## Political Budget Cycle in log-Capital Spending of Municipalities at Population Thresholds Relevant for Mayor's Wage. RD-FE Estimates (2004-2006)

## Source details

**Canonical URL:** [Political Budget Cycle in log-Capital Spending of Municipalities at Population Thresholds Relevant for Mayor's Wage. RD-FE Estimates (2004-2006)](https://www.imf.org/-/media/files/publications/wp/wp1706.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/wp1706.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/wp1706.pdf.json)

---

### Introduction and main claim
- Fiscal rules can help moderate the political budget cycle (increases in government spending or the deficit, or decreases in taxes, in an election year or pre-election years driven by incumbents’ re-election incentives).
- Context: Italian municipalities in the early 2000s under the Domestic Stability Pact (DSP) introduced in 1999.
- Identification lever: municipalities below 5,000 inhabitants are exempt from the DSP; municipalities above 5,000 are subject to the rule.
- Core quantitative findings:
  - The political budget cycle increases real capital spending by about 35 percent on average in the three years prior to municipal elections.
  - The sub-national fiscal rule reduces these figures by about two thirds.

### Institutional setup (DSP and local electoral rules)
- Domestic Stability Pact (DSP):
  - Introduced in 1999 to include sub-national authorities in meeting European fiscal targets.
  - Operational target for municipalities changed over time (limits to spending in 2005–06; overall balance prior to 2005 and from 2007 onward).
  - Penalties for non-compliance: limits on hiring, on spending, and on borrowing for investments.
  - Since 2001, municipalities with less than 5,000 residents have been exempted from the DSP.
  - The DSP was discontinued in 2015 and replaced with a budget balance rule for all local authorities.
- Municipal governance and elections:
  - Decision bodies: mayor (Sindaco), Executive committee (Giunta comunale), municipal council (Consiglio comunale).
  - Electoral systems:
    - Municipalities with less than 15,000 inhabitants: simple plurality, each candidate supported by a single list.
    - Over 15,000: multiple lists possible and run-off if no absolute majority.
  - Since 1993 municipal elections every four years; since 2000 mayoral term extended to 5 years (with exceptions).
  - Elections usually held in May and June.

### Data (period, coverage, key aggregates)
- Analysis focuses on capital spending; period limited to before 2007. Main dataset covers 2004–2006.
- Data combined: municipal budgets, municipal election dates, mayor characteristics (age, education, gender, party).
- Aggregate and financing facts (2004 baseline year and dataset descriptions):
  - Total spending of municipalities in ordinary-statute regions represented almost 5 percent of GDP in 2004.
  - Capital spending represented about 38 percent of total spending.
  - In real per capita terms, municipalities spent about €600 annually in investment.
  - Transfers from regions and the central government over the period represented about 40 percent of overall revenues; own revenues covered the rest.
  - Main municipal taxes: ICI (real estate tax on home property) provided about 43 percent of municipal tax revenues; IRPEF surcharge amounted to about 6 percent of municipal tax revenues.
- Notes on fiscal-rule operational details in the analysis window:
  - 2004: difference between current spending and current revenues could not be higher in real terms than in 2003.
  - 2005: current and capital spending should have been lower than the average over 2001–03 increased by 10 percent.
  - 2006: current spending should have been lower by 6.5 percent with respect to 2004 (by 8.1 percent for municipalities with per capita spending over 2002–04 greater than population class average); capital spending should not have exceeded the 2004 value increased by 8.1 percent.
  - Anticipation effects: capital spending expected to be included from 2005, so municipalities anticipated investment project timing in 2004.

### Identification strategy
- Conceptual mechanism: incumbents can signal competence via pre-election expansionary fiscal policy (Rogoff and Sibert, 1988; Rogoff 1990). Fiscal rules raise the ex-post cost of pre-electoral fiscal expansion.
- Empirical strategy:
  - Regression-discontinuity (RD) around the 5,000 population threshold combined with diff-in-diff logic.
  - Sample restricted in baseline analysis to municipalities with less than 15,000 inhabitants to maintain homogeneous electoral rules.
  - McCrary (2008) density test at 5,000 in 2006: no evidence of manipulation of population to avoid the DSP threshold (discontinuity estimate (log-difference in height) = -0.02 (standard error 0.20)).
  - Outcome variable for political budget cycle: log-difference of pre- vs. post-election per-capita capital spending where “pre” = election year and two prior years (t = −2, −1, 0) and “post” = two years after (t = 1, 2).

