tives for banks to undertake actions that save taxpayers' resources, rather than their own, are reduced.
Mistrust between creditors and debtors also constrains debt reduction ( Claessens and Diwan 1989; Cohen 1989). Because of the intertemporal nature of the credit relation, efficient solutions have to involve promises by both parties to undertake future actions. The banks cannot credibly promise to debtors that they would extract a small share of future resources if the debtor imposes austerity and increases investments, and the debtor cannot credibly promise to invest new loans if they were forthcoming. This mutual lack of confidence is akin to the bad equilibrium in a prisoner's dilemma. Coordinated action can increase the available resources that will have to be shared, but it will not occur in the absence of a commitment mechanism. While such a mechanism is available to debtors in the form of the multilateral institution's conditionality, complex financial contracts would be needed for the banks to commit to low resource extraction in bad times without creating serious moral hazard ( Krugman 1989; Cohen 1989).
The debt-overhang argument has proven to be extremely important in policy debate. However, it should not obstruct the fact that for the HICs, bargaining remains the main determinant of capital flows. More empirical research is needed to make the overhang argument operational, and public policy should try to alleviate the constraints to debt reduction.
I would like to thank S. Claesssens for his comments. The views expressed in this discussion should not be attributed to the World Bank.
Bouchet Michel, and Jonathan Hay. ( 1989). "The Rise of the Market Based Menu Approach and Its Limitations." In Dealing with the Debt Crisis, ed. I. Husain and I. Diwan. Washington, D.C.: World Bank.
Bulow Jeremy, and Kenneth Rogoff. ( 1988). Multilateral Developing-Country Debt Rescheduling Negotiations. IMF Working Paper 88/35. Washington, D.C.: International Monetary Fund.