Discretion: A Precondition for Democratic Accountability

Discretion means that policymakers are able to determine freely instruments at their disposal to serve a number of objectives under well-defined constraints. Hence, discretion would appear to be a precondition for holding any elected government accountable for the democratic mandate received by it from the electorate. But irrespective of the specific spending priorities and financing strategies reflecting that mandate, policymakers cannot persistently ignore the fiscal preconditions for macroeconomic stability. There are three aspects of this: first, fiscal policy should remain consistent with government solvency (public debt sustainability); second, public finances should be resilient in the face of unexpected shocks (contained fiscal risk); and third, the fiscal stance should contribute to macroeconomic stabilization (countercyclicality), which could occur in part by implementing measures that insure liquidity constrained individuals against adverse macroeconomic shocks.

In an ideal world, if the above preconditions were to be met, fiscal policy would never result in sustained increases in the public debt-to-GDP ratio, the budget would not be exposed to unmanageable implicit liabilities, and the fiscal stance would play a significant part in dampening the business cycle, or at least it would not aggravate the cycle. To the extent that voters clearly understand these desirable macroeconomic aspects of fiscal policy, a rational and democratically accountable government would have clear incentives not to deviate from such policies.

In practice, however, fiscal policy often turns out to be inconsistent with the requirements for macroeconomic stability. As noted in Chapter 1, the past three decades have witnessed a persistent tendency for countries to run large fiscal deficits, and a near universal tendency to procyclicality. This has led to sustained public debt increases in many countries, repeated debt crises in developing countries, mounting implicit liabilities worldwide, and often an aggravation of countries’ cyclical conditions.

Admittedly, debt financing and procyclical policies are at times optimal. Debt can finance valuable investment projects that pay for themselves in the long run so that solvency is not at risk. It can also prevent transitory expenditure variations to result in undesirably frequent modifications to the tax system. Likewise, unavoidable fiscal adjustments may have to take place in bad times. The problem lies in the very persistent nature of debt buildups and in repeated procyclical expansions. In the face of such evidence, the notion that there exist significant distortions in fiscal policy-making can hardly be rejected.

It is often tempting to attribute these unsatisfactory outcomes to discretion itself, so that suppressing it might appear acceptable. However, as the experience with monetary policy reforms shows, the underlying problem does not lie with discretion as such, but with the incentives shaping the behavior of those who exercise it. This would suggest that rather than remove discretion and put policy on automatic pilots, institutional reforms aimed at correcting incentives would be preferable. Indeed, in this respect, the analogy with monetary policy is instructive: central bank reforms have, in general, not eliminated discretion—official dollarization and currency boards remain exceptions. Instead, they have sought to create a framework, and to provide clear institutional guarantees that discretion would not be misused.

The effectiveness of an institutional framework in improving policy outcomes ultimately rests on how it affects the perceived (electoral or reputational) costs for policymakers to deviate from desirable policies. Before discussing available options for institutional fiscal reform, it is useful to review the key distortions underlying fiscal behavior. These suggest that fiscal policymakers may be subject to excessive complacency in the face of high deficits and rising debt—the so-called deficit bias—and that they find it difficult to resist fiscal expansions in good times, leading to systematic procyclicality. Of course, procyclicality may also be due to factors not directly related to policymakers’ incentives such as lags in decision making and in implementing policy measures after a shock has been identified.¹ But these do not raise deep political economy issues, as the reforms needed to allow for a swift response to changing macroeconomic conditions are generally likely to be of a procedural nature.²

The role of distorted incentives in explaining these phenomena is corroborated by peripheral issues such as frequent attempts to reduce fiscal transparency and undermine democratic accountability. Specifically, unduly optimistic projections are often used to hide ex ante the adverse effects of unrealistic political demands on the budget (Stein, 1994), whereas “creative accounting” produces the ex post illusion of more favorable outturns.

