The IMF has addressed the moral hazard problem by imposing conditions onthe lending programs that it oVers, attempting to force states to adopt more
responsibleWscal policies. While initially some IMF members opposed the use of
such conditionality, arguing that the organization’s role was to provide funding as
needed, the major creditors (especially the United States) have insisted on impos-
ing conditions. The number and types of conditions has expanded substantially
over the years. Governments wishing to conclude a program with the IMF must
typically commit to reduce public spending, increase collection of taxes, liberalize
their international economic relations, and even improve other areas of govern-
ance. Of course, such conditions are not popular for the governments that must
accept them. Even if they are economically justiWed (a point that some would
dispute), there are occasions on which the major creditor states would prefer looser
conditions for purely political reasons. For example, it is widely understood that
the United States opposed the imposition of tough conditions on Russia in the
early 1990 s, wishing to assure Russia’s political stability. In addition, states that are
home to private creditors with substantial exposure in the crisis country are likely
to prefer looser conditions andXows of capital.
This basic strategic problem—potential beneWts from capitalXows, but a moral
hazard problem—has led many scholars to use a principal–agent framework to
study the IMF and, less extensively, the World Bank. In this framework, the
members of the IFIs, especially the major creditors, are treated as the principals
that use the IFI to implement their preferred policies. IFIs, as agents, have their own
interests, usually understood as technocratic economic interests. The question is
then the extent to which the IFIs can pursue their own agenda vs. responding to the
speciWc demands of their principals. As such, the ongoing tug-of-war between rules
and power describes the dynamics of the IFIs.
Some analysts, such as Strom Thacker, demonstrate that the IMF’s patterns of
lending respond to the geopolitical interests of the United States, its dominant
member (Thacker 1999 ). The ‘‘public choice’’ school has studied the IMF as a self-
interested organization attempting to assert itself in the face of constant political
demands from its powerful member states. This work, like Thacker’s, illustrates that
these states are often able to exert substantial inXuence over the IMF’s activities.
Thus, while the IMF is an agent with some autonomy, it has a hard time escaping its
political conWnes. Dreher and Vaubel ( 2004 ) apply this analysis to examine the
evolution of conditionality over time, asking about the content and number of
conditions imposed. They demonstrate that the IMF can usefully be studied as a
bureaucracy with its own internal rules and interests. In this sense, they posit that it
has more autonomy than others have recognized. They reason that an autonomous
IMF should impose stringent conditionality on states that have a poor record of
living up to past commitments, andWnd evidence to support this argument. Overall,
the evidence suggests that the IMF is an agent constrained by the political interests of
its principals, but one that is able to exert autonomy under certain conditions.
international economic institutions 667