Electoral Cycles in Fiscal Rules: International Empirical Evidence

Abstract

Using panel data from 99 countries over the period of 1985–2020, this study aims to uncover that fiscal rules work in electoral cycles. The results show that fiscal rules are less likely to be implemented in the pre-election year. We also find that this loosening in fiscal rules is especially prominent in cases when the incumbent political party can wield significant political power, where countries are less developed economically and politically, and where the management of fiscal rules is more likely to be exposed to discretionary interventions. Our findings are aligned with the opportunistic political budget cycles, in which politicians exploit fiscal policies and systems to serve their own political incentives. These results suggest that even fiscal rules, which are designed to prevent political pressure, are also subject to political process. (JEL codes: D72, D78, E32, E62, H11, and H60).

1. Introduction

Political budget cycles generally refer to increases in government spending or deficits, or decreases in taxes, prior to an election, which are perceived as being motivated by the incumbent’s desire to maximize political gains (Nordhaus 1975; Rogoff and Sibert 1988; Brender and Drazen 2005; Shi and Svensson 2006). As a result, fiscal policies become distorted. In particular, the president and ruling party can manipulate government policies to increase the likelihood of winning the next election, and fiscal policies and governmental budget can be a useful tool in this regard. For instance, large-scale national projects or social welfare-related expenditures may be introduced in the year prior to the election to influence outcome in a favorable direction by incumbent politicians. This tendency gives rise to the political budget hypothesis, which suggests that variables related to fiscal policy can be predicted based on the electoral cycle.

The political budget cycle arises because the budget is fundamentally under the discretion of politicians and bureaucrats with authority, who act as rational agents seeking to maximize private interests (Downs 1957; Niskanen 1971; Wildavsky and Caiden 1988). Consequently, the political budget cycle is essentially related to opportunistic behavior. From this perspective, politicians’ discretionary intervention in fiscal policies according to the electoral cycle can lead to rent-seeking activities and economic fluctuations, resulting in various side effects (Cerra et al. 2023).

Fiscal rules refer to a framework for managing fiscal policy that establishes specific quantified objectives for fiscal indicators such as the fiscal balance, government expenditure, and national debt, and implements legislative measures to achieve these objectives (Kopits and Symansky 1998). In particular, fiscal rules can limit the discretionary fiscal management of fiscal policy authorities by making the budget operate according to pre-determined rules (Fatás and Mihov 2003; Bova et al. 2015; Yared 2019). Introducing fiscal rules is considered as a method to address concerns over chronic fiscal deficit bias and ensure fiscal soundness and sustainability.

The preceding research has primarily focused on the effects or outcomes of fiscal rules (Potrafke 2023). In examining the relationship between political business cycles and fiscal rules, the existing literature has predominantly concentrated on the efficacy of fiscal rules in preventing the political budget cycle (Rose 2006; Bonfatti and Forni 2019; Gootjes et al. 2021). However, it should be noted that rules and systems are susceptible to political influence and may be manipulated at the discretion of political stakeholders (North 1991). Recent emerging literature has begun to garner attention by moving beyond merely verifying the effectiveness of fiscal rules to scrutinizing how they operate in different contexts (Badinger and Reuter 2017a). Consequently, there has been an increasing emphasis on studies examining the broader societal context in which fiscal rules operate, including factors influencing their adoption and compliance (Reuter 2019; Ulloa-Suárez 2023). Given the political nature of fiscal rules, examining their operational conditions within the context of elections can significantly contribute to the existing body of research. Addressing this gap in the literature, this article seeks to explore how the implementation of fiscal rules does not entirely eliminate political influence (Benito et al. 2013; Heinemann et al. 2018) but rather functions within an electoral cycle.

