Electoral Results and Opportunistic Policies: An Integrated Approach Toke Aidt University of Cambridge (UK) Francisco José Veiga University of Minho (Portugal) Linda Gonçalves Veiga 22/04/2019 FEMES 2009

This paper integrates two strands of literature: Summary This paper integrates two strands of literature: The literature on political business cycles (PBC), which suggests that politicians systematically manipulate economic conditions before elections. The literature on vote and popularity functions, which suggests that economic conditions systematically affect election outcomes. 22/04/2019 FEMES 2009

Summary We use Rogoff’s (1990) model of the rational PBC to derive the two-way relationship between the win-margin of the incumbent politician and the size of the opportunistic distortion of fiscal policy. This relationship is estimated, for a panel of 275 Portuguese municipalities, from 1979 to 2005, as a system of simultaneous equations. 22/04/2019 FEMES 2009

Literature Precursors: VP function: PBC: Rational PBC: Goodhart and Bhansali (1970), Mueller (1970), Kramer (1971). PBC: Nordhaus (1975), Hibbs (1977). Rational PBC: Alesina (1987), Rogoff and Sibert (1988), Rogoff (1990). 22/04/2019 FEMES 2009

Literature Many papers followed, but VP functions and opportunistic PBCs are generally investigated separately, although the theory suggests that they should be analyzed together. Exceptions: Frey and Schneider (1978), Akhmedov and Zhuravskaya (2004), Drazen and Eslava (2005). But, these studies estimate opportunistic PBC and VP equations separately. The purpose of this paper is to fill a gap in the literature by estimating these functions jointly as a system of equations. 22/04/2019 FEMES 2009

The Model We consider a simple two-period economy (t = 1, 2) populated with a continuum of citizen-voters. They care about private consumption (ct) and two types of public goods (g1,t and g2,t+1). The life-time utility function of a representative citizen-voter is: uv = c1 + ln g1,1 + θ ln g2,1 + β (c2 + ln g1,2 + θ ln g2,2) Each citizen-voter is endowed with y units of a non- storable good each period, pays the lump sum tax τt and consumes ct = y − τt. 22/04/2019 FEMES 2009

The Model Public goods are produced from tax revenues by an elected politician using a simple linear technology: g1,t + g2,t+1 = τt + εt , where εt is a stochastic competency term. Each period a citizen-voter is elected. He/she is either competent (εt = εH) or incompetent (εt = εL<εH) as a politician. The probability of being competent is 0<ρ<1. Politicians receive the ego-rent m per period in office. In addition to competence, citizen-voters also care about the ideology of their elected politician. The advantage of the incumbent at time t is at = m - svt where m and s are parameters and vt captures ideological shocks. 22/04/2019 FEMES 2009

The Model Timing of events: At the beginning of period 1, the incumbent observes his competency ε1 and decides on how to allocate resources between the two public goods (g1,1, g2,2). Voters observe the incumbent’s ideological advantage (α1) and how much is provided of the observable public good (g1,1). At the end of period 1, an election takes place where the incumbent runs against a randomly chosen challenger. At the beginning of period 2, the incumbent, if reelected, decides how much to invest in the observable public good. If the challenger is elected she observes her competency (ε2) and decides on how much to invest in the observable public good. The structure described above is a sequential game of incomplete information and the natural solution concept is Perfect Bayesian Equilibrium (PBE). 22/04/2019 FEMES 2009

The Model Main Results: Fiscal policy is distorted before the election because competent politicians need to convince rational voters that they are indeed competent. This is the Rational Political Business Cycle (RPBC). The larger the natural advantage of the incumbent is, the lower is the incentive of the competent incumbent to signal. Having a natural advantage increases the reelection chance of all types of incumbents. 22/04/2019 FEMES 2009

The Data We believe that our dataset is a promising testing ground for a study of the interrelationship between the VP function and the opportunistic political business cycle: First, we have gathered a large and detailed data set covering all Portuguese mainland municipalities (278) since 1979 to 2005. Second, the mayor is a principal decision-maker in the allocation of resources and the distribution of investment in the municipality. Third, the institutional structure of local governments and the policy instruments available are the same for all Portuguese localities. Finally, election dates are fixed and exogenous from the perspective of the local authorities, and all municipalities have elections on the same day. 22/04/2019 FEMES 2009

Empirical Specification Empirical implications: In our model, the opportunistic distortion (OD) and the win-margin of the incumbent (WM) are jointly determined at equilibrium. We can therefore write the structural form of the model as: A vote and popularity function represented by the win-margin WM = h (OD, Z) An equation for the opportunistic distortion OD = k (WM, X) 22/04/2019 FEMES 2009

