THEORETICAL PERSPECTIVES

In federal-type systems, the central government must find ways to restrain unsustainable borrowing by states, borrowing which, if left unchecked, can lead to disastrous spillover effects on neighboring states and ultimately to demands for federal bailouts. Markets alone fail to adequately discipline sovereigns. If they did, as Bayoumi, Goldstein, and Woglom (Reference Bayoumi, Goldstein and Woglom1995, 1046) explain, then yields on sovereign debt would “rise smoothly at an increasing rate with the level of borrowing, thereby providing the borrower with an incentive to restrain excessive borrowing.” But in fact yields often do not rise smoothly and excessive borrowing is not effectively restrained. Rather, sovereigns often build up substantial debt at low borrowing costs, and then face some type of fiscal crisis which prompts market panic, a rapid spike in borrowing costs to unsustainable levels. Instead of playing a helpful role deterring the accumulation of excessive debt, markets often react too late and then so forcefully that they help drive states to default.

To make up for the shortcomings of pure market discipline, many scholars and policy makers have called for the imposition of binding fiscal rules, including balanced budget rules. With the imposition of such rules, law and courts are expected to force politicians to engage in responsible fiscal policy. One approach to examining the effects of formal fiscal rules is to analyze the outcome of public finances. Focusing on the institutional variation across American states, Bohn and Inman (Reference Bohn and Inman1996) find that strict balanced budget rules have a greater deficit-reducing effect than more lenient rules. Furthermore, their results suggest that retrospective rules such as the no-carryover rule are more effective in lowering the likelihood of a state running a deficit in a given year than prospective rules. The benefits of strict fiscal rules are not restricted to lower deficits, however. A number of studies find that states with strict fiscal rules tend to issue less debt (Bunche 1991; Eichengreen Reference Eichengreen1992; Kiewiet and Szakaly Reference Kiewiet and Szakaly1996; von Hagen Reference von Hagen1991), have lower expenditures (Rueben 1996), and are quicker to restore fiscal balance following an economic shock (Poterba Reference Poterba1994).

Similar results emerge from outside the United States. In a sample of Latin American countries, Alesina et al. (Reference Alesina, Hausmann, Hommes and Stein1999) find that strict fiscal rules may contribute to lower fiscal deficits. Feld et al. (2011) find that the existence and strength of fiscal rules in Swiss cantons are associated with lower bond risk premia of bonds. Finally, cross-national studies on E.U. member states find that fiscal rules appear to be effective at lowering public sector debt levels and deficits (Debrun et al. Reference Debrun, Moulin, Turrini, Ayuso-i-Casals and Kumar2008; Hallerberg and Wolff Reference Hallerberg and Wolff2006).

At the same time however, other studies question the effectiveness of fiscal rules. Some argue that fiscally responsible countries do not need rules and fiscally irresponsible countries may not achieve fiscal discipline even with rules (Kennedy and Robbins Reference Kennedy and Robbins2001). If so, then fiscal prudence is less a function of fiscal rules than country-specific characteristics, such as political institutions. Briffault (Reference Briffault1996) charges that in the U.S. fiscal rules—state constitutional debt provisions in particular—are inconsistent, vague, and routinely circumvented by creative financing devices.Footnote ⁵ The International Monetary Fund (IMF) finds no significant effect of fiscal rules on bond spreads of 22 OECD countries (IMF 2009). In the E.U., Iara and Wolff (Reference Iara and Wolff2010) examine the relationship between numerical fiscal rules and government bond spreads and find that fiscal rules exert only a weak effect and that any impact occurs only in times of high risk and market stress. For reasons we explain below (see footnote 8), numerous studies suggest that the fiscal rules in place in German states have proven relatively ineffective in restraining deficits.