Using panel analysis of data from over 99 countries during the 1985–2020 period, we find evidence that the implementation of fiscal rules is looser in the run-up to elections with uncertain outcomes, where incumbent governments may have the strongest incentive to interfere with fiscal systems. Our results remain robust across multiple model specifications. Additionally, we find that these effects are more pronounced in environments where the president and ruling party have greater political power, in settings with lower levels of economic, political, and social development, and in cases where the operation of fiscal rules is subject to discretionary interventions. These findings suggest that the management of fiscal rules, which are intended to prevent the political budget cycle, is not immune to political interference. Our evidence supports the idea that incumbent governments use fiscal systems to strengthen their political power, consistent with models of opportunistic political cycles.

A major contribution of our study is to expand the scope of prior research by confirming the existence of electoral cycles in the operation of fiscal rules. Earlier studies have focused on macroeconomic outcome variables such as Gross Domestic Procuct (GDP), inflation, unemployment, and fiscal policy variables that are under direct government control.¹ Recent studies have broadened the scope to include other indices such as tax reform, credit, labor, rules, and regulations, finding evidence of political budget cycles in these areas as well.² To the best of our knowledge, our study is the first to empirically investigate the electoral cycles of fiscal rules. Moreover, our heterogeneity analysis contributes to the literature on conditioning political budget cycles and demonstrates that the electoral cycle analyzed from the perspective of fiscal rules also varies depending on environmental and institutional factors.

Our study is organized as follows. Section 2 develops our main hypotheses based on existing literature. Section 3 introduces the data and empirical strategy. Section 4 presents our estimation results. Section 5 concludes the study.

2. Related Literature and Hypotheses Development

The literature has noted that government budgeting and fiscal management are inherently political processes (Wildavsky and Caiden 1988). Politicians and bureaucrats often manipulate the budget and its system to serve their own personal and professional interests (Niskanen 1971). Accumulating evidence suggests that a range of political actors, particularly the president, ruling party, and powerful lawmakers with fiscal authority, utilize fiscal policies for rent-seeking activities. For instance, the president may exploit fiscal authority to enhance the political power of their affiliated party (Larcinese et al. 2006; Berry et al. 2010), and a similar phenomenon may arise when lawmakers possess budgetary powers (Levitt and Snyder 1995; Albouy 2013).

Elections are a crucial aspect of modern politics in representative democracies. Theoretically and empirically, it has been argued that fiscal policies fluctuate in accordance with major electoral cycles, a phenomenon known as political budget cycles (see reviews by Drazen 2000; Dubois 2016; Mandon and Cazals 2019). Political budget cycles have been systematically studied since Nordhaus (1975), who introduced the opportunistic model, arguing that the incumbent president or ruling party uses various measures to boost the economy before an election to win the election and implements austerity measures after the election to slow down inflationary pressures caused by expansionary policies. Rogoff (1990) developed this model and introduced the assumption of rational voters into the opportunistic model, resulting in the rational opportunistic model. He suggested that voters support the incumbent party when it uses fiscal spending to boost the economy before an election because voters have incomplete information about the ability of the president under the circumstances of asymmetric information.

The core point of these theoretical discussions is that political budget cycle occurs because the rulers use expansionary policies to ensure that the economy is booming before the election to win the election, while after the election, they implement austerity measures to curb inflation. The president and ruling party act as economic entities that seek to maximize their interests, and their political utility is maximized when they maintain their political status through re-election and continuing their preferred policies. To achieve this, the incumbent politicians with fiscal authority may introduce new spending programs or tax cuts before the election to increase their chances of re-election, resulting in systematic fluctuations of fiscal policies by periodic electoral cycles. Electoral motivations for politicians to engage in public spending that benefits voters, and the subsequent occurrence of political budget cycles, have been widely supported by empirical evidence in various contexts, including both developed and developing countries.³

Large business cycles can hinder economic growth, exacerbate unemployment, income inequality, and inflation (Hamilton 1988; Parker 1998; Fatas 2000; Cerra et al. 2023). In particular, the political budget cycle caused by changes in fiscal policies due to politicians’ rent-seeking behaviors leads to problems, such as resources being used for the benefit of the state and the public interest being wasted. Bureaucrats may introduce poorly considered, unnecessary, or low-demand projects to expand their own influence, which can undermine fiscal democracy and budget efficiency.

