FISCAL RULES, FISCAL INSTITUTIONS AND FISCAL PERFORMANCE
265
• Designing electoral rules promoting political accountability and competition
and increasing the politicians’ incentives to deliver the policies voters most
prefer;
• Designing the decision-making processes over public finances that induce
policy makers to recognise more fully the marginal social benefits and costs
of their policies.
In this paper, we review and discuss recent research into these three
approaches and their relative success in mitigating the principal agent and
common pool problems. In Section II, we discuss fiscal rules as instruments to
limit the principal agent problem and the common pool problem. In Section III,
we consider the role of electoral institutions in shaping and limiting the
principal agent problem. In Section IV, we look at the institutional aspects of
decision-making processes regarding public finances. In Section V, we
conclude with some remarks on institutional reform.
II FISCAL RULES
Most state governments in the US and provincial governments in Canada
are subject to balanced budget requirements or debt ceilings, and many state
constitutions in the US include numerical expenditure limits or require public
referenda for raising tax rates. History suggests that these controls were often
imposed by taxpayers, who were angry about the spending profligacy of their
elected representatives (Eichengreen and von Hagen, 1996; Millar 1997).
There is a fair amount of variation in the scope and strictness of these
constraints, documented by ACIR (1987) and Strauch (1998). Von Hagen and
Eichengreen (1996) and Stein et al. (1999) document fiscal rules imposed on
sub-national governments in other countries. Subjecting policy makers to such
ex ante controls seems the most straightforward approach to controlling their
behavior and they seem attractive for simplicity and transparency.
But how successful are such constraints? Empirical evidence for US state
governments suggests that they do limit the size of the annual deficit in the
current budget and reduce the government debt ratio, if “full faith and credit”
debt is considered (Strauch 1998; Eichengreen 1990). But if other types of debt
instruments including the debt of off-budget entities are considered,
numerical constraints have no effect on public debt ratios (von Hagen, 1991).
This indicates that governments subject to stringent numerical deficit
constraints tend to substitute debt instruments not covered by the legal rule
for full faith and credit debt. Kiewiet and Szakalay (1996) find another
substitution effect, namely that state governments subject to more restrictive