While recognizing that the literature remains somewhat ambiguous, it seems fair to conclude that the preponderance of studies on balanced budget rules suggest that they at least have some positive effect on restraining sovereign debt and lowering sovereign borrowing costs. But how do these laws work given that they are rarely, if ever, actually enforced in court?Footnote ⁶ Most of the literature has simply ignored this question, arguing that balanced budget rules work better where they are more “strict,” “stringent,” or “strong” but never questioning why these rules might be obeyed. The only explicit, theoretically grounded effort to address this question suggests that balanced budget rules work through a form of legal deterrence. Bohn and Inman (Reference Bohn and Inman1996) spell out this logic in their seminal article on balanced budget rules in U.S. states:

We find this causal explanation of how balanced budget rules influence government behavior unconvincing. First, Bohn and Inman’s (Reference Bohn and Inman1996) assertion that, “Perhaps the last place the state legislature and governor want their budget decisions made is in the state court,” is not compelling. The prospect of the judiciary making budget decisions is not necessarily anathema to law makers. In fiscal policy, as in many sensitive areas of policy making, law makers may be willing to punt sensitive questions to the courts (Frymer Reference Frymer2003; Lovell Reference Lovell2003; Whittington Reference Whittington2005). As Yoo and Gaziano (Reference Yoo and Gaziano2011) put it, “If legislators know that courts will be lurking in the background, they may simply abdicate their own responsibilities to comply with a balanced-budget requirement.” Similar dynamics seem to be at work in Germany, a country where there are routine violations of state balanced budget rules and where there have been a number of legal actions seeking to enforce them—with apparently little effect on state lawmakers.Footnote ⁸

Second, the experience of state courts enforcing fiscal rules in the US federal system suggests that they are inclined to show political branches considerable deference in this field—a posture which would hardly strike fear into the hearts of lawmakers. Though there are few cases of litigation involving balanced budget requirements, there is an extensive case law concerning other fiscal rules. As Briffault (Reference Briffault2003, p. 939–940) explains, in adjudicating cases involving such rules state courts have treated fiscal limits not as issues of fundamental rights to be subject to strict scrutiny by the courts, but rather as matters of public policy best left to legislatures. The New Jersey Supreme Court summarized this accommodating judicial attitude in a 1972 ruling explaining, “the modern science of government has found a method of avoiding [the debt] clause, and the courts have approved it.”Footnote ⁹ Moreover, as we discuss below in reference to a recent ruling of the Portuguese Constitutional Court, when courts are asked to balance the enforcement fiscal rules against the protection of other constitutionally guaranteed norms, they may allow the latter to take precedence over the former.

If lawmakers have less to fear from the prospect of judicial enforcement of balanced budget rules than the existing literature has theorized, how could it be that these rules nevertheless reduce state government deficits and state government borrowing costs?

A distinctive line of research on balanced budget rules developed by Goldstein and Woglom (Reference Goldstein, Woglom, Canzoneri, Grilli and Masson1992), Bayoumi, Goldstein, and Woglom (Reference Bayoumi, Goldstein and Woglom1995), Poterba and Rueben (Reference Poterba and Rueben1997), and Lowry and Alt (Reference Lowry and Alt2001) suggests that it is the bond markets rather than the courts that discipline states who violate such fiscal rules. Building on this line of research, we argue that balanced budget rules can serve as a coordinating device for bond markets. These laws provide a focal point, helping bond investors to coordinate on when to punish states for running excessive deficits by demanding higher yields. The dissuasive effect of these rules comes not from the expectation that a court will enforce them strictly, taking control of fiscal policy away from elected politicians. Rather, where these rules exist, politicians recognize that any indication that they are violating the rule is likely to be met with a punishment imposed by bond markets (i.e., a spike in borrowing costs). This decentralized punishment would be felt before and would be far more painful than any penalty that might eventually be meted out by a court. To the extent that states obey balanced rules, they do so not due to fear of centralized judicial enforcement, but due to fear of the bond markets.

Though the scholars mentioned above have emphasized the role bond markets play in enforcing fiscal rules, and though some of them (Lowry and Alt Reference Lowry and Alt2001) have explicitly suggested that balanced budget rules can play a role as a coordinating device, our argument differs from theirs in several ways. Lowry and Alt (Reference Lowry and Alt2001) suggest that investors lack adequate information on the intentions of government officials and that a particular form of balanced budget rule can help investors to “extract information from noisy signals” and coordinate their beliefs about how governments will react to one-time deficits (Reference Lowry and Alt2001, 52).