Fiscal rules serve as a means to achieve fiscal soundness and prevent discretionary abuse of fiscal policies. A fiscal rule is a fiscal framework that legislates fiscal management goals with specific targets for fiscal indicators such as government debts, revenues, and expenditure. It requires three major factors: legal foundation, fiscal goal, and sanctions on violations (Kopits and Symansky 1998). Since the 1970s, fiscal rules have been introduced to cope with the increase in macroeconomic instability caused by the dismantling of the Bretton Woods system, the surge in national debt due to oil price shocks, and financial and foreign exchange crises. Recently, fiscal rules have been introduced to preemptively prepare for an increase in welfare expenditure and future financial burdens due to an aging population and to ensure fiscal sustainability.

The fiscal rules are largely divided into four types: budget balance rules (BBR), debt rules (DR), expenditure rules (ER), and revenue rules (RR). BBR is a method of setting the target variable as the fiscal balance and keeping the fiscal balance in a balanced or small surplus every fiscal year or for a certain period of time. DR is a method of imposing restrictions on government authorities to maintain or gradually reduce the ratio of national debt to GDP at a certain level. DR is not a direct control of fiscal income or expenditure but a method of managing the final national debt as a result. ER is a method of securing fiscal soundness by imposing constraints on total expenditure, total expenditure growth rate, and total expenditure ratio to GDP. RR stipulates the highest or lowest limit on income to set an upper or lower limit on revenue to increase revenue collection and prevent an excessive tax burden.

Fiscal rules are commonly implemented at the national level, but they are also used by supranational organizations such as the European Union (EU). Examples of fiscal rules include the Gramm-Rudman-Hollings Act and the Budget Enforcement Act, which were enacted by the US government for fiscal austerity, the Stability and Growth Pact, which allows European Monetary Union member states to maintain fiscal deficits and national debt within 3% and 60% of GDP, and the Golden Rule, which ensures that most states in the USA maintain a current fiscal balance.

The literature has investigated whether fiscal rules function as a means of fiscal discipline by focusing on various aspects such as fiscal balance (Grembi et al. 2016; Caselli and Reynaud 2020), national debt (Azzimonti et al. 2016; Asatryan et al. 2018), fiscal soundness (Bergman et al. 2016; Badinger and Reuter 2017b), government expenditure (Dahan and Strawczynski 2013), and government accounting (Milesi-Ferretti 2004). An increasing number of studies also examine the impact of fiscal rules on non-fiscal variables, including the macroeconomic business cycle (Levinson 1998; Fatás and Mihov 2006; Gootjes and de Haan 2022), as well as government financing costs and interest rate spreads (Thornton and Vasilakis 2018; Afonso and Jalles 2019).

Recently, studies have linked fiscal rules and political budget cycles, focusing on whether fiscal rules can prevent political budget cycles. The effects of these studies are inconclusive, with some suggesting that fiscal rules can be effective (Rose 2006; Bonfatti and Forni 2019; Gootjes et al. 2021; Strong 2023), while others suggest the effects are uncertain (Buti and Van den Noord 2004; Heinemann et al. 2018), depending on the context of the studies.

Several studies point out that fiscal rules themselves are flawed in controlling for political influence on fiscal policies, and various channels do not effectively control the political budget cycle (Benito et al. 2013; Reuter 2019). For instance, Benito et al. (2013) argue that although governments are constrained by such rules, they engage in electoral-induced fiscal manipulation such as circumventing legal debt limits, expenditure deferment, and creative accounting. They point out that governments create room for maneuver to implement expansionary fiscal policies in electoral years.

It is noteworthy that fiscal rules are also an institutional rule that politicians can ultimately manipulate. As the institutional system is also politically formed (North 1990; March and Olsen 1996; Acemoglu et al. 2005), fiscal rules also work in the political budget cycle. Specifically, fiscal rules may not be introduced in the year just before an election because incumbents seeking to implement expansionary fiscal policies to gain political support do not want to tie their hands on fiscal policies. Moreover, this tendency may be more prominent in the case of BBR and ER that constrain expenditure from a short-term perspective compared to DR and RR, which are more long-term and difficult to control. Thus, the following hypothesis can be constructed:

Hypothesis 1: Fiscal rules are less likely to be implemented in the year preceding an election, and this tendency is more pronounced for rules that are easier to be influenced by incumbent politicians in the short-term perspective.