### Regression-discontinuity specifications and covariates
- Baseline RD specification includes:
  - Fifth-degree polynomials in normalized population (order p), interactions with treatment indicator and electoral dummy (pre-election years).
  - Treatment indicator ܼ equals one if municipality subject to DSP (>5,000) and zero otherwise.
  - Covariates include mayor characteristics (gender, age, education, party affiliation, political color), total per-capita transfers received, proportion aged 15-64, taxable per-capita income.
  - Municipality fixed effects and year effects included.
- Robustness checks: different polynomial orders and windows; results generally robust but lose significance when polynomial order below three or population window above 3,000 inhabitants.

### Key empirical results (RD-FE estimates, 2004–2006)
- Selected coefficients at the 5,000 threshold (fifth-degree polynomial; windows 0–15,000 and 4,000–6,000):
  - >5000: (1) 0.070 (0.112); (2) 0.051 (0.105); (3) 0.383* (0.203); (4) 0.272 (0.205)
  - Pre-election years: (1) 0.365*** (0.090); (2) 0.348*** (0.085); (3) 0.525*** (0.186); (4) 0.473*** (0.182)
  - Pre-election years x >5000: (1) -0.238** (0.114); (2) -0.182* (0.109); (3) -0.721*** (0.276); (4) -0.631** (0.289)
- Interpretation:
  - Capital spending is 36 percent higher in pre-electoral years (baseline).
  - The DSP reduces electoral (pre-election) capital expenditure by more than 60 percent relative to municipalities not subject to the rule (interaction term magnitude and sign).
- Sample sizes and fit:
  - R-squared range: 0.062 to 0.188 across columns.
  - Municipalities: 5,155 (columns 1–2) and 620 (columns 3–4).
  - Observations: 14,848; 14,149; 1,700; 1,627 (respectively).

### Robustness on mayor wage thresholds
- Tests at population thresholds relevant for mayor wage increases (1,000; 3,000; 5,000; 10,000):
  - No significant pre-election interaction effects at 1,000, 3,000, 10,000 thresholds.
  - Significant effect at 5,000 threshold consistent with DSP effect (replicates Table 2 findings).
- Selected coefficients (5,000 threshold):
  - >threshold (5,000): (5) 0.383* (0.203); (6) 0.272 (0.205)
  - Pre-election years (5,000): (5) 0.525*** (0.186); (6) 0.473*** (0.182)
  - Pre-election years x >threshold (5,000): (5) -0.721*** (0.276); (6) -0.631** (0.289)
- Conclusion: changes in mayor wage at other thresholds (including a 50 percent wage increase at 3,000) do not show significant effects on the political budget cycle, reducing concern that mayor wage discontinuities drive the main result.

### Evidence on enforcement of the DSP
- Directly-collected lists of municipalities that did not comply with the DSP (Interior Ministry, 2004–06) allow testing of enforcement.
- Empirical evidence:
  - Hiring and long-term borrowing were remarkably lower for non-complying municipalities in the year following a breach relative to complying municipalities.
  - Current spending evidence is consistent though less striking.
- Interpretation: breaching the DSP carried penalties in terms of fiscal aggregates, supporting the premise that the rule was enforced in the sample period.

### Balance and validity checks
- Covariate continuity around 5,000 threshold (Figure 7):
  - Mayor gender, age, education, party affiliation, per-capita transfers, taxable income, and proportion aged 15-64 show no evident discontinuities.
- McCrary density test: discontinuity estimate -0.02 (standard error 0.20) — no evidence of population manipulation around 5,000 threshold.