We suggest a different understanding of the coordinating role that balanced budget rules can play. Balanced budget laws help signal individual buyers the point at which other bond buyers are likely to withhold further financing; these rules thus create a focal point that can help investors coordinate on when to punish sovereigns by demanding far higher rates. In most cases, the fear of this coordinated market punishment is enough to restrain sovereigns from violating the balanced budget rules. On rare occasions where the rules are violated, the imposition of market discipline may quickly pressure sovereigns to rein in their deficits.

Scholars of political economy typically regard law as a sanctioning device, its impact on behavior a function of how strict and well-enforced it is. Thus in most of the research on the impact of fiscal rules, scholars focus on the strictness of the rules under the implicit assumption that these rules work under the threat of legal enforcement. However, as McAdams (2000) points out, law can affect behavior “independent of the sanctions it threatens to impose, that law works by what it says in addition to what it does” (emphasis added). Law can help actors overcome coordination problems by acting as a focal point, which helps provide information and expectations on how others will behave; law is expressive.

Given its expressive function, law can help participants coordinate punishment against a violator even in the absence of a formal enforcement mechanism (Hadfield and Weingast Reference Hadfield and Weingast2011, Reference Hadfield and Weingast2012, Reference Hadfield and Weingast2013). Hadfield and Weingast (ibid.) extend McAdams (2000) and others’ arguments by focusing on the characteristics that such a law must have. They argue that “an important role of legal process, therefore, is to make rulings that reduce ambiguity”. It is not sufficient that a law is common knowledge, it must also be clear. While other analyses examine how law, by acting as a focal point, can facilitate equilibrium selection, Hadfield and Weingast (Reference Hadfield and Weingast2011) contend that a coordination equilibrium exists only if players prefer the payoffs from coordinating successfully to those without coordination.

Following Hadfield and Weingast (Reference Hadfield and Weingast2011), our model assumes that individuals often hold different views, which makes coordination challenging and decentralized punishment difficult. We also make a similar assumption that each individual’s information is private and unobservable by others. Unlike Hadfield and Weingast (Reference Hadfield and Weingast2011) however, we make two modifications. First, the players in our model move simultaneously and are unable to observe the actions of other players prior to their decisions, as opposed to the sequential nature of Hadfield and Weingast’s (Reference Hadfield and Weingast2011) model, where players are willing to reveal information by taking costly actions if the benefits from doing so are sufficiently large. Second, we analyze explicitly how the clarity of a law—which Hadfield and Weingast (Reference Hadfield and Weingast2011) stress as one of the key features of a coordinating device—affects its ability to facilitate coordination.

To do so, we draw on recent advances in game theory—the theory of global games, pioneered by Carlsson and van Damme (Reference Carlsson and van Damme1993) and brought to the forefront of modern game theory by Morris and Shin (Reference Morris and Shin1998). Coordination games give rise to multiple equilibria. The problem of equilibrium selection in such games is due to the assumption that all relevant features of the game are common knowledge, or known to every player.Footnote ¹⁰ A global game removes the common knowledge of payoffs; the game becomes one of incomplete information. Instead, players’ payoffs are a function of some unknown “state” variable, a feature of the environment that affects the payoffs of all players. In our model, the state variable is a government’s fiscal situation.Footnote ¹¹ In addition, players’ payoffs depend on the actions of other players in a particular way—the greater the number of players taking the same action, the higher the payoff to each individual player; players’ actions are strategic complements.

In brief, our model shows that the clearer a balanced budget rule, the more effective it is as a coordinating device. Only when a rule is sufficiently clear can it serve as an effective focal point to facilitate decentralized enforcement.Footnote ¹² This implies that balanced budget rules can be effective in enabling states to borrow at sustainable rates while simultaneously preventing them from building up unsustainable debts—but that their effectiveness depends on their clarity and not on the strictness of their legal enforcement mechanisms.