Previous studies have posited that the presence of the political budget cycle is not homogeneous and varies depending on the environment. These disparities are characterized by the conditional political budget cycle theory (Shi and Svensson 2006; Alt and Rose 2009; Potrafke 2010; Aidt et al. 2011; De Haan and Klomp 2013; Veiga et al. 2019). Moreover, recent series of studies such as those by Badinger and Reuter (2017a), Reuter (2019), and Ulloa-Suárez (2023) have analyzed the determinants of fiscal rule implementation, an area that remains relatively underexplored. They examine the impact of various economic, social, and political factors, and the characteristics of fiscal rules themselves on the adoption and operation of fiscal rules. Their research highlights that determinants such as government and political environment, institutional features, and economic and social environments may particularly be relevant to the stringency and compliance of fiscal rules.

To specify, heterogeneities can be observed from a political perspective, where the concept of political rent creation within institutional systems suggests that interest groups manipulate government mechanisms to secure special advantages (Hartle 1983; Boucher 1991). In contexts where political checks and balances are ineffective, powerful political factions may wield significant influence over the implementation of fiscal rules. These rules, crafted to ensure fiscal stability, might hinder the establishment and execution of fiscal policies by influential political entities seeking public support in elections. Consequently, such groups may be motivated to circumvent the fiscal rule system.

Furthermore, disparities between developed and developing countries, stemming from variations in institutional quality and socioeconomic development, may manifest in our findings. Seminal works by Acemoglu et al. (2005) underscore the evolution of institutions over centuries, adapting to local conditions and technological advancements, leading to inclusive and participatory systems in advanced countries. In contrast, institutions in developing nations, often imposed during colonial periods or hastily established post-independence, may not align well with local needs or social norms, resulting in weaker enforcement and compliance. Governance challenges such as corruption, lack of accountability, political instability, and lower levels of social capital further undermine the effectiveness of institution in many developing countries. These characteristics lead us to infer the heterogeneities in the realization of electoral cycles in fiscal rules across multiple socioeconomic dimensions.

Lastly, disparities in the electoral cycle may arise from the perspective of fiscal rules themselves. The lack of monitoring, enforcement, and legal foundations of fiscal rules may enable political and economic stakeholders to circumvent the systems and provide channels for operation under political influence, rendering the enforcement of the original system nominal (Reuter 2019; Ulloa-Suarez and Valencia 2022).

Taken together, these discussions imply that the execution of fiscal rules under electoral cycles can vary greatly depending on specific circumstances. It is reasonable to suspect that the operation of fiscal rules is greatly influenced by political and socioeconomic factors, such as the political status, electoral rules and form of government, the presence of political checks and balances, the level of economic development, institutional quality, as well as specific characteristics of fiscal rules, such as the availability of independent monitoring, the possibility of enforcement, and the establishment of legal foundations.

Following this, we infer that there are three cases in which fiscal rules are vulnerable to electoral cycles: first, when the incumbent’s political power can be greatly exercised; second, when the degree of economic, social, and democratic development is not mature; and third, when the basis for fiscal rules is not securely established. Thus, we can derive Hypothesis 2 as follows:

Hypothesis 2: The extent of electoral cycles in fiscal rules varies depending on the political situation, socioeconomic development, and the operating principles of fiscal rules.

3. Method

We constructed a panel data set of 99 countries⁴ from 1985 to 2020 by collecting relevant variables from various sources, such as the International Monetary Fund (IMF), the Database of Political Institutions by the World Bank, the World Development Indicators (WDI), and previous studies.