### Conclusions and policy implication
- Main conclusion:
  - The political budget cycle increases real capital spending by about 35 percent on average in the years prior to municipal elections.
  - The sub-national fiscal rule (DSP) reduces these figures by about two-thirds relative to municipalities not subject to the rule.
- Additional points:
  - Evidence suggests DSP penalties were enforced in 2004–06.
  - This is presented as the first robust evidence that fiscal rules can limit the political budget cycle.
  - Welfare implications:
    - While fiscal rules that contain deficits may or may not be welfare improving, limiting the political budget cycle addresses an inefficiency (electoral-motivated distortion).
    - In the Italian case, reductions in deficits were achieved mainly by compressing capital spending; evaluating welfare requires weighing deficit reductions and reduced political-cycle distortions against costs of lower capital spending.

*Source: IMF Working Paper — “Political Budget Cycle in log-Capital Spending of Municipalities at Population Thresholds Relevant for Mayor's Wage. RD-FE Estimates (2004-2006)” (content unit wp1706).*

### References  _______________________________________________________________________________________  18

### References

### Figures
- 1. Log-Capital Spending per-capita by Electoral Year and Size of Municipality ____________________9
- 2. Average Hiring of Municipalities (per 1,000 Inhabitants) by DSP Compliance in Previous Year (2005-2007) __________________________________________________________________________________9
- 3. Mean per-Capita Long-Term Borrowing (Accrual) of Municipalities by DSP Compliance in Previous Year (2005-2007) _________________________________________________________________ 10
- 4. Mean per-Capita Purchase of Goods and Services (Cash) of Municipalities by DSP Compliance in Previous Year (2005-2007) _______________________________________________________________  10
- 5. Checking Continuity of the Population Distribution Around the 5,000 Inhabitant Threshold __ 11
- 6. Difference in pre- vs. post- Electoral Log-Capital Spending per capita ________________________  12
- 7. Checking Continuity of Covariates Around the 5,000 Inhabitant Threshold. Population below 15,000 (2004-2006) ________________________________________________________________________ 14

### Tables
- 1. Summary Statistics (2004–2006) _________________________________________________________________7
- 2. Political Budget Cycle in log-Capital Spending of Municipalities at the DSP Threshold. RD-FE Estimates (2004-2006)  _____________________________________________________________________  13
- 3. Legislative Thresholds of Municipalities (2004-2006) __________________________________________  15

*Source: wp1706 - References*

### 4. Political Budget Cycle in log-Capital Spending of Municipalities at Population Thresholds Relevant

### 4. Political Budget Cycle in log-Capital Spending of Municipalities at Population Thresholds Relevant for Mayor's Wage. RD-FE Estimates (2004-2006)

### Introduction and main claim
- Fiscal rules can help moderate the political budget cycle, defined as increases in government spending or the deficit, or decreases in taxes, in an election year or pre-election years driven by incumbents’ re-election incentives.
- Context: Italian municipalities in the early 2000s under the Domestic Stability Pact (DSP) introduced in 1999.
- Identification lever: municipalities below 5,000 inhabitants are exempt from the DSP; municipalities above 5,000 are subject to the rule.
- Core quantitative findings:
  - The political budget cycle increases real capital spending by about 35 percent on average in the three years prior to municipal elections.
  - The sub-national fiscal rule reduces these figures by about two thirds.

### Institutional setup (DSP and local electoral rules)
- Domestic Stability Pact (DSP):
  - Introduced in 1999 to include sub-national authorities in meeting European fiscal targets.
  - Operational target for municipalities changed over time (limits to spending in 2005–06; overall balance prior to 2005 and from 2007 onward).
  - Penalties for non-compliance: limits on hiring, on spending, and on borrowing for investments.
  - Since 2001, municipalities with less than 5,000 residents have been exempted from the DSP.
  - The DSP was discontinued in 2015 and replaced with a budget balance rule for all local authorities.
- Municipal governance and elections:
  - Decision bodies: mayor (Sindaco), Executive committee (Giunta comunale), municipal council (Consiglio comunale).
  - Electoral systems:
    - Municipalities with less than 15,000 inhabitants: simple plurality, each candidate supported by a single list.
    - Over 15,000: multiple lists possible and run-off if no absolute majority.
  - Since 1993 municipal elections every four years; since 2000 mayoral term extended to 5 years (with exceptions).
  - Elections usually held in May and June.

