Fiscal rules and budget forecast errors of Italian municipalities

We study the impact of the domestic stability pact on the accuracy of budget forecasts of Italian municipalities. Identification of the causal effect exploits a quasi-natural experiment generated by the removal in 2001 of the fiscal restraints on budget decisions for municipalities with fewer than 5000 inhabitants and by stricter budgetary restrictions and severe penalties for noncompliers in 2002. We find that relaxing fiscal rules had a sizeable impact on budget forecast errors, especially in 2002. In fact, revenue (expenditure) forecast errors for municipalities with fewer than 5000 inhabitants became 26% (22%) larger than in the past.     

The unintended composition effect of the subnational government fiscal rules: The case of Italian municipalities

Although numerical fiscal rules may be introduced to achieve several objectives, to date the maintenance of fiscal sustainability is their predominant goal. This is particularly true at subnational level; maintaining fiscal discipline in a decentralized setting is challenging and subnational government fiscal rules are considered one of the most valid solutions to the problem. While theoretical and empirical literature has mainly focused on their effectiveness in containing subnational deficit and/or debt, little attention has been paid to the possible trade–offs and side effects of the rules on the composition of subnational expenditure. The aim of this paper is to fill this gap by exploiting the case of Italian municipalities, which have been subject for fifteen years (1999–2015) to a set of rules called Domestic Stability Pact. The Italian DSP framework – imposing rules only on municipalities above a population threshold (5000 inhabitants) – allows us to implement a quasi–experimental technique to investigate the unintended composition effects of the rules. A difference–in–discontinuities design permits to find rigorous empirical evidence that the switching in 2007 to rules which are more binding in terms of fiscal discipline leads to a recomposition of municipal expenditure against investment spending. The analysis is then integrated by evaluating the impact of the rules on six categories of investment expenditure. Investment in human capital and infrastructure seems to be the most affected.     

Exposure to corruption and political participation: Evidence from Italian municipalities

This paper aims to study the effects of exposure to corruption on all the aspects of political participation. Focusing on Italian municipalities in the period 1999–2014, we generate a daily and local measure of exposure to corruption, screening newspaper articles of the main Italian press agency. We concentrate on local elections and, in an event-study analysis, we find three main results. First, corruption exposure affects citizens' participation in election by reducing voter turnout. Second, corruption impacts on politicians’ participation: the number of candidates and electoral lists decreases after a scandal and candidates with political tenure are more likely to run. Finally, these changes affect local political outcomes as tenured politicians are more likely to be elected, while freshmen lose ground. These results suggest that exposure to corruption has general and negative effects on political participation, leading people to lose interest in politics.     

The effect of female representation on political budget cycle and public expenditure: Evidence from Italian municipalities

This study analyzes the impact of the gender composition of political institutions on the political budget cycle (PBC) and on the size and structure of public expenditure. An instrumental variable approach is implemented to evaluate the influence of female politicians in municipal councils. The introduction of gender quotas for Italian municipalities is used as an exogenous variation in female participation in politics. The results show that: (i) fluctuations in local public spending are only slightly affected by the presence of a wider female representation; (ii) an increase in the number of elected women reduces the overall amount of public expenditure; (iii) this reduction involves fields typically affected by PBC (e.g., roads' maintenance) except those related to females' needs (e.g., kindergarten, primary education, and social care).     

The electoral migration cycle

This paper proposes a new test of Tiebout sorting that relies on the exogenous time structure of recurrent local elections. The test is based on the idea that competitive elections represent periodic perturbations to the Tiebout equilibrium of local public good provision and allocation of households to communities, so that their schedule should affect the timing of households' sorting decisions. On the other hand, internal migration flows that have nothing to do with the demand for public goods over which localities vote recurrently ought to be orthogonal to the timing of elections in a reduced-form migration equation. I exploit the staggered schedule of mayoral elections in Italy to analyze migration, elections, and public budget data across several thousands of municipalities, and find evidence of a systematic influence of the electoral calendar on the timing of sorting decisions.     

Disentangling political and institutional determinants of budget forecast errors: A comparative approach

This paper provides a unique comparison between French and Portuguese local governments with respect to the nature and determinants of budget forecast errors. It starts by documenting and comparing their statistical properties. The results point at biased and inefficient budget forecasts, which seem to have been more cautious in French departments than in Portuguese municipalities. Second, we examine the political, institutional and economic determinants of forecast biases. Overall, we find that they are essentially driven by electoral motivations and by institutional differences across the two countries. In particular, opportunistic forecasting is more prevalent where governments enjoy greater margin of maneuver, and there is evidence of conservatism in French departments where fiscal autonomy is greater.     