Regarding the dependent variable, the implementation of fiscal rules, given the diverse array of fiscal rule types and operational contexts across different countries and years, it is crucial to ensure consistency and comparability in the fiscal rule data used for country-level panel analysis. This necessitates the utilization of reliable data structured in a standardized format for cross-country and cross-year comparisons, as well as the conversion of varied fiscal rule characteristics into comparable numerical indicators.

To construct the indicator of fiscal rule implementation, our study relies on information regarding fiscal rules provided by the IMF’s Fiscal Rules Dataset (Davoodi et al. 2022), which encompasses four categories of national and supranational fiscal rules: ER, RR, BBR, and DR. Specifically, this study considers national fiscal rules across these four categories, with observations of each specific fiscal rule represented by 0-1 dummies at the country-year level. We then computed the total value of dummy variables across the years.⁵

Subsequently, we evaluated the stringency of the operational principles of fiscal rules by considering factors such as the monitoring of compliance outside the government, enforcement procedures, and legal basis.⁶ For each type of fiscal rule, we calculated the total value of dummy variables representing the application of these stringency conditions.

Figure 1 illustrates the implementation status of fiscal rules over the specified period. Consistent with similar references (Jung and Kim 2021), our constructed indicator demonstrates reliability, as it aligns with the trend observed in many countries where fiscal rules have been increasingly adopted and strengthened in recent years (Davoodi et al. 2022).

The election year by country is obtained from Cruz et al. (2021). This study uses both presidential and parliamentary elections as explanatory variables, considering that the powers of the executives and legislatures vary depending on whether it is a presidential or parliamentary cabinet system (Foremny et al. 2018). Figure 2 illustrates presidential and legislative elections by country and year.

Table 1 presents descriptive statistics, detailing the sources and descriptions of each variable. Concerning the dependent variable, among the year-country observations, about one out of four fiscal rules has been implemented, with the budget balance rule and debt rule being the most prevalent. As for the explanatory variable, the occurrence rate of presidential and legislative elections in the data is 11% and 24%, respectively.

In this model, i is a country and t is a year. The dependent variable (⁠$Fiscal Ruleit$⁠) serves as an indicator for the implementation of fiscal rules. The explanatory variable (⁠$Electioni,t+1$⁠) represents either presidential or legislative elections and is equal to 1 if an election is held at year t + 1. Following Sever and Yücel (2022), we account for the possibility that electoral cycles in fiscal rules may be exhibited in the year prior to the election. The covariates (⁠$Xit$⁠) include time-varying economic, fiscal, demographic, and political characteristics of each country. We also include the country (⁠$λi$⁠) and year (⁠$μt$⁠) fixed effects to control for time-invariant country characteristics as well as macroeconomic trends. Standard errors are clustered at the country level.

4. Results

Tables 2 and 3 present the results for Hypothesis 1. Columns (1)–(4) display the outcomes for the presidential election as the explanatory variable, and columns (5)–(8) for the legislative election.

Consistent with the theory of opportunistic electoral cycles, we find that the implementation of fiscal rules is less likely in the year before a presidential election when only fixed effects are controlled (column 1). The results remain consistent when we include control variables, except for fiscal indicators (column 2). When we include fiscal indicators such as government gross debt and fiscal balance, which may be correlated with the implementation of fiscal rules, the number of observations decreases due to data limitations; however, the results remain similar to those in columns 1 and 2 (column 3). On average, 1.174 fiscal rules were in effect during the analysis period. The coefficients are approximately −0.059 for presidential elections and −0.048 for parliamentary elections. These negative values suggest that electoral cycles influence the implementation of fiscal rules, although the effect size is relatively modest. Specifically, this corresponds to an estimated decrease of about 5% of a standard deviation in the likelihood of enforcing fiscal rules in the year leading up to an election, as indicated by our numerical index.

To more rigorously support the existence of electoral cycles in fiscal rules, we test whether the results are significant only in the year preceding the election. The findings confirm our expectation that the implementation of fiscal rules is less likely just before the election, with little evidence found in other adjacent years (column 4). The parameter estimate indicates that the fiscal rule index was by 0.078 points lower (around 6% of a standard deviation of the dependent variable) in pre-election years than in the years after a presidential election. This implies that the incumbent politicians exploit fiscal policies and systems to gain support from the public, resulting in electoral cycles in fiscal rules.