THE MODEL

There is a continuum of buyers in a bond market indexed by i ∈ [0, 1]. Each risk-neutral buyer chooses some amount of bonds, $b_{i} \in \mathbb {R^+}$ , to purchase from a government. The simultaneous decisions of buyers form the aggregate level of financing the government receives, $\mathcal {B}$ . We assume that each buyer is “small” in the sense that an individual buyer’s stake is negligible as a proportion of the whole.Footnote ¹³

Two components enter into the decision-making process of each buyer. First, each buyer cares about the government’s fiscal situation, θ. A buyer would not want to purchase bonds from a government that is running unsustainable fiscal policies and may be potentially insolvent. Conversely, a government in good fiscal health makes for a sound investment. This uncertainty over the government’s fiscal situation is known as fundamental uncertainty. Second, each buyer also cares about other buyers’ beliefs and their decisions to finance the government’s debt, an uncertainty referred to as strategic uncertainty. Players care about one another’s beliefs and decisions about financing a governments debt, because they recognize that the sustainability of the government’s debt is itself a product of the bond markets’ willingness to finance that debt at reasonable rates. In other words, rational behavior for the individual bond buyer is not simply based on an assessment of economic fundamentals, but on their beliefs about the likely actions of other bond buyers.Footnote ¹⁴

Let π denote the weight that each buyer places on other buyers’ actions. That is, 0 < π < 1 captures the extent of complementarity. Each buyer’s concern for choosing the right value of financing given the government’s fiscal situation is thus 1 − π. We can then write each buyer’s utility as follows:Footnote ¹⁵

\begin{eqnarray*} u_{i} &=& \Big [\underbrace{(1 - \pi )\theta }_{\text{concern for government's fiscal situation}}\\ &&+\, \underbrace{\pi \mathcal {B}}_{\text{concern for other buyers' actions}}\Big ]b_{i}. \end{eqnarray*}

The optimal strategy for buyer i is given by the first-order condition:

\begin{eqnarray*} b_{i}^* = (1-\pi )E_i(\theta ) + \pi E_i(\mathcal {B}),\vspace*{-4pt} \end{eqnarray*}

where $\mathcal {B} = \int _0^1 b_{i}\ di$ , the average action of the mass of buyers, and E_i (·) is the expectation operator for i.Footnote ¹⁶

The government’s true fiscal situation θ is unknown to the buyers. As such, each buyer relies on her research to obtain information and form beliefs about θ. Since buyers’ prior experiences are different, they do not share the same information.Footnote ¹⁷ Let this private information be represented by

\begin{eqnarray*} x_{i} = \theta + \epsilon _{i}, \epsilon _{i} \sim N(0, \sigma _{x}^2),\vspace*{-4pt} \end{eqnarray*}

where ε _i represents buyer i’s idiosyncratic uncertainty over θ, is independent of θ, independent and identically distributed across buyers with mean zero and variance σ² _x over the standard normal. x_i is a private signal in the sense that it is not observable by others and only known to buyer i.

Suppose the government announces a balanced budget rule that is observable to all buyers; this rule is common knowledge. Let this public signal be denoted by

\begin{eqnarray*} y = \theta + \tau , \tau \sim N(0, \sigma _{y}^2), \end{eqnarray*}

where τ is independent of θ, independent and identically distributed over all buyers with mean zero and variance σ² _y drawn from the standard normal. The distribution of τ is referred to as the common prior.

Whereas buyers have different priors in the absence of a balanced budget rule, the buyers can now update their heterogeneous priors using the public signal and thus share a common prior. That is, each buyer now receives a common signal about the government’s fiscal situation, which influences their expectations about other buyers’ likely actions. The balanced budget rule serves two purposes: (1) it conveys information about fundamentals and (2) provides a focal point which signals individual buyers the point at which other bond buyers are likely to withhold further financing.

