Michael Mussa has written a convincing and useful chapter on the questionable logic surrounding parts of the PSI debate. His two principal conclusions can be summarised as follows. First, the problems that are commonly associated with moral hazard and "excessive" IMF lending are greatly exaggerated. Second, two different, and perhaps more pernicious, forms of moral hazard can nevertheless arise whenever the IMF deviates from the principles embodied in its Articles of Agreement.
Mussa argues that large IMF programmes have not encouraged reckless behaviour on the part of creditors or debtors. The sizable costs that each group has borne in the recent string of crises, coupled with the limited relief that they can reasonably expect from any official financing in the future, both work against this popular but flawed perception. Problems can nevertheless arise from two other related sources. These occur, in the first instance, whenever the IMF is encouraged to lend for geopolitical rather than economic reasons. Another, equally troubling situation arises when countries with evidently unsustainable debts are given emergency assistance. 2 Although policymakers in the affected countries should, in principle, act in the best interests of the people they represent, this does not always occur in practice. Instead of initiating a prompt and orderly restructuring of their country's debts, policymakers often delay the day of reckoning as long as possible, "gambling for resurrection".
The major policy implication that Mussa draws from all this is that "greater involvement of private sector creditors in absorbing losses from such crises - beyond the substantial amount which already occurs - is not needed to correct the popularly perceived problem of moral hazard from so-called IMF bail-outs". Greater emphasis should instead be given to following existing guidelines and lending official money only when appropriate. Since I agree with most of what Mussa has said, there is little that I can offer by way of useful criticism or comment concerning his main arguments. Moral hazard, at least as commonly perceived, does not seem to provide a very convincing rationale for enhanced PSI. Rather than dis-