Fiscal Discipline: from Theory to Practice

International Monetary Fund. Western Hemisphere Dept.
Felipe Larraín, Luca Ricci, Klaus Schmidt-Hebbel, Hermann González, Metodij Hadzi-Vaskov, and Andrés Pérez
Published Date:
October 2019
  • ShareShare
Show Summary Details

Charles Wyplosz is Emeritus Professor of International Economics at the Graduate Institute in Geneva where he was director of the International Centre for Money and Banking Studies. He currently serves as policy director at the Center for Economic Policy Research (CEPR). A French national, Charles Wyplosz holds a degree in Engineering from Ecole Centrale, Paris, and a PhD in Economics from Harvard University.

Constraining budgetary decisions is needed for two main reasons. First, governments are simultaneously subject to intense demands for more spending and to reluctance to increase taxes, which leads to a deficit bias. Second, financial markets can quickly switch from trust to mistrust and abruptly cut lending to governments, leading to highly disruptive crises. Preventing a debt buildup requires providing governments with counter-incentives to tame the deficit bias and reduce the odds of sovereign debt crises.

Many lessons have been drawn from the growing number of experiments that aim at establishing fiscal discipline. All sorts of arrangements have been conducted. They typically include the adoption of a rule, sometimes backed by an independent fiscal council. The results have been mixed. A key reason is that the rules often lack a solid foundation, which makes them difficult to uphold when political pressure becomes intense.

Fiscal discipline is needed because every government is subject to a budget constraint. This constraint is not about deficits over one year or two, it is about the longer-term public debt. It is not even about the size of the debt in a particular year, it is about its evolution. A large debt that is set to decline is compatible with discipline while a debt that grows continuously is not. This principle carries a number of implications.

It explains why a number of rules are improperly designed. The most popular rule, setting numerical targets for the annual budget balance, focuses on the short-run. Not only annual budget outcomes have a very limited impact on the long-term evolution of the debt, but they also ignore that a key function of fiscal policy is to counteract disturbances that lead to expansions and recessions. In addition, a numerical target pins the rule on a number that is arbitrary and therefore hard to justify when the rule binds.

This observation has led many countries to target instead the structural balance. A structural balance target allows for the automatic stabilizers to operate, a clear improvement in principle. In practice however, it suffers from serious defects. It is difficult to compute the structural balance with any degree of precision. This leads to try and improve the methodology used for that purpose, but then the results of computations change. This is what happened three times in Chile (in 2004, 2005 and 2009). In two cases the revisions entailed revisions of some 2% of GDP, a number that is very large relative to the target. Changing the guidepost along the way hardly inspires confidence in the rule. In addition, the concept is complicated to explain and therefore most unlikely not to be trusted by policymakers and the public at large, seriously undermining the rule. Furthermore, there are cases when the automatic stabilizers are too small to stabilize the level of activity. This is why the numerical target has been changed twice in Chile in the wake of the global financial crisis of 2008. Once the target has been changed, it loses some credibility.

Another popular target is expenditures, or their annual growth rate. The main advantage of this rule is that it is under relative precise control by the government. Its fatal disadvantage is that it is not directly related to fiscal discipline. The adequacy of an expenditures target rests on the evolution of government revenues. Controlling one blade of the scissors is simply not enough. In addition, the size of public spending is fundamentally a political choice. Constraining this choice lacks economic and political justification, which undermines the rule’s legitimacy.

Faced with limits of numerical targets, the tendency has often been to add new targets to existing targets. A multi-target rule makes matters even worse. It does not improve the logic of each target. It is nearly impossible to bring coherence to the various targets, so that in any given year some may be satisfied but not others. Governments then can pick the target that is most convenient one year, and another target another year, without ever achieving discipline. Much the same can be said about escape clauses that list situations when the rule can be suspended. Even with the greatest imagination, it is impossible to anticipate all the situations that may arise, so escape clauses stand to be invoked for other reasons than intended, for good or bad reasons alike.

A good fiscal rule rests on a single target, not numerical. Theory says that the proper target is the evolution of the debt to GDP ratio in the long-term. It might seem that this is unrealistic, because long-term targets are impossible to assess and are too weak to act as a restraint over the short term, the policy horizon. This is not the case. The long-term debt depends on the whole path of budget balances and on assumptions about the interest and growth rates. Of course, no one can know what the interest and growth rates will be over decades. But we can make assumptions and even assess the likelihood of these assumptions. This is done by the IMF in its Debt Sustainability Analysis, by the US Congressional Budget Office and by various other agencies. The result can be shown as a fan chart that gives a good sense of where the debt is likely to be heading given the path of budget balances. Reverse-engineering this calculation, we can assess which paths of budget balances are compatible with a stable – or declining where it is too large to start with – debt to GDP ratio. This approach has a number of important advantages.

First, the stability of the long-term debt path is consistent with the concept of fiscal discipline (or solvency, or sustainability, or prudence). Second, it correctly inscribes forthcoming budget balances within a long-term commitment to discipline. Third, it makes it clear that the fiscal instrument can be used to stabilize incomes in the face of shocks, provided that deviations from the stable path are eventually corrected. Fourth, it is intuitive and therefore understandable outside narrow technical circles. Fifth, many budget balance paths are compatible with a long-term debt stability. Thus, the approach does not tie down a government to an arbitrary predetermined numerical target while requiring the government to make the case that its budgetary decisions are disciplined. Finally, it indicates what successive governments need to do, without unrealistically tying them down, thus allowing the public to detect unreasonable behavior.

The downside of this approach is that it rests on highly technical work, which involves making difficult assumptions and judgment calls. Complexity and technicality result in opacity, and opacity opens the door to manipulation. For this reason, the task cannot be entrusted to the government since the purpose of a fiscal rule is to provide the government with incentives to act in a way that may not always be politically appealing. The solution is to delegate the task to an independent and highly-skilled fiscal institution, which is required to be fully transparent about its work. This is done in the New Zealand and the Netherlands, for example. This is also what the US Congressional Budget Office does, but without a fiscal rule.

    Other Resources Citing This Publication