With the common prior, buyer i now forms her posterior belief by updating her private signal with the public signal. For ease of exposition, let

\begin{eqnarray*} \alpha &=& \frac{1}{\sigma _y^2} \nonumber \\ \beta &=& \frac{1}{\sigma _x^2} \end{eqnarray*}

Buyer i’s expected value of θ is then

\begin{eqnarray*} E_i(\theta | x_{i}, y) = \frac{\alpha y + \beta x_i}{\alpha + \beta }. \end{eqnarray*}

Definition 1. An equilibrium of the game consists of a set of beliefs for buyer i and a set of strategies $b_i^*:\mathbb {R}^2 \rightarrow \mathbb {R}$ such that for all (x_i , y),

(1) \begin{equation} u_i(b_i|x_i, y) = \arg \max _{b_i} (1-\pi )E_i(\theta ) + \pi E_i(\mathcal {B}). \end{equation}

Definition 2. A linear equilibrium is any strategy that satisfies 1 and is linear in x_i and y.

We solve for a linear equilibrium of this game and show that not only does such an equilibrium exist, it is unique.

Proposition 1. There exists a linear equilibrium in the game, b* _i = ϕx_i + (1 − ϕ)y, where

\begin{eqnarray*} \phi &=& \frac{\beta (1 - \pi )}{\alpha + \beta (1 - \pi )}, \\ \alpha &=& \frac{1}{\sigma _y^2}, \\ \beta &=& \frac{1}{\sigma _x^2}. \end{eqnarray*}

See Technical Appendix for proof.

Proposition 2. (Morris and Shin 2002) The equilibrium identified in proposition 1 is the unique equilibrium of the game.

Given the equilibrium strategy of the buyer, we consider now how buyer i’s optimal decision, b* _i , changes with respect to σ _x and σ _y , the clarity of buyer i’s private signal and the government’s balanced budget rule respectively. Recall that

\begin{eqnarray*} E_i(\theta | x_{i}, y) = \frac{\alpha y + \beta x_i}{\alpha + \beta }. \end{eqnarray*}

For a given σ _x , a reduction in σ _y —which corresponds to a clearer signal—leads buyer i to place greater weight on the public signal relative to her private signal and consequently improves coordination. Where a balanced budget rule is sufficiently clear, violations of that rule are more likely to trigger the bond markets to withhold financing. Knowing that violations of clear balanced budget rules would elicit a decentralized punishment by bond markets, governments will avoid violating those rules in the first place. On the contrary, a vague signal, i.e., σ _y is large, leads buyers to discount the information contained in the (noisy) public signal and instead place greater weight on their private signals, thereby making coordination more difficult. In other words, where a balanced budget rule is not sufficiently clear, violations of that rule will be more difficult to observe and will not provide bond markets with a clear focal point. Knowing that violations of a vague rule are unlikely to elicit a decentralized punishment by bond markets, governments have little incentive to respect the rule.

In sum, the model illustrates that a balanced budget rule can act as a focal point for the coordination of decentralized punishment by bond markets. More importantly, it shows that these rules need not be backed by the threat of legal enforcement to be effective; instead they can be backed by market deterrence. At the same time, focal points are not created equal. The extent to which a balanced budget rule can facilitate coordination is a function of its clarity. In particular, changes in the degree of clarity between private and public signals can affect the extent of coordination and thus the effectiveness of the rule. The clearer a balanced budget rule, the better it is as a focal point and, therefore, the more effective it is.

With this in mind, we specify the following hypotheses:

H1. The clearer a balanced budget rule, the greater its effectiveness.

H2. If a balanced budget rule is unclear, increasing the strictness of legal enforcement will not increase its effectiveness.

EMPIRICAL APPLICATIONS: THE U.S. AND THE E.U.