### Data (period, coverage, key aggregates)
- Analysis focuses on capital spending; period limited to before 2007. Main dataset covers 2004–2006.
- Data combined: municipal budgets, municipal election dates, mayor characteristics (age, education, gender, party).
- Aggregate and financing facts (2004 baseline year and dataset descriptions):
  - Total spending of municipalities in ordinary-statute regions represented almost 5 percent of GDP in 2004.
  - Capital spending represented about 38 percent of total spending.
  - In real per capita terms, municipalities spent about €600 annually in investment.
  - Transfers from regions and the central government over the period represented about 40 percent of overall revenues; own revenues covered the rest.
  - Main municipal taxes: ICI (real estate tax on home property) provided about 43 percent of municipal tax revenues; IRPEF surcharge amounted to about 6 percent of municipal tax revenues.
- Notes on fiscal-rule operational details in the analysis window:
  - 2004: difference between current spending and current revenues could not be higher in real terms than in 2003.
  - 2005: current and capital spending should have been lower than the average over 2001–03 increased by 10 percent.
  - 2006: current spending should have been lower by 6.5 percent with respect to 2004 (by 8.1 percent for municipalities with per capita spending over 2002–04 greater than population class average); capital spending should not have exceeded the 2004 value increased by 8.1 percent.
  - Anticipation effects: capital spending expected to be included from 2005, so municipalities anticipated investment project timing in 2004.

### Identification strategy
- Conceptual mechanism: incumbents can signal competence via pre-election expansionary fiscal policy (Rogoff and Sibert, 1988; Rogoff 1990). Fiscal rules raise the ex-post cost of pre-electoral fiscal expansion.
- Empirical strategy:
  - Regression-discontinuity (RD) around the 5,000 population threshold combined with diff-in-diff logic.
  - Sample restricted in baseline analysis to municipalities with less than 15,000 inhabitants to maintain homogeneous electoral rules.
  - McCrary (2008) density test at 5,000 in 2006: no evidence of manipulation of population to avoid the DSP threshold (discontinuity estimate (log-difference in height) = -0.02 (standard error 0.20)).
  - Outcome variable for political budget cycle: log-difference of pre- vs. post-election per-capita capital spending where “pre” = election year and two prior years (t = −2, −1, 0) and “post” = two years after (t = 1, 2).

### Regression-discontinuity specifications and covariates
- Baseline RD specification (see paper for full equation) includes:
  - Fifth-degree polynomials in normalized population (order p), interactions with treatment indicator (>5,000) and electoral dummy (pre-election years).
  - Treatment indicator ܼ equals one if municipality subject to DSP (>5,000) and zero otherwise.
  - Covariates include mayor characteristics (gender, age, education, party affiliation, political color), total per-capita transfers received, proportion aged 15-64, taxable per-capita income.
  - Municipality fixed effects and year effects included.
- Robustness checks: different polynomial orders and windows; results generally robust but lose significance when polynomial order below three or population window above 3,000 inhabitants.

### Key empirical results (RD-FE estimates, 2004–2006)
- Table 2 (RD-FE estimates at the 5,000 threshold; fifth-degree polynomial; windows 0–15,000 and 4,000–6,000):
  - Coefficients at the 5,000 threshold (selected estimates reported in table):
    - >5000: (1) 0.070 (0.112); (2) 0.051 (0.105); (3) 0.383* (0.203); (4) 0.272 (0.205)
    - Pre-election years: (1) 0.365*** (0.090); (2) 0.348*** (0.085); (3) 0.525*** (0.186); (4) 0.473*** (0.182)
    - Pre-election years x >5000: (1) -0.238** (0.114); (2) -0.182* (0.109); (3) -0.721*** (0.276); (4) -0.631** (0.289)
  - Interpretation:
    - Capital spending is 36 percent higher in pre-electoral years (baseline).
    - The DSP reduces electoral (pre-election) capital expenditure by more than 60 percent relative to municipalities not subject to the rule (interaction term magnitude and sign).
  - Sample sizes and fit:
    - R-squared range: 0.062 to 0.188 across columns.
    - Municipalities: 5,155 (columns 1–2) and 620 (columns 3–4).
    - Observations: 14,848; 14,149; 1,700; 1,627 (respectively).
- Additional robustness on mayor wage thresholds (Table 4):
  - Tests at population thresholds relevant for mayor wage increases (1,000; 3,000; 5,000; 10,000) show:
    - No significant pre-election interaction effects at 1,000, 3,000, 10,000 thresholds.
    - Significant effect at 5,000 threshold consistent with DSP effect (replicates Table 2 findings).
  - Selected coefficients from Table 4 (columns aligned with thresholds, other covariates No/Yes):
    - >threshold (5,000): (5) 0.383* (0.203); (6) 0.272 (0.205)
    - Pre-election years (5,000): (5) 0.525*** (0.186); (6) 0.473*** (0.182)
    - Pre-election years x >threshold (5,000): (5) -0.721*** (0.276); (6) -0.631** (0.289)
  - Conclusion: changes in mayor wage at other thresholds (including a 50 percent wage increase at 3,000) do not show significant effects on the political budget cycle, reducing concern that mayor wage discontinuities drive the main result.