Possible Causes for Deficit Bias

At the most basic level, voters themselves may be unable to appreciate fully the macroeconomic features of optimal fiscal policy. Specifically, an apparently insatiable appetite for additional public goods or transfers may result in pressures to spend revenue windfalls in good times, which then may leave no option but to retrench in bad times, leading to procyclicality. Also, Ricardian equivalence notwithstanding, voters may not grasp fully the mechanics of the intertemporal budget constraint by which today’s deficits are inevitably linked to tomorrow’s taxes and noninterest spending capacity. This lack of understanding has two effects. First, a rational policymaker may find it useful to use fiscal expansions as a way to increase reelection chances, resulting in a political business cycle (Calmfors, 2005). Second, voters’ myopia and an incumbent’s willingness to stay in office may cause undue delays in much needed fiscal adjustments. Overall, deficits will tend to be higher, and the cyclical response of fiscal policy will be asymmetric, as discretionary policy will tend to undermine automatic stabilizers in good times only, further contributing to deficit bias.

The fact that voters’ preferences may be the root cause of deficit bias and procyclicality might appear potentially problematic for the design of fiscal frameworks aimed at enhancing fiscal discipline. First, it is unclear how such reforms would receive the support of the majority. Second, even if adopted, the credibility of a given fiscal framework would be in doubt, as respecting it could entail electoral costs, whereas flouting it could bring about gains, at least in the near term. However, increasing empirical evidence indicates that the situation is likely to be more hopeful, in that voters’ awareness of the government’s intertemporal budget constraint is generally greater than that of elected officials. In particular, a higher degree of direct democracy seems associated with better fiscal outcomes, suggesting that policymakers’ willingness to spend exceeds voters’ demand for public goods and services. The rest of this chapter focuses on distortions emanating from the policymaking process itself, which in principle can be addressed through institutional reform.

A key distortion underlying inadequate fiscal discipline arises from the tendency of governments to have shorter time horizons than voters. The reason for policymakers’ myopia lies in the electoral uncertainty inherent in the democratic process. As elected officials focus primarily on the consequences of their own (discretionary) actions while in office, their interest for future policies is lessened due to the risk of losing the next election. Myopia implies a relative neglect for the future tax hikes and primary expenditure cuts inevitably attached to present deficits. A rational government that balances the perceived marginal cost of deficits with the marginal gains will thus opt for deficits above those desired by voters (Alesina and Tabellini, 1990; and Rogoff, 1990). For similar reasons, fiscal adjustments tend to be delayed or backloaded, and revenue windfalls in good times are less likely to be saved, pointing once again to an asymmetric cyclical pattern in deficits.

In addition to myopia, discretionary policies can be “time inconsistent” when a government finds it desirable ex post to deviate from policies that were deemed optimal ex ante. One manifestation of time inconsistency is the tendency to try to alleviate the symptoms of structural problems through macroeconomic (monetary or fiscal) stimulus (Kydland and Prescott, 1977). While a low deficit is ex ante optimal, consumers understand that the corresponding expectation of low inflation (reflected in nominal wage contracts) would increase the effectiveness of a fiscal stimulus. Hence, even though there is broad agreement on the desirability of a low deficit policy, the latter would turn out not to be credible. While time inconsistency was a key argument behind the reform of monetary institutions (Rogoff, 1985), those very reforms may have aggravated the time-inconsistency problem of fiscal policy.

Another manifestation of time inconsistency is the difficulty for liquidity constrained governments (particularly in developing economies) not to spend revenue windfalls in good times. Although the lack of financing in bad times makes it optimal to save revenue windfalls in good times, the very realization of such windfalls triggers spending pressures that prove hard to resist. This, in turn, undermines the country’s repayment capacity and encourages financial markets to ration credit in bad times, perpetuating the liquidity constraint at the origin of the problem (Eichengreen, Hausmann, and von Hagen, 1999). Overall, time inconsistency results in excessive average deficits and debt accumulation. It can also be related to the procyclicality of fiscal policy in countries with only intermittent access to financial markets, which in turn rationalizes the procyclical nature of access to financing.

The political and distributive conflicts among different interest groups that result in “common pool” problems similar to the “tragedy of the commons” also give rise to distortions. Special interest groups and constituencies view government revenue as a “common pool” of resources open to competing demands. These groups have a basic tendency to use the available resources for specific distributive purposes without regard to the overall budgetary situation (Eichengreen, Hausmann, and von Hagen, 1999). Faced with such pressures, the government has an incentive to spend revenue windfalls and incur excessive deficits. Such fiscal expansion in good times directly contributes to procyclicality in the short run. But, once again, the cyclical asymmetry of such pressures implies that the windfalls are not used to improve the underlying position whereas the fiscal position worsens during bad times, leading to a rising trend in public debt.