We begin the empirical assessment of our arguments with an analysis of data on the impact of the clarity of balanced budget rules in the U.S. Our quantitative analysis supports the hypotheses developed in the previous section, namely that the effectiveness of a balanced budget rule increases with its clarity (H1) and that if a balanced budget rule lacks sufficient clarity, increasing the strictness will not increase the rules effectiveness (H2). Moreover, in a brief qualitative analysis, we show that what occurred in one of the rare instances when a state government blatantly violated its balanced budget rules is more consistent with our “law as focal point” causal argument than with existing explanations. Next, we turn to the E.U. to examine the likely consequences of the 2012 Fiscal Compact Treaty, which seeks to reform eurozone governance in part by requiring all signatories to incorporate structural balanced budget rules into domestic law. While there is a rich literature on the effects of fiscal rules and fiscal institutions in E.U. member countries (Hallerberg and Wolff Reference Hallerberg and Wolff2006), balanced budget rules at the national level are a recent phenomenon in most member states, the Fiscal Compact Treaty only took effect from January 2013, and a number of signatories have yet to ratify and/or implement it. As such, it is far too early to conduct any sort of systematic quantitative analysis of the impact of the Fiscal Compact Treaty’s balanced budget requirements. Therefore, we offer a more qualitative, prospective analysis, assessing the empirical implications of our argument in the E.U. We suggest that the Fiscal Compact Treaty’s balanced budget rules lack the clarity needed to provide a focal point that could coordinate decentralized punishment by bond markets of excessive sovereign borrowing. Because the treaty focuses on strengthening the legal enforceability of balanced budget rules but leaves the rules themselves extremely vague, it is unlikely to prove effective.

Independent Variables

Rule Clarity

To construct our rule clarity index, we use the following dichotomous items from Alt, Lassen, and Rose’s (2006) fiscal transparency index (items 1, 3, 6, 8, and 9) of the American states:

1. 1. Is the budget reported on a Generally Accepted Accounting Principles (GAPP) basis?

2. 2. Is the frequency of the budget cycle annual?

3. 3. Are all appropriations included in a single bill?

4. 4. Is the legislature prohibited from passing open-ended appropriations?

5. 5. Does the budget require published performance measures?

Each question is scored 1 if the answer is yes, and 0 otherwise. The final index ranges from 0 to 5, where higher values indicate greater clarity.Footnote ²⁰

Rule Strictness

Like most studies of state balanced budget rules, we use the ACIR index of rule strictness as our indicator of the strictness of legal enforcement.Footnote ²¹ This measure ranges from 0 (states with lax requirements) to 10 (states with stringent requirements). Vermont is the only state that lacks a formal balanced budget rule.

Controls

We include a set of economic control variables at the state level: the state-wide unemployment rate (from the U.S. Bureau of Labor Statistics), per capita income (from the U.S. Bureau of Economic Analysis), general revenue, and debt outstanding (both from the U.S. Census Bureau). Unemployment, as well as higher levels of outstanding debt, may hinder a state’s ability to service its debt and negatively impact its credit rating. Richer states, as captured by per capita income, ought to have better credit ratings.

Finally, we include a standard set of political (dummy) controls covering the 1990s to account for the possible effects of fiscal institutions: revenue limit, expenditure limit, debt issuance restriction, and popular vote requirement (all coded one if such a restriction exists) from Poterba and Rueben (Reference Poterba and Rueben1997, 35).Footnote ²² Revenue and (tax or) expenditure limits typically restrict the growth rate of revenues as well as general fund expenditures (coded 1 if present, 0 otherwise in each case). Debt issuance restrictions refer to limits on the ability of states to issue long-term general obligation debt (coded 1 if present, 0 otherwise). Finally, the popular vote requirement captures states that require a popular vote to issue debt (coded 1 if present, 0 otherwise). Following Poterba and Rueben (Reference Poterba and Rueben1997), we focus on binding limits—limits that cannot be overridden by a simple legislative majority.

Table 1 presents summary statistics of the variables in our analysis.

TABLE 1.

Summary Statistics of Variables

Given our ordinal outcome, we estimate a series of ordered probit regressions typical in the literature. The first examines the effect of rule clarity and rule strictness with the consolidated bond credit ratings from all three ratings agencies as the outcome. The second includes a two-way interaction between clarity and strictness (clarity*strictness). As a robustness check, a third set of models uses bond credit ratings from each individual agency (Moody’s, Standard and Poor’s (S&P), and Fitch) as the outcome.