### Evidence on enforcement of the DSP
- Directly-collected lists of municipalities that did not comply with the DSP (Interior Ministry, 2004–06) allow testing of enforcement.
- Empirical evidence (Figures referenced in paper):
  - Hiring and long-term borrowing were remarkably lower for non-complying municipalities in the year following a breach relative to complying municipalities.
  - Current spending evidence is consistent though less striking.
- Interpretation: breaching the DSP carried penalties in terms of fiscal aggregates, supporting the premise that the rule was enforced in the sample period.

### Balance and validity checks
- Covariate continuity around 5,000 threshold (Figure 7):
  - Mayor gender, age, education, party affiliation, per-capita transfers, taxable income, and proportion aged 15-64 show no evident discontinuities.
- McCrary density test: discontinuity estimate -0.02 (standard error 0.20) — no evidence of population manipulation around 5,000 threshold.

### Conclusions and policy implication
- Main conclusion:
  - The political budget cycle increases real capital spending by about 35 percent on average in the years prior to municipal elections.
  - The sub-national fiscal rule (DSP) reduces these figures by about two-thirds relative to municipalities not subject to the rule.
- Additional points:
  - Evidence suggests DSP penalties were enforced in 2004–06.
  - This is presented as the first robust evidence that fiscal rules can limit the political budget cycle.
  - Welfare implications:
    - While fiscal rules that contain deficits may or may not be welfare improving, limiting the political budget cycle addresses an inefficiency (electoral-motivated distortion).
    - In the Italian case, reductions in deficits were achieved mainly by compressing capital spending; evaluating welfare requires weighing deficit reductions and reduced political-cycle distortions against costs of lower capital spending.

*Source: IMF Working Paper — “Political Budget Cycle in log-Capital Spending of Municipalities at Population Thresholds Relevant for Mayor's Wage. RD-FE Estimates (2004-2006)” (content unit wp1706).*

