LDC Debt
What price buy-backs?

The difficulty experienced by some less developed countries (LDCs) in servicing their sovereign debt and the willingness of lenders to sell their claims on LDC debt at substantial discounts in the secondary market are evidence of a substantial LDC debt problem. At a lunchtime meeting on 8 May, Daniel Cohen criticized proposals for one-off writing-down of LDC debts. Instead he advocated a `generalized buy-back' scheme, under which all transfers from debtor to creditor, that is both debt service and repurchases, would take place at a discount representing the market value of the debt.
Cohen is a Research Fellow in CEPR's International Macroeconomics programme and is currently engaged with Richard Portes (CEPR and Birkbeck College, London) in a research project on the secondary market for LDC debt, supported by the Overseas Development Administration and the Bank of England. The research underlying Cohen's talk is reported in Discussion Paper No. 312. Financial support for the meeting was provided by the Ford and Alfred P Sloan Foundations as part of their support for the CEPR's International Macroeconomics programme. The views he expressed were his own, and not those of any donor organization or of CEPR, which takes no institutional policy positions.
Cohen began by noting that the 1985 Baker plan had correctly acknowledged that economic growth on the part of the debtor nations resembled a public good. It would benefit all parties aside from the domestic benefits, it would enable creditors to receive a reasonable return on their loans and provide larger markets for exporters but it was hard to see how the market alone could facilitate such growth. The Baker plan was designed in part to overcome the downward spiral of low growth and no new lending. The problem with it, Cohen argued, was that breaking the downward spiral required a definite commitment to new lending in the future, yet Baker had contemplated no such binding commitment. The Brady plan, put forward earlier this year, had recognized the need for some measure of debt relief, but was vulnerable to the same criticism, according to Cohen. Because it included no binding requirement on banks to lend in the future, it offered nothing that market participants would not do anyway.
The idea that some LDCs' debt is now `too big' to be economically efficient is supported by the existence of accumulated arrears, failed reschedulings and the willingness of lenders to sell their debt claims at a discount in the secondary market. Economists have developed two approaches to analysing this inefficiency, Cohen noted. Some have argued that the existence of any divergence between the face value of debt and its price in the secondary market constitutes a relative price distortion which must inevitably be a source of some economic inefficiency. The second approach, initiated by Sachs and by Krugman, suggests that there is a `debt Laffer curve'. At low levels of debt, the total return to the lender increases with the face value of the debt. The debt could, however, reach such a level that the debtor defaults or its economy has been so damaged that its ability to afford repayments is reduced. At or above this level, lenders may actually increase the total returns they receive by reducing the face value of the debt. Reappraising this problem in Discussion Paper No. 312, Cohen maintained that these two approaches are in fact equivalent: the underlying proposition of the debt Laffer curve approach is the relative price distortion revealed in the secondary market.
Cohen argued that the key to the debt problem is not, as often proposed, a willingness once and for all to write down the face value of debt, but a commitment to reducing the flow of funds over time from debtor to creditor. In order to demonstrate this, he observed that the secondary market valuation of debt reflects the market's assessment of the state of each debtor's economy and its total ability to service debt. Writing off a limited proportion of a country's debt will have a negligible impact on the country's real ability to afford to service debt. Consequently there will be a corresponding rise in the market value of the remaining debt, and its total value on the secondary market will be approximately unchanged. Buy-backs and repurchases on the secondary market may therefore have little or no effect on a country's debt burden, Cohen argued, unless they involve an unrealistically high proportion, say a half, of the debt. At worst the money spent on schemes like the Brady plan may prejudice tax-payers against future, more effective schemes.
If debt write-offs may have so little impact on a country's total effective indebtedness, what would reduce the debt problem? In his theoretical work, Cohen had defined the optimal rescheduling strategy, that is the maximum present discounted value of repayment that a lender could possibly expect. In proposing his own solution, he again drew attention to the relative price distortion revealed in the secondary market. Investors can purchase claims on a country's debt at a discount, but the debtor country itself must pay the full nominal value in order to repay its debt. This distortion can be removed, and optimal rescheduling achieved, Cohen argued, by allowing debtors to service their debt at a value reflecting the price at which it trades in the secondary market. He called this a `generalized buy-back' scheme, because all transfers from debtor to lenders would be priced according to the debt's market value. The actual secondary market value should not be used, however, to set the discount at which the debt should be serviced, as this might create incentives for debtors to manipulate their economy or economic indicators so as to reduce the debt's market value. Instead creditors and debtor should agree on an equilibrium market price as part of their rescheduling agreement.
Such a scheme would be economically efficient, Cohen noted, as it would eliminate the relative price distortion. It would be good for debtors, as it would mean that economic advance and trade surpluses would not automatically be lost to higher repayments. It would be good for the banks, as it would provide debtors with an incentive to pay and it would not reduce the market value of banks' claims on the debt. The only respect in which the scheme may not be optimal is in the event of a sequence of beneficial, unanticipated shocks to the debtor's economy. Such shocks may raise the sum of resources which the lender could expect the debtor to transfer, leading the lender to regret having allowed a large discount. Nevertheless, Cohen argued, the difference between a generalized buy-back and a one-off debt write-down is still central: in the former case, subsequent flows from debtor to creditor can be upgraded to take account of the beneficial shock, whereas when the stock of debt is written down once and for all, such a contingent adjustment would be impossible.
Members of the audience raised a number of questions about Cohen's proposal. One asked why debt write-offs did not benefit debtors by stimulating saving, but Cohen argued that debt was so high that reductions would have a negligible effect on savings ratios. He also disagreed that his scheme would cause distortions in commodity prices, arguing that the relative price of commodities and debt was irrelevant as lenders did not want to be repaid in goods. Finally, Cohen agreed that it was necessary for his scheme to be comprehensive across countries, even those that presently service their debt fully. He accepted that there were problems in treating those who over-borrowed in the 1970s the same as those who had borrowed within their means.