NEW STRUCTURALLY BALANCED BUDGET RULES IN THE EUROZONE

The eurozone sovereign debt crisis revealed the failure of the governance regime that had been put in place at the inception of the euro to prevent eurozone states from accumulating excessive deficits. As E.U. leaders reform eurozone governance in the wake of the crisis, they seek to reduce political discretion over fiscal policy, empowering courts to enforce balanced budget rules to ensure member state fiscal discipline. This approach is manifest in the Fiscal Compact Treaty, which requires eurozone governments and other signatories to impose legally binding, “structurally balanced budget” rules, to be enforced by national courts. However, the analysis of the impact of balanced budget rules we developed above suggests that the E.U.’s effort to judicialize fiscal policy through balanced budget provisions of the Fiscal Compact Treaty is unlikely to succeed.

To place the E.U.’s current approach in perspective, we must turn briefly back to the regime for eurozone governance initially established under the Maastricht Treaty. In the Maastricht Treaty, the E.U. put in place two devices intended to prevent member states from accumulating excessive debt. First, under the Stability and Growth Pact (SGP) states were required to limit their deficit and debt levels. Second, to protect against moral hazard and bolster market discipline, the Maastricht Treaty contained a “no bailout” clause (Article 104b, now Article 125 TFEU) stating that neither the Community nor any individual member state could bail out (i.e., assume the debts of) another member state. Ultimately, however, both pillars of the Maastricht regime broke down. The Stability and Growth Pact crumbled in the mid-2000s after Germany and France, with the support of other member states, blocked a Commission effort to enforce the SGP and managed to violate its provisions with impunity (Heipertz and Verdun 2010, chaps. 7–8). This led in 2005 to a reform of the SGP that substantially loosened its requirements. The no bailout rule was violated in the Greek crisis, when the E.U.—unwilling to countenance a disorderly Greek default—agreed to cooperate with the IMF on a bailout of Greece. Bailouts for Ireland and Portugal followed soon after, and in February 2011, E.U. leaders agreed to establish a permanent bailout fund, the European Stability Mechanism (ESM).

As the cornerstones of the Maastricht eurozone governance regime were crumbling, E.U. leaders moved rapidly—if rather haphazardly—to construct a new regime designed to limit the moral hazard of member states running up unsustainable debts and turning again and again to the E.U. for bailouts. In December 2011, EU lawmakers agreed to the so-called “Six-Pack” of economic governance reforms, which strengthened the European Commission’s powers to monitor member state fiscal performance and to sanction those that violated E.U. requirements (European Commission 2011). In February 2012, E.U. leaders went a step further, with 25 governments signing the Fiscal Compact Treaty. In part, the Compact simply restated and raised the profile of measures already agreed to in the “Six-Pack,” but the Compact did change the E.U.’s approach in one crucial respect. Previous approaches to ensuring fiscal discipline had precluded a role for the judiciary and had relied on E.U.-level political institutions such as the Council and Commission. But the Compact Treaty sought to both judicialize and decentralize fiscal discipline, requiring all 25 signatory states to enshrine a balanced budget rule and an automatic correction mechanism (debt brake) in domestic law and to enforce these rules through their national courts. In a sense, the Compact Treaty tried to create by fiat a system with important similarities to that found in the U.S., where 49 of the 50 states have embedded balanced budget rules in state law (see footnote 2 above).

As discussed above, this approach to fiscal discipline has had considerable success in the U.S. Nevertheless, our analysis suggests that the Fiscal Compact Treaty’s approach to ensuring member state fiscal discipline is unlikely to work. To be fair, a number of the measures in the Compact Treaty and the Six-Pack—such as the strengthening of the Commission’s surveillance of national budgets—will enhance transparency and prevent the sort of fiscal deception that went on in Greece. These reforms will also make it easier for markets to discipline governments. However, other central elements of the new regime—those that focus on judicializing fiscal policy—are unlikely to function as hoped, in part because their design is based on a misunderstanding of the role law and courts play in maintaining fiscal discipline. Two sets of factors explain why the national balanced budget laws called for in the fiscal compact treaty are unlikely to work.

