Creditor Moral Hazard
Creditor Moral Hazard
The problem of creditor moral hazard arises from the track-record of IMF lending, which creates an expectation among creditors that the Fund will invariably bailout troubled sovereign debtors – and by extension their creditors – in order to avoid a costly default. With this safety net in place, private creditors engage in excessive and risky lending to sovereigns, allowing for the unsustainable buildup of sovereign debt and contributing to the bankruptcy crises that follow. Mitigating moral hazard “requires a clear and credible threat of losses through restructuring when lending has been excessive” (Schadler 2012: 13). But the IMF has a hard time making credible commitments of this sort because of the so-called ‘dynamic inconsistency problem’ – that is, the potential for actions that may be optimal before the fact to be sub-optimal after the fact if some condition is not satisfied. As James Haley (2013) explains:
In the case of sovereign bankruptcy, the commitment not to provide IMF (or other official sector) resources is entirely sensible in that it creates incentives for sovereign borrowers and their private creditors to come to a timely restructuring. But if such a restructuring is not forthcoming, and the result is a financial crisis that has external effects on global markets, the commitment not to provide financing may look decidedly sub-optimal, particularly for an institution whose mandate is global financial stability.
The solution to this problem is to put in place formal restructuring procedures that stipulate the conditions under which restructurings will take place, and that reduce the costs of pre-emptive restructurings enough to be credible (Schadler 2012).
While many see it as a central concern, empirical evidence on the existence of creditor moral hazard is inconclusive. Some scholars find robust evidence in support of the moral hazard thesis, while others find no empirical foundation for this argument.[1] The evidentiary ambiguity surrounding this question is at least partly a result of the difficulty of accurately identifying or measuring cases of moral hazard.
[1]Haldane and Scheibe (2004), Dell’ Ariccia et al. (2002), and McBrady and Seasholes (2000) find evidence of moral hazard. Lane and Phillips (2000), Jeanne and Zettelmeyer (2001), and Kamin (2002) find no evidence of moral hazard.