved from severe recession five years ago to economic recovery. Economic growth increased from an average -2 percent a year in 1982-83 to an average + 3.4 percent in 1984-87 for Latin America as a whole. The oil price collapse of 1986 caused a temporary setback for Mexico and other oil exporters, but Mexico restored positive growth in 1987 and should increase the rate modestly this year.
Commodity prices have been an area of disappointment. In the global recovery of 1984 they rebounded from their extreme lows of 1982, but they then eroded to even lower levels by mid-1986. They have been on an uptrend since, however, and by the fourth quarter of 1987 commodity export prices had risen by 75 percent from their 1986 trough in Chile and Peru, by 21 percent in Argentina, and by 9 percent in Brazil (where the low point was in early 1987).
Progress in emerging from the debt problem has been greatest where governments have implemented sound economic policies. Sharp devaluation and fiscal adjustment in Mexico to confront lower oil prices have boosted manufactured exports by 40 percent annually for the past two years. Proper fiscal and exchange rate policies have enabled Chile to reduce its interest/export ratio from 50 percent to 27 percent since 1982 while achieving growth of almost 5 percent and an inflation rate of less than 15 percent annually from 1984 through 1987. Even where major policy mistakes were made, as in the excessive demand and frozen exchange rate under Brazil’s 1986 cruzado plan, reduction of these distortions has permitted a recovery in trade (with Brazil’s trade surplus back up from zero in late 1986 to $11 billion in 1987). By early 1987 Brazil ended the moratorium it had declared on long-term bank debt the previous year (with President Jose Sarney calling the moratorium the biggest error his government had made), and returned to negotiations toward debt rescheduling and new money.
Strategy for major debtors
Although progress on Latin American debt has been greater than generally recognized, it has not been sufficient to restore voluntary lending (as opposed to new money packages coordinated under pressure). Even if the interest/export ratios had already subsided to the range of 25 percent or less that would be compatible with restored creditworthiness, the banks with large exposure would continue for some time to prefer to diversify their portfolios away from Latin America after their narrow escape from disaster. Yet the region needs capital to invest in growth and should be receiving new loans to cover at least a portion of interest due.
One solution repeatedly proposed by some legislators, academics, and politicians in Latin America is for the banks to forgive an important part of the debt. For the major debtor countries, this remedy is unnecessary and would be counterproductive. Defaults in Latin America in the 1930s locked the countries out of the capital markets for three decades, and forced forgiveness today could do the same.
There is a need, however, for a medium-term strategy for assuring growth in debtor countries and achieving the transition back to voluntary lending. Its outlines could be as follows. First, the banks would negotiate multi-year new money arrangements with the debtor countries. These packages would assure new financing on a scale that would decline over time. While avoiding any rigid link to interest due, the multi-year schedules for new lending could in practice provide for the refinancing of a significant but descending portion of anticipated interest obligations. For example, in early 1988 the banks agreed in principle to lend Brazil an amount equivalent to about half of interest owed on long-term debt for 1987-88. A multi-year program could provide for further lending at say four-fifths of this amount in 1989, three-fifths in 1990, and so on, declining to zero by 1993. Arrangements for international monitoring of country policy by the IMF or other institutions would be a part of these programs.
Second, to compensate for the declining amounts of new loans, the country would obtain funds from other sources by gradually shifting toward voluntary private lenders and, ideally, the World Bank and other public sources. It would offer incentives to attract such voluntary lending (e.g., options to lend directly to private sector clients and to convert debt into- equity, bonds convertible into commodities, project related loans, interest spreads of 1½ to 2 percent instead of under 1 percent, and so forth). As the recent issue of $100 million in bonds by Venezuela demonstrates, gradual reentry into the private capital market is possible.
Third, the banks would agree that the country could use any excess income from unanticipated export success for the purpose of buying back debt at a discount on the secondary market, rather than losing the new money pledged under the multi-year package. At a typical 40 to 50 cents on the dollar, debt bought from the secondary market would help alleviate the country’s debt burden, while giving the selling bank immediate liquidity. The smaller, more impatient banks would exit from the new-money process and clear their books of Latin debt. The larger banks with a longer time horizon in the region would benefit from the reduction in the country’s debt to other banks, and the resulting improvement in the country’s ability to honor remaining claims in full. Repurchases of debt would put upward pressure on the secondary market price, and a revival of prices on this market much closer to 100 cents on the dollar will be an ultimate prerequisite for a full return to voluntary lending.
Fourth, the banks would accept the principle that all would participate either through the provision of new lending or through “exit bonds;” the process should not break down because of “free rider” banks that collect interest but provide no support to the debtor country. There have been two notable attempts at exit bonds. In early 1987 Argentina offered 25 year 4 percent bonds in exchange for bank claims with the understanding that takers would be exempt from future new lending. There were no takers, because the bonds paid too little and retained the risk of non-payment. In early 1988 Mexico, with the help of Morgan Guaranty, issued bonds backed by US Treasury zero-coupon bonds. The idea was to obtain great leverage, but the attraction of the collateral was judged by the market at no more than about 15 cents on the dollar, because it guaranteed principal but not interest, while interest provides the great bulk of the value of a 20-year instrument. Even so, Mexico managed to convert $3.6 billion in outstanding debt into only $2.5 billion in new bonds, for a significant savings.
When evaluating exit bonds, banks face the traditional portfolio decision of trading off risk versus reward. For a bank to give up its claim to 100 cents on the dollar in return, say, for only 50 or 60 cents, it must be convinced that the risk drops sharply to compensate for the concession. If the country and all of its bank creditors formally agreed that exit bonds held senior status for repayment purposes and interest would be punctually serviced at the front of the queue even if other claims had to be rescheduled, the risk factor would largely disappear and more banks would be prepared to adopt the exit bonds even without zero-coupon treasury bonds or other instruments as guarantees. The exit bond with seniority would be an appropriate instrument for “voluntary debt relief based on the market, because it would permit the country to receive much of the discount present in the secondary market.