In the particular case of monetary unions, the centralization of monetary policy can reduce individual countries’ incentives for fiscal discipline. Normally, the unpleasant prospect that excessive public debt may ultimately increase future inflation and interest rates is likely to impose some self-restraint on governments. In a monetary union, however, this effect is likely to be diluted, particularly for small countries, and could lead to excessive debt accumulation (Beetsma and Bovenberg, 1999). In addition, monetary unions can entail a moral hazard related to the greater likelihood of a bailout by other member states or by the common central bank.

These fiscal policy biases can, in turn, affect monetary policy. Left unchecked, excessive deficits and rapid debt accumulation could undermine the capacity of monetary institutions to fulfill their mandates. On the one hand, the pressure to raise the inflation tax may become hard to resist in the face of high public debt, resulting in fiscal dominance. On the other hand, to the extent that governments use fiscal policy to influence aggregate demand, conflicts between monetary and fiscal authorities could reduce the benefits of central bank independence (Debrun, 2000; and Dixit and Lambertini, 2003).

Inadequacy of Market Discipline

Financial markets could be expected to respond to the deficit bias and procyclicality in a way that promotes greater fiscal discipline. This could be the result of three effects. First, higher deficits could raise the level of interest rates as national savings decline, and inflation expectations rise if there is a perceived possibility of future debt monetization. Second, higher deficits could lead to a widening of credit spreads on public debt and potentially on the external borrowing of the economy due to a higher country risk premium even in the absence of an impact on the level of interest rates. Both these effects would tend to impose costs on policymakers: they would raise the budgetary cost of public borrowing and thus reduce the room for politically more attractive spending or require tax increases, and they would burden the economy at large with potential negative repercussions on growth and employment. Third, as debt rises, at some point the government could face borrowing constraints that would require politically painful fiscal adjustment and could be accompanied by a broader economic crisis.

The size of these effects is likely to depend particularly on the degree of economic openness and financial development, and the credibility of policies and institutions. Greater economic openness could, on the one hand, tend to weaken the effect of profligate fiscal policy on domestic interest rates both in the short run and the medium run. On the other hand, freer movement of capital could increase the penalty imposed by an increase in credit risk. Limited financial depth, in turn, would tend to magnify crowding out. And credit spreads are likely to react more harshly to fiscal profligacy for countries that defaulted in the past or whose policies and institutions have low credibility, as reflected by a history of high inflation or opaque public finances. These considerations suggest that market discipline is likely to be more stringent for developing than industrial countries. Moreover, market discipline is likely to be weakened in a monetary union. Similarly, it is possible that financial globalization could weaken the effect of fiscal policy on domestic interest rates (see Hauner and Kumar, 2006).

A large empirical literature suggests that markets are unlikely to effectively constrain a deficit bias. If anything, the market seems to penalize fiscal profligacy in a discontinuous fashion only at a late stage. For industrial countries, most studies of the determinants of real interest rates in industrial countries found effects that seem too small to impose a significant political cost on governments.³ Credit spreads and ratings do not seem to substantially penalize fiscal profligacy of industrial countries either (Balassone, Franco, and Zotteri, 2006). Less evidence is available on the determinants of interest rates in developing countries, but there is some indication that the effects of fiscal policy are stronger than in industrial countries, possibly due to some of the above considerations. However, the large literature on the determinants of emerging market country spreads and ratings yields conflicting results as to whether there is a substantial impact of fiscal variables. In any case, experience suggests that market discipline is certainly not a sufficient deterrent against unsustainable fiscal policies in developing countries either.

Improving Incentives Through Institutional Reform

As discussed above, the evidence of significant biases in fiscal policy points to persistent distortions in policymakers’ incentives. Moreover, the response of financial markets to weak fiscal performance appears insufficient to alleviate those distortions. Institutional reforms aimed at reshaping policymakers’ incentives could thus help ensure that fiscal policy choices remain consistent with the intertemporal budget constraint, and the maintenance of macroeconomic stabilization.