Legislators' behaviour and electoral rules: Evidence from an Italian reform

We explore how electoral rules and the degree of civicness of the voters interact in shaping elected officials' behaviour. We examine the expenditure proposals sponsored by Italian Senators from 1994 to 2013 and exploit the 2005 electoral reform that transformed a mainly majoritarian system into a mainly proportional one. First, we find that legislators elected in first-past-the-post districts are more likely to sponsor pork-barrel bills and to put effort into legislative activity than those elected with a closed-list proportional system. Second, more importantly, we show that the effects of the electoral rules are muted in areas with high civicness.     

Political budget cycles revisited: Testing the signalling process

A widespread view in the ‘political budget cycles’ literature is that incumbent politicians seek to influence voters’ perceptions of their competence and/or preferences by using the composition of the fiscal budget as a signalling tool. However, little is known about whether voters actually receive and perceive the signal in that way. To empirically assess the relevance of the signalling channel at the municipal level, we conducted a survey among 2000 representative German citizens in 2018. Only a small fraction of voters feel well-informed about the fiscal budget signal and use the information it contains to decide whether to vote for the incumbent politician. Persons paying more attention to the signal sent by local politicians live in smaller municipalities, are more satisfied with their economic situation, are more educated, and do not feel that they are being electorally manipulated. Our analysis raises doubt about the relevance of budget composition as a signalling mechanism for voters at the local level.     

The influence of political fragmentation on public enterprises: Evidence from German municipalities

This study investigates the influence of politically fragmented municipal councils on the investment behaviour of municipally owned public enterprises. We argue that common public choice theories can also be applied to public enterprises. For our observation period (2002–2014), we use a data set encompassing 8,685 municipally owned enterprises subject to commercial accounting located in 3,237 municipalities drawn from all 13 German federal area states. On average, our results point to a negative effect of fragmentation on investment per capita, supporting a war of attrition or veto player approach. Effects are heterogeneous among different sectors, possibly due to higher visibility to voters. Thus, veto players might actively try to block prestigious and publicly perceivable investment projects to hinder their opponents. Political budget cycles additionally influence this effect for some sectors. However, testing non-linearities reveal that the common-pool hypothesis cannot be completely ruled out.     

Gender quotas and the selection of local politicians: Evidence from French municipal elections

Using data on the universe of elected politicians in French municipalities, this paper studies the impact of a gender quota law on the political representation of women and on the composition of municipal councils. The empirical strategy, a Difference-in-Discontinuities design, takes advantage of the fact that the policy applies to cities above a population threshold, and that this threshold has been modified over time. I find that the quota policy has a substantial impact on the share of female candidates and elected politicians, but fails to promote female mayors and list leaders, even in cities that have been exposed to the policy for 13 years. Women do not reach leadership positions because they are more likely to resign than male politicians. This higher propensity of women to leave politics is correlated with local gender norms concerning the place of women in society, and also varies with individual characteristics such as age and professional background. In a second part, I show that quotas have little effect on the composition of municipal councils in terms of socio-economic background, age, and political experience.     

Nonlinear effects of government debt on private consumption: Evidence from OECD countries

This paper investigates nonlinear effects of government debt on private consumption. The estimated consumption function shows smooth regime switching depending on the debt-to-GDP ratio, and a higher level of government debt crowds out private consumption to a greater extent.     

Birthplace favoritism and the distribution of budget amendments in Brazil: Evidence from nondistrict elections

In this paper, we investigate whether federal deputies in Brazil display birthplace favoritism in allocating discretionary resources to municipalities via budgetary amendments. The data include information on three electoral cycles and on more than five thousand municipalities. Unlike previous research, we match data on an elected deputy’s share of votes by municipality and the value of the legislative amendments per capita obtained for each municipality by that deputy. Our results suggest that although resource distribution through amendments is smaller than targeted pork spending, birthplace favoritism is still an important driver of local politicians’ behavior. On average, when a Brazilian municipality is the place of birth of a deputy, it obtains 8.7% more amendments per capita than its counterparts. Additionally, our findings suggest that some deputies display birthplace favoritism because they plan to run for mayor in their hometown in subsequent elections.     

Term limit extension and electoral participation. Evidence from a diff-in-discontinuities design at the local level in Italy

We study the effect of term limits on voter turnout in Italian local elections. Since 2014 the Italian law allows mayors in municipalities with a population size lower than 3,000 inhabitants to re-run for a third term, whereas mayors in cities with a number of residents above the cut-off still face a two-term limit. The introduction of the reform permits us to implement a difference-in-discontinuities design exploiting the before/after with the discontinuous policy change. We find that voters negatively react to the introduction of the reform: electoral participation decreases by about 5 percentage points in municipalities eligible to the treatment compared to municipalities in the control group. This negative effect is essentially driven by a decrease in the political competition. We also find that relaxing term limits does not improve the quality of politicians running for election.