First, our argument suggests that balanced budget laws can work because they provide a focal point that enables investors to coordinate on when to punish governments for running excessive deficits—not because they establish a credible threat that the judiciary will step in to force governments to balance their budgets. From this perspective, the crucial attribute of a balanced budget rule is not the strength of its legal enforcement mechanisms, but rather its clarity. The clearer the rule in question, the more obvious it is when a government violates that rule and the better investors can coordinate on enforcing the rule (see Hypotheses 1 and 2 above). However, the “structurally balanced budget” rules called for in the Fiscal Compact Treaty are anything but clear. The Treaty focuses on the legal enforceability of the rules, but leaves the rules themselves extremely vague. Therefore, they are unlikely to help coordinate decentralized punishment of irresponsible governments.

Contrary to widespread reporting, the Fiscal Compact Treaty does not impose a clear-cut balanced budget rule on member states. Instead, the Treaty relies on the much more elastic concept of a structurally balanced budget and even then provides substantial discretion by allowing violations of the structural balanced budget rule in exceptional circumstances. A structurally balanced budget rule does not require a government to balance its budget every year, but rather requires that the budget be balanced across the business cycle. This allows for governments to engage in Keynesian countercyclical policies during economic downturns, but then requires them to make up for this by running surpluses during periods of growth.Footnote ²³ Treaty Article 3(1)(b) explains that a state’s budget will be considered to be “balanced” if, “the annual structural balance of the general government is at its country-specific medium-term objective, as defined in the revised Stability and Growth Pact, with a lower limit of a structural deficit of 0.5% of the gross domestic product at market prices.”Footnote ²⁴ Article 3(1)(c) further allows that states “may temporarily deviate from their respective medium-term objective or the adjustment path towards it only in exceptional circumstances.” Precisely what constitutes a cyclically adjusted, structurally balanced budget is subject to interpretation and debate as what might constitute allowable “exceptional” circumstances. As a letter from Professor David Cameron to the Financial Times put it, “Fiscal pact requires reliance on a statistic that cannot be observed.” (Cameron Reference Cameron2012). These vague concepts will not serve as clear focal points that could coordinate decentralized punishment by bond markets of excessive sovereign borrowing.Footnote ²⁵

Supporters of the Treaty’s fiscal rules have made much of the fact that states are required to embed them in national law “through provisions of binding force and permanent character, preferably constitutional, or otherwise guaranteed to be fully respected and adhered to throughout the national budgetary processes” and to back them with an automatic “correction mechanism” (Article 3(2)). This legal enforceability might be important if it was the fear of legal enforcement that restrained government borrowing. But we suggest that fear of legal enforcement is not the causal mechanism through which balanced budget rules keep governments in check. Indeed, court rulings concerning fiscal policy are just as likely to undermine government efforts to cut deficits as they are to aid such efforts. A recent ruling by the Portuguese Constitutional Court nullifying several austerity measures underscores that courts may not simply be counted on to enforce balanced budget rules; rather, they may be just as likely to override deficit reduction measures in the name of other constitutional values (Wise Reference Wise2013). If it is decentralized punishment by bond markets that discourages governments from violating fiscal rules, then the clarity of those rules is paramount. And that clarity is strikingly absent in the provisions of the Fiscal Compact Treaty.

Secondly, in the E.U. the background conditions that surround these balanced budget rules undermine the credibility of the rules. Scholars of fiscal federalism emphasize that market forces can only effectively discipline subnational borrowing if there is no perceived chance of a bailout by the central government or central bank (Bordo, Markiewicz, and Jonung Reference Bordo, Markiewicz and Jonung2011, 6, 26; Hallerberg Reference Hallerberg2010). Thus, for instance, in the U.S., state level balanced budget rules have been able to restrain state deficits because the federal government had established a credible commitment not to bailout states. With the establishment of the ESM, there is now a permanent structure in place designed to provide bailouts to E.U. member states. The possibility of bailouts under the ESM is likely to undermine the effectiveness of market discipline. Nevertheless, because of the remaining uncertainty about the availability of bailouts for any given state and the extent of “private sector involvement” (including losses imposed on bondholders), bond markets can still play a constructive role in restraining excessive state borrowing.Footnote ²⁶ However, this will only be the case where bond markets are presented with sufficiently clear focal points.