1. Introduction
This paper aims to test the main implication of the theoretical model presented
in Marchesi and Thomas (1999), namely that agreement to follow a programme
could be a signal of an indebted country’s willingness and ability to successfully
reform (and use any new money provided for investment rather than consumption
purposes), which is thus rewarded with a debt relief.
Throughout the 1980s and 1990s, many developing countries have struggled
to repay large amounts of their external debts to both commercial banks and in-
dustrial countries’ governments. In the early ’80s Paris Club creditors provided
reschedulings for low-income countries on non-concessional terms and on market-
related interest rates.1 In the late eighties (1989-’94) the Brady deals addressed
commercial bank lending to government debtors (generally middle-income coun-
tries) and involved a combination ofan IMF agreement and debt and debt-service
reduction and rescheduling from commercial banks.2
In the same period, Paris Club creditors agreed to provide low-income coun-
tries with concessional reschedulings, conditional on the adoption of IMF adjust-
ment programme, under the Toronto (1988), Trinidad (1990), and Naples terms
(1994).3 Since the onset of the debt crisis while the debt situation of middle-
1 Paris Club creditors are o^ial (bilateral) creditors of government debt, while London Club
creditors are commercia l creditors of private i ntern at iona l debt.
2On 10 March 1988, US Treasury Secretary N.J. Brady an noun ced a new plan for interna-
tional debt that wou ld shift the strategy from co-ordinated lending to the reduction, or partial
forgiveness, of existing debt. The Brady Plan was premised on a vo lu ntary, ma rket oriented
approach to debt reducti on.
3They recogn ised that mostof low-incom e countries requ ired an actua l redu cti on of their level
of debt, more than repeated reschedulings on “standard terms”. A “concessiona l” reschedu ling
implies a redu cti on of the net present value of the rescheduled amount.