Several recent proposals would interject the government into this process by creating an official intermediary that would buy up debt at a discount and pass the discount along to the country. Plans that would make such sales of existing debt mandatory (including by punitive regulatory and accounting measures) have all of the drawbacks of mandatory debt forgiveness in general. Proposals that are truly voluntary still face the issue of whether the public sector is prepared to commit large resources to the purchase of debt and transfer of the risk from the private to the public sector. In addition, the transition from market-based, voluntary debt relief to more formalized, government-sponsored programs for discounted conversion runs a considerable risk of sending the signal to debtor countries that they need not honor their debts because the governments of industrial countries would now take care of the problem. There could be counterproductive changes in economic policies of the debtors, and increased pressure politically within debtor countries toward insistence on debt forgiveness (raising the “moral hazard” issue), with negative long-term effects on the country itself.
The four-point approach outlined here amounts to an enhanced version of the Baker plan, including its flexible “menu approach” which tailors arrangements to the preferences of individual banks. Avoidance of a severe global recession or a renewed explosion in international interest rates remains a precondition for success of the basic Baker plan strategy. Because the oil shocks and high inflation of the mid-1970s and early 1980s are behind us, it should be possible to keep any new economic slowdown short and follow it with compensating above-average growth. If policymakers in the industrial countries can do their part to assure the right international economic environment, those Latin American countries that follow sound economic policies should be able to emerge from the debt problem in a way that restores their creditworthiness and access to capital markets, and thereby facilitates integration into the international economy and sustained growth over the longer term.
Both the problem and the solution are different for many of the African countries (and for a short list of low-income countries in Latin America such as Bolivia). For an important part of Africa, outright forgiveness of a significant portion of existing claims held by foreign official agencies, together with conversion of new official development assistance to a grant basis, may be necessary.
At the end of 1985, 21 Sub-Saharan African countries (Group A) accounting for $35 billion in external debt had ratios of debt to exports of goods and services exceeding 300 percent. Their weighted average ratio of debt to exports of goods and services was 471 percent. In contrast, 23 countries (Group B) owed $48 billion (of which Nigeria alone accounted for one-third) but had a weighted debt/export ratio of only 168 percent. Only the first group seems likely to require debt forgiveness.
In sharp contrast to Latin American countries, many African countries have not been reducing the growth of debt/export ratios by running trade surpluses. Instead, their trade balances have been in large deficit. Group A African countries had a merchandise trade deficit of $4.4 billion in 1986, or 60 percent of their exports of goods and services. Moreover, Africa pays an average interest rate of 6 percent on its debt, surprisingly high for countries eligible for concessional assistance. While nominal export growth of perhaps 10 percent would potentially overshadow “inherited” growth of debt from interest on past debt, and thereby permit gradual reduction in debt/export ratios, the trade deficit swamps export growth and causes the ratio to spiral upward.
Many African countries, therefore, acutely need to increase export growth. While commodity prices are beyond their control, there is much that the governments can do to increase exports and reduce trade deficits, especially by adopting realistic exchange rates and agricultural prices. Beyond self-help, foreign official debt relief may be required. In contrast to the large Latin American nations, countries in Sub-Saharan Africa owe 85 percent of their debt to official sources, not banks or other private creditors. Most of these nations are low-income countries eligible for concessional assistance from the International Development Association and bilateral sources. For these countries, debt forgiveness would have much less potential adverse impact than for countries that rely more on private capital sources.
Debt relief for Group A African countries is essentially a matter of strategy in concessional development assistance. Some donors want their earlier debts to be serviced even as they provide new aid. In this way they retain the right to impose policy conditionality. However, a more coordinated and comprehensive solution may well require forgiving a substantial portion of official claims on African countries, and limiting the bulk of new official lending to grants so that debt service obligations do not mount. The IMF’s Structural Adjustment Facility is an important potential source of official finance that can supplement IDA and bilateral programs, and still retain conditionality to spur economic reform.
It would be possible to supplement official debt forgiveness for Group A African countries with buyback programs for the small amount of private debt that these countries owe. The Bolivian case provides a precedent: with funds provided by official donors, Bolivia expects to repurchase about half of the debt owed to banks at 11 cents on the dollar or less. Because Group A countries in Africa owe only $5 billion to private creditors; a similar program would cost well below $1 billion in additional official support.
It is crucial to recognize, however, that forgiveness of official claims on low-income countries should not be assumed to be a valid prescription for the debt problem as a whole. Even within Africa it is unlikely that forgiveness of debt for Group B countries (including Nigeria and CÒte d’lvoire) is either necessary or favorable for the countries themselves in the longer run.
In sum, there has been considerably more progress on international debt problems as a whole than is generally recognized. Reinforcement of the Baker plan with multi-year new money packages, exit bonds with seniority, and the right for successful exporters to use extra earnings to repurchase debt at a discount in the secondary market, should be sufficient for most of the major debtors to achieve gradual restoration of creditworthiness along with adequate domestic growth. In contrast, for a special list of heavily indebted, low-income countries (primarily in Africa), some forgiveness of primarily official debt may be appropriate. While a significant number of countries is on this list, their aggregate debt is only a small fraction of the total for developing nations. For all debtor countries, progress on debt will depend not only on arrangements on debt per se, but more fundamentally on appropriate domestic policies (especially on fiscal balance and the real exchange rate) and successful management of the international economy by the leading industrial countries.