Empirical Specification We measure the win-margin (WM) of the incumbent as the difference between the vote share of the mayor’s party and that of the largest opposition party. We measure the opportunistic distortion (OD) as the percentage deviation of investment expenditures from the election term average. Since RPBC are more likely to occur in budgetary items whose timing of implementation is controlled by the mayor and are visible to the electorate, we concentrate our analysis on investment expenditures. These are also the ones for which Veiga and Veiga (2007) found greater evidence of political business cycles. 22/04/2019 FEMES 2009

Results – Table 2 22/04/2019 FEMES 2009 GMM 1 2 3 3SLS 4 FIML 5 Equation (18): Win-margin Opportunistic distortion (% Deviation of Investment Expenditures from their Term Mean) .060 (2.31)** .075 (3.13)*** .080 (3.31)*** .048 (1.73)* .085 (3.41)*** Investment Expenditures (Term Mean) .003 (1.12) .006 (1.98)** .004 (1.54) .005 (1.82)* (1.28) Years Mayor -.359 (-4.49)*** -.300 (-3.73)*** -.321 (-4.12)*** -.329 (-4.05)*** -.330 (-3.88)*** Run for Re-election 8.421 (8.16)*** 8.963 (9.43)*** 8.857 (9.28)*** 8.901 (8.48)*** 8.873 (7.36)*** Win-margin in previous election .581 (16.5)*** .451 (11.9)*** 0445 (11.7)*** .442 (15.3)*** .416 (15.0)*** Government’s Party * Unemployment Rate (national) 1.979 (3.70)*** .694 (1.33) Government’s Party -13.886 (-4.01)*** -8.445 (-2.47)** -4.096 (-5.40)** -3.748 (-4.73)*** -3.397 (-4.16)*** Unemployment Rate (national) .478 (.96) .471 (.98) Municipal Employment .002 (.51) Average Real Wages (1.49) # Observations 1489 1740 Adjusted R2 .22 .18 .17 22/04/2019 FEMES 2009

Results – Table 2 22/04/2019 FEMES 2009 GMM 1 2 3 3SLS 4 FIML 5 Equation (19): Opportunistic distortion (% Deviation of Investment Expenditures from their Term Mean) Win-margin -.674 (-2.50)** -.508 (-4.42)*** -.532 (-4.62)*** -.665 (-5.45)*** -1.117 (-7.76)*** Win-margin squared .007 (1.01) Investment Expenditures (term mean) .003 (2.24)** .041 (3.11)*** .037 (2.85)*** .045 (3.81)*** .046 (4.03)*** Years Mayor 0.012 (.07) -.242 (-1.41) Run for Re-election 7.290 (2.67)*** 8.333 (3.38)*** 8.674 (3.49)*** 9.947 (3.90)*** 14.909 (5.21)*** Capital Transfers (Term Mean) -.007 (-.47) -.051 (-3.83)*** (-4.00)*** -.047 (-4.34)*** -.045 (-4.66)*** % Change in Capital Transfers (From Previous Year) .344 (11.2)*** .343 (12.4)*** .349 (15.6)*** .353 (17.8)*** % Population Over 65 Years Old -.089 (-.43) -.044 (-.23) Population Density .173 (1.30) -.022 (-.19) Right -6.768 (-3.36)*** -7.647 (-4.64)*** -7.426 (-4.52)*** -6.948 (-4.01)*** -5.268 (-3.87)*** # Observations 1489 1740 Adjusted R2 .26 .19 .17 .11 22/04/2019 FEMES 2009

Miscellaneous Constructions Results – Table 3 Total Expend. Current Expend. Capital Expend. Miscellaneous Constructions 1 2 3 4 Equation (18): Win-margin Opportunistic distortion (% Deviation of Expenditures from their Term Mean) .070 (1.68)* -.077 (-1.02) .051 (2.05)** .088 (3.24)*** … # Observations 1767 1766 1489 Adjusted R2 .19 .20 Equation (19): Opportunistic distortion (% Deviation of Expend. from their Term Mean) Win-margin -.217 (-5.11)*** -.001 (-.48) -.295 (-4.27)*** -.610 (-4.97)*** … # Observations 1767 1766 1489 Adjusted R2 .33 .26 .35 .17 22/04/2019 FEMES 2009

Conclusions Empirical results clearly support the hypothesis that opportunism pays off: Greater expenditures in the election year (when compared to the term mean or, simply in euros per capita) lead to greater vote differences between the incumbent and her main opponent. 22/04/2019 FEMES 2009

Conclusions The hypothesis that the magnitude of opportunism is inversely proportional to the estimated win-margin also receives empirical support. Thus, the opportunistic distortion is smaller when the incumbent expects to win by a comfortable margin, and is greater when the election is close or if the incumbent lags behind her main opponent. 22/04/2019 FEMES 2009