The challenge in designing fiscal institutions arises from the need to discourage the undesirable manifestations of fiscal discretion while retaining policymakers’ flexibility to respond to unexpected developments and to fulfill their democratic mandate. One practical difficulty in doing so is the thin line between ways to discourage the misuse of discretion and outright limitations to it. There has been an extensive literature rationalizing “fiscal frameworks” that could help meet this challenge. These frameworks can often help improve policy outcomes and stabilize expectations regarding fiscal behavior, reducing uncertainty and volatility, and have a positive impact on investment and growth.

An array of institutional arrangements can potentially improve fiscal outcomes, but they all have in common two essential components: (1) an explicit and transparent characterization of what the government views as a desirable (unbiased) fiscal policy, and (2) ways to enhance the (political or reputational) costs of deviations from unbiased policy. The choice among available options would reflect the nature of the underlying distortions to fiscal behavior, the extent of the resulting fiscal bias, and the broader political and institutional landscape of a country.

At one end of the institutional spectrum, one could consider explicit political commitments supported by strong mechanisms of democratic accountability. This is perhaps the most natural way in a democracy to contain distortions arising from the political decision-making process. By publicly exposing broken promises, errors, or biases, the framework simply seeks to maximize the costs of breaching commitments, without formally impinging on discretion. Medium-term expenditure frameworks and transparency requirements will typically be part of such fiscal frameworks.

However, formal commitment as such may be seen as insufficient, and there could be a need to directly increase the costs to policymakers of biased policies. The most common channel for doing so is to define ex ante explicit boundaries of acceptable fiscal outcomes. Fiscal rules—on deficits, debts, expenditure, or revenues—are widely used. Yet evidence of their effectiveness is mixed, reflecting mainly how governments perceive the costs of breaching them. Fiscal rules can be embedded in the more comprehensive setup of a fiscal responsibility law, which also provides for transparency requirements, and accountability mechanisms aimed at further increasing the political costs of breaching the rule.

One important issue in the ongoing debate on fiscal rules is whether they are conceived as instruments to curtail discretion (hard rules), or as mere guideposts to enhance the public debate on fiscal policy (soft rules). It is common in the literature to retain the hard-rule interpretation. In that sense, a rules-based framework seeks to restrain flexibility, and is bound to prevent an appropriate fiscal response in at least some circumstance. Although the ideal would be to adopt rules that allow for the highest credibility while limiting reduction in flexibility, this is far from straightforward. Simple and transparent rules entailing automatic sanctions tend to bring policy credibility only at potentially high cost in terms of forgone flexibility.

It is for this reason that other institutional reforms are favored by some because they could effectively contribute to reducing fiscal policy biases without excluding flexible policy response when needed. They include nonpartisan fiscal agencies mandated to monitor and assess fiscal policy, and “procedural” fiscal responsibility laws establishing strict reporting and transparency requirements. These institutions primarily aim at maximizing the (reputational or electoral) costs of deviations from desirable policies. At the same time, they preserve policy flexibility when it is needed because discretion itself is not curtailed. Given that fiscal rules have often been undermined by their rigidity under unfavorable circumstances, an improvement in flexibility as provided by institutional reform might also be more effective in raising policy credibility.

As emphasized above, the effectiveness of these mechanisms ultimately rests on their ability to discourage the misuse of discretion. Even hard rules can only succeed if the political costs of removing them, changing them, or breaching them openly would be unacceptable. Their self-imposed nature makes it difficult to believe that they could as such limit an elected official from exercising discretion.

Fiscal rules and institutional innovations could potentially complement each other. This can be seen considering the example of the fiscal agencies discussed in Chapter 6. On the one hand, these institutions can strengthen the enforcement of rules by increasing the political cost of breaching them, but also by making their implementation more flexible under exceptional circumstances that may merit a deviation from the rules in the short term—without undermining their credibility in the long term: for example, such an institution could determine when exceptions from rules would be warranted, similar to the role played by the constitutional court in Germany in implementing its Golden Rule. On the other hand, the existence of rules established through the democratic process can provide clear benchmarks against which institutions such as fiscal agencies could assess the performance of fiscal policy, such as envisaged in the proposals by the Committee on Stabilization Policy (2002) of Sweden and the European Commission (2004). Similarly, the effect of rules could be strengthened by giving the judiciary a role in their implementation, such as envisaged in the proposals by De Haan, Berger, and Jansen (2004) and Inman and Rubinfeld (1996).