1930s EXPERIENCE SUPPORTS SELECTIVE DEBT RELIEF

The 1930s experience of sovereign default offered clear lessons for managing the 1980s debt crisis, said CEPR Director Richard Portes at a lunchtime meeting on October 21. Strategies must differentiate among borrowers, he argued. Analysis of the 1930s experience suggests that defaults were not random and could be explained by simple economic variables. The research also indicates that despite the widespread defaults in the 1930s, which did shift some of the debt burden to the creditors, the ultimate settlements on government-backed loans yielded very respectable rates of return. Portes argued that selective write- offs or easing of repayment terms might be indicated for countries otherwise likely to default. This might offer a bearable and historically justifiable redistribution of the debt burden.

Portes, who is Professor of Economics at Birkbeck College (University of London) first wrote on debt eight years ago in Foreign Affairs, when he analyzed why East European borrowing in the 1970s was unsustainable and predicted Poland's 1980-81 debt crisis. The research he discussed at the lunchtime meeting was done jointly with Barry Eichengreen (Harvard and CEPR), and is described in CEPR Discussion Paper No. 75, entitled 'Debt and Default in the 1930s: Causes and Consequences'. The views expressed by Portes were his own, however, and not attributable to Eichengreen or to CEPR, which takes no institutional policy positions.

The multiple defaults of the 1930s offer suggestive parallels to reschedulings in the 1980s, Portes noted. Illiquidity was not then confined to any one country or region, and the debts arose from widespread balance-of-payments disequilibria in both the 1920s and the 1970s. In both periods borrowers' problems were due partly to their domestic policies, partly to disturbances in the world economy including real interest rate shocks, commodity price fluctuations and exceptionally severe recessions in industrialized regions, and partly to systemic features leading to a collapse of lending to the debtors. Now as in the 1930s, there is a significant political dimension to the creditor-debtor relation which was largely absent from nineteenth century debt problems.

Yet the institutional arrangements governing international lending have changed fundamentally. The switch from bond to bank finance, with different loan contracts and fewer creditors party to negotiations, is said to facilitate rescheduling of debt as an alternative to outright default. The IMF can serve as an external lender of last resort to illiquid borrowers, and it gives the capital market information on domestic adjustment programmes. Macroeconomic stabilization policies should preclude a business cycle downturn like the Great Depression, which would in turn hit borrowing countries. On the other hand, the financial institutions were merely intermediaries in selling bonds in the 1920s. Now their capital is at risk, and this has implications for the stability of the entire financial system.
To create a basis for comparison, Portes and Eichengreen had employed regression analysis to show that across countries and over time, levels of debt during 1930-38 were closely related to GDP and positively associated with the degree of 'openness' of the economy, as measured by the ratio of imports to GDP. This confirmed similarities to the 1970s-1980s.

Existing accounts of countries' decisions to default in the 1930s cannot explain why and how the incidence and extent of default varied so greatly among debtors. These accounts cite idiosyncratic national circumstances or their converse, irresistible 'bandwagon effects'. Some simply maintain that default was the only feasible alternative for debtor developing countries facing so great an external shock. Portes offered new results showing that the proportion of a country's debt in default during 1934-38 was positively related to its debt/GDP ratio, the extent of deterioration of its terms of trade from 1929 onwards, and the percentage increase in its government budget deficit during 1929-31. These results, he argued, were inconsistent with the view that there was no alternative to default for developing countries during the Great Depression, and also with the view that non-economic factors were decisive. On the contrary, Portes argued, simple economic variables go a long way towards explaining the incidence and extent of default in the 1930s and might similarly indicate appropriate candidates for partial debt relief today.

Portes then presented new estimates by Eichengreen and himself of the rate of return on foreign loans floated in the 1920s. The internal rate of return on default-free loans was close to 6% for both dollar and sterling issues, but default hit dollar bondholders much harder than those holding sterling bonds. For example, the average realized internal rates of return (weighted by issue value) on the full samples of government-backed loans are 3.25% for dollar issues and 5.4% for sterling issues. Not only are both rates positive, but the latter compares favourably with the average yield on consols during the period. Despite the preoccupation with sovereign default, investors who lent directly to national governments ultimately received respectable rates of return; settlements appear to have been relatively favourable.

Would such rates of return be unacceptable for the 1980s-1990s? Portes argued that creditors may be unable to continue to insist on payment in full. Bank profitability, even solvency, are not obviously more important to systemic stability than debtor country ability to maintain a regular pattern of (possibly reduced) payments without their debt continuing to increase faster than exports. Both considerations are important for stabilty, and the appropriate policy, according to Portes, involved sharing of the burden created by internal and external economic and political shocks.

Although interest rates are still high relative to export growth rates, the adjustment programmes implemented under IMF conditionality were typically more than adequate to guarantee solvency - if they are sustainable. The export surpluses generated in 1983-84 by these programmes were actually large enough to ensure that the present value of future debt would fall to zero, as Daniel Cohen demonstrates in the first issue of Economic Policy. Portes argued, however, that these programmes are not likely in any case to prove sustainable. His analysis of the 1930s experience suggests where the strains will appear and how to manage them.

Rate of Return on Foreign Loans Floated in the 1920s:
New Estimates by Eichengreen and Portes