The results for legislative elections are similar to those for presidential elections. The coefficients are statistically significant and negative when we include all control variables and fixed effects (column 7). Moreover, we do not find evidence of a loosening of fiscal rules in the post-election year or in distant years from the election, indicating that our findings are indeed driven by the pressure of upcoming elections on incumbent politicians (column 8).

Table 3 presents the results using four categories of fiscal rules (expenditure rule, revenue rule, budget balance rule, and debt rule) as dependent variables. Distinct outcomes emerge for each type of election. In presidential elections, the implementation of ER is less probable before the election (column 1), indicating an estimated decrease of about 10% of a standard deviation in the likelihood of implementing ER in the year before a presidential election, as reflected in our numerical index. For legislative elections, the execution of the budget balance rule is approximately 3% less likely (column 7).⁷

Tables 4–6 present the heterogeneities of the results for Hypothesis 2. Columns (1) and (3) present the case of presidential elections as an explanatory variable, while columns (2) and (4) present the case of legislative elections.

Table 4 presents the results by dividing the samples according to the possibility of political influence. The electoral cycles in fiscal rules can be noticeable when the president and the ruling party are under substantially favorable conditions to operate fiscal authority, and our findings support this prediction. We find that loosening in fiscal rules before the presidential election is more prominent in the case of a presidential system (Panel A). Furthermore, the results show that the electoral cycles are noticeable in the circumstances when the incumbent politicians are in power, such as both houses of Congress controlled by the ruling party (Panel B), the first party in Congress being the ruling party (Panel C), and shares of the ruling party being greater than those of the first opposition party (Panel D).

Table 5 shows the results of heterogeneity according to the extent of economic and political development. We find that the electoral cycles in fiscal rules are remarkable in cases with a low degree of economic development (Panel A) and a low degree of democracy (Panel B). Using the Organization for Economic Cooperation and Development (OECD) as a criterion for developed countries, the coefficients are only significant in non-OECD countries (Panel C). Additionally, the political budget cycles in fiscal rules are prominent in countries outside the EU with little experience with fiscal rules (Panel D).

Finally, Table 6 reports the results by the operational foundations of fiscal rules. Consistent with our expectations, we find that the electoral cycles are significant when there is less function of monitoring of compliance outside the government (Panel A), enforcement procedures are weak (Panel B), and legal foundations are below average (Panel C).

These results indicate that conditional political budget cycles are also applied in the case of electoral cycles of fiscal rules. Our findings that the political budget cycles depend on the degree of political power, economic and social development, and the functional characteristics of fiscal rules suggest that establishing environmental and institutional systems in operating fiscal rules is crucial.

5. Conclusion

Fiscal rules have become an increasingly important part of fiscal systems worldwide. They aim to ensure fiscal efficiency and soundness by binding fiscal management to rules linked to fiscal indicators and preventing discretionary intervention by politicians. While a rapidly growing academic literature studies the impacts of fiscal rules on fiscal performance, as well as the role of preventing electoral cycles in fiscal policies through fiscal rules, this study takes a step forward by showing empirically that the implementation of fiscal rules also exhibits a systematic electoral cycle.

Using panel data from 99 countries covering the period 1985–2020, the results reveal that fiscal rules are less likely to be executed just before the year of the election. These patterns are consistent with the argument put forth by the opportunistic political business cycles theory, which suggests that incumbent politicians with fiscal authorities make fiscal management more discretionary and expansive just before the election to secure more support from the public. We also find that our findings are more prominent when the president and ruling party can yield more influential power, when the extent of political and economic development is immature, and when the operating principles of fiscal rules are prone to discretionary actions. These results raise the question of whether political economy factors can be a constraint that rationalizes the use of prudential fiscal rules.

Funding

We appreciate financial support from Seoul National University.

Footnotes

References

Appendix

Country Considered in This Study