### REFERENCES

### REFERENCES

### Cited works

- Acconcia, A., Corsetti G., and Simonelli, S. (2014). Mafia and Public Spending: Evidence on the Fiscal Multiplier from a Quasi-experiment, American Economic Review 104(7), 2185-2209.  
- Alesina, A. (1987). Macroeconomic policy in a two party system as a repeated game. Quarterly Journal of Economics 102(3), 651-678.  
- Alesina, A. (1988). Credibility and Policy Convergence in a Two-Party System with Rational Voters. The American Economic Review, 78(4), 796–805.  
- Alesina, A., and Paradisi, M. (2014). Political budget cycles: evidence from Italian cities, NBER Working Paper 20570.  
- Alesina, A., Cohen G.D. and Roubini, N. (1997). Political Cycles and the Macroeconomy, The MIT Press, Cambridge (MA, United States) and London (UK).  
- Alesina, A., Troiano, U., and Cassidy, T. (2015). Old and Young Politicians, NBER Working Papers 20977, National Bureau of Economic Research, Inc.  
- Beetsma, R.M.W.J., and Debrun, X. (2004). Reconciling Stability and Growth: Smart Pacts and Structural Reforms, IMF Staff Papers 51/3, International Monetary Fund.  
- Beetsma, R.M.W.J., and Debrun, X. (2007). The new stability and growth pact: A first assessment, European Economic Review, 51(2), 453-477.  
- Brender, A., and Drazen, A. (2005). Political budget cycles in new versus established democracies, Journal of Monetary Economics 52(7), 1271-1295.  
- Brender, A., and Drazen, A. (2008). How Do Budget Deficits and Economic Growth Affect Reelection Prospects? Evidence from a Large Panel of Countries, American Economic Review 98(5), 2203-20.  
- Brollo, F. and Nannicini T. (2012). Tying Your Enemy’s Hands in Close Races: The Politics of Federal Transfers in Brazil, American Political Science Review 106(4), 742-761.  
- Calonico, S., Cattaneo, M.D., and Titiunik, R. (2014). Robust data-driven inference in regression-discontinuity design, Stata Journal 14(4), 909-46.  
- Chiades, P., and Mengotto, V. (2013). Il calo degli investimenti nei Comuni tra Patto di stabilità interno e carenza di risorse, Bank of Italy Occasional papers 121.  
- Cioffi, M., Messina, G., and Tommasino, P. (2012). Parties, institutions and political budget cycles at the municipal level, Temi di discussione (Economic working papers) 885, Bank of Italy.  
- Coelho, C., Veiga, F.J., and Veiga L.G. (2006). Political business cycles in local employment: Evidence from Portugal, Economics Letters 93, 82-87.  
- Debrun, X., Moulin, L., Turrini, A., Ayuso-i-Casals, J., and Kumar, M.S. (2008). Tied to the mast? National fiscal rules in the European Union, Economic Policy 23, 297-362.  
- Drazen, A., and Eslava, M. (2010). Electoral manipulation via voter friendly spending: Theory and evidence, Journal of Development Economics 92(1), 39-52.  
- Ferraresi M., Rizzo, L., and Zanardi, A. (2015). Policy outcomes of single and double-ballot elections. International Tax and Public Finance, 22(6), 977-998.  
- Foremny, D., and Riedel, N. (2014). Business taxes and the electoral cycle, Journal of Public Economics, 115, 48-61.  
- Gagliarducci S., and Nannicini, T. (2013). Do Better Paid Politicians Perform Better? Disentangling Incentives from Selection, Journal of the European Economic Association 11(2), 369-398.  
- Grembi, V., Nannicini, T., Troiano, U. (2016). Do fiscal rules matter?, American Economic Journal: Applied Economics (forthcoming).  
- Gupta, S., Mulas-Granados, C., and Liu, E. (2015). Now or Later? The Political Economy of Public Investment in Democracies, IMF Working paper 155.  
- Lee, D., and Lemieux, T. (2010). Regression Discontinuity Designs in Economics, Journal of Economic Literature 48, 281-355.  
- Lindbeck, A. (1976). Stabilization Policies in Open Economies with Endogenous Politicians, American Economic Review Papers and Proceedings 1-19.  
- McCrary, J. (2008). Manipulation of the Running Variable in the Regression Discontinuity Design: A Density Test, Journal of Econometrics 142(2), 698-714.  
- Milesi-Ferretti, G.M., Perotti, R., and Rostagno, M. (2002). Electoral Systems and Public Spending, Quarterly Journal of Economics 117 (2): 609-657.  
- Nordhaus, W. (1975). The political business cycle, Review of Economic Studies 42, 169-190.  
- Persson, T., and Tabellini, G. (2000). Political Economics: Explaining Economic Policy, MIT Press, Cambridge and London.  
- Petterson-Lidbom, P. (2008). Do Parties Matter for Economic Outcomes? A Regression-Discontinuity Approach, Journal of the European Economic Association, 6(5), 1037-1056.  
- Rogoff, K. (1990). Equilibrium political budget cycles, American Economic Review 80(1), 21-36.  
- Rogoff, K. and Sibert, A. (1988). Elections and Macroeconomic Policy Cycles. The Review of Economic Studies, 55.  
- Shi, M., and Svensson, J. (2006). Political budget cycles: Do they differ across countries and why?, Journal of Public Economics 90(8-9), 1367-1389.  
- Veiga, L., and Veiga, F. (2007). Political business cycles at the municipal level, Public Choice 131(1), 45-64.

*Source: wp1706 - REFERENCES (PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/wp1706.pdf_