The political hourglass: opportunistic behavior in local government policy decisions

Figures & data

Table 1. Description of the variables.

Variable	Description	Source
Contracting out (dependent variable)	Dichotomous variable that takes the value 1 when the service is contracted out and the value 0 otherwise	Official Bulletin of Province (BOP)
Public-private partnership (dependent variable)	Dichotomous variable that takes the value 1 when the service is PPP and the value 0 otherwise	Official Bulletin of the Province (BOP)
Public externalization (dependent variable)	Dichotomous variable that takes the value 1 when the service is public externalization and the value 0 otherwise	Official Bulletin of the Province (BOP)
Election	Dichotomous variable that takes the value 1 in the pre-electoral years and the value 0 in the post-electoral ones	Derived by the authors
Capital spending-Postelectoral years	Interaction between capital spending and postelectoral years	Derived by the authors from data published by the General Secretariat for Local and Regional Coordination
Current spending-Preelectoral years	Interaction between current spending and preelectoral years	Derived by the authors from data published by the General Secretariat for Local and Regional Coordination
Political ideology	Dichotomous variable that takes the value 1 when the party in government has a conservative ideology and the value 0 when it has a progressive one	Ministry of the Interior
Conservative absolute majority	Dichotomous variable that takes the value 1 when a conservative party governs with an absolute majority and the value 0 when it governs in coalition or when another party governs.	Derived by the authors
Progressive absolute majority	Dichotomous variable that takes the value 1 when a progressive party governs with an absolute majority and the value 0 when it governs in coalition or when another party governs.	Derived by the authors
Change of mayor	Dichotomous variable that takes the value 1 when a different mayor takes office and the value 0 when the same mayor continues in office	Derived by the authors using data supplied by the Ministry of the Interior
Change of mayor-Postelectoral years	Interaction between political change and postelectoral years	Derived by the authors
Gender	Dichotomous variable that takes the value 1 when a woman serves as mayor and the value 0 when the mayor is male	Derived by the authors using data supplied by the Ministry of the Interior
Tax pressure	Tax revenues (Chapter 3 of the spending budget) divided by the number of inhabitants	Derived by the authors
Tax pressure-Preelectoral years	Interaction of tax pressure and preelectoral years	Derived by the authors
Mandatory tax 1 (Property)	Local tax on property (real estate)	General Secretariat for Local and Regional Coordination
Mandatory tax 2 (Economic activity)	Local tax on economic activity	General Secretariat for Local and Regional Coordination
Mandatory tax 3 (Vehicles)	Local tax on vehicles	General Secretariat for Local and Regional Coordination
Population	Ln of municipal inhabitants	National Institute of Statistics (INE)

Source: Compiled by the authors.

Table 2. Features of delivery forms of public services.

Delivery forms / Features	Private Sector participation	Risk transfer	Responsibility	Citizens’ perception
Contracting out	Total	Total	Low	Negative
PPP	Partial	Partial	Shared	Negative
Public externalization	None	None	Total	Positive

Source: Compiled by the authors.

Table 3. Correlations matrix.

	Elect.	Inter. 1	Inter. 2	Progr. Ideology	Inter. 3	Inter. 4	Political change	Inter. 5	Gender	Tax pressure	Inter. 6	Mandat. tax 1	Mandat. Tax 2	Mandat. tax 3	Populat.
Election	1
Inter. 1	−.502*	1
Inter. 2	−.363*	−.407*	1
Progr. Ideology	.003	.008	−.004	1
Inter. 3	.21*	−.007	−.02	−.012	1
Inter. 4	.006	−.01	−.002	.003	−.274*	1
Political change	.422	−.190*	−.175*	.020*	−.055*	−.077*	1
Inter. 5	−.058*	.114*	−.070*	.056*	−.063*	−.044*	.356*	1
Gender	.006	.013	−.017	−.002	−.051*	−.052*	.027*	.003	1
Tax pressure	−.031*	.005	.030*	.004	.009	−.031*	−.006	.005	−.009	1
Inter. 6	−.172*	−.289*	.477*	−.002	−.010	−.015	−.080*	−.033*	−.010*	.633*	1
Mandatory tax 1	−.088*	.132*	−.081*	.009	−.065*	.073	−.03*	.041*	−.006	.355*	.088*	1
Mandatory tax 2	−.024*	.022*	−.002	.009	−.041*	.023*	−.011	.003	−.008	.403*	.153*	.378*	1
Mandatory tax 3	−.025*	.003	.01	.006	−.053*	.042*	−.013	−.002	−.011	.426*	.171*	.315*	.547	1
Population	−.001	.035*	.024*	.000	.033*	−.055*	−.008	−.003	.024*	.016	.006	−.019*	.062*	.031*	1

*= p<.01.

Table 4. Results.

Variables	Model 1	Model 2	Model 3
	Contracting Out	Public-Private Partnership	Public Externalization
	Odds ratio^1	Odds ratio^1	Odds ratio^1
Election	.019***(.010)	.219**(.606)	8.801*(11.585)
Capital Spending-Postelect.	.776***(.029)	.922**(.161)	1.220**(.107)
Current Spending-Preelect.	.704***(.024)	.917***(.168)	1.181**(.098)
Progressive ideology	.978(.091)	.572(.246)	.895(0.188)
Absolute conservative majority	1.101*(.135)	1.122**(.109)	.897(.238)
Absolute progressive majority	.687***(.077)	1.950(1.154)	1.724**(.369)
Political change	.845(.134)	.374(.425)	.750(.332)
Political change-Postelect. years	.889(.296)	1.330(.002)	.666(.590)
Gender	1.361**(.170)	.990(.654)	.454**(.140)
Tax pressure	.994***(.001)	.994*(.003)	.999(.001)
Tax pressure-Preelectoral years	1.010(.001)	.999(.003)	.999(.001)
Mandatory tax 1 (property)	.430***(.130)	.639(.570)	1.622*(.460)
Mandatory tax 2 (economic activity)	1.728***(.295)	2.022(1.428)	1.236(.299)
Mandatory tax 3 (vehicles)	1.031***(.004)	1.018(.019)	1.006(.007)
Population	1.000***(.000)	.999(.000)	.999***(.000)
Test of Rho	.8437***	.5811***	.7131***
Hausman test	7.67 (.000)	1.40 (.000)	1.47 (.000)

***= p < 0.01; **= 0.01<p < 0.05; *= 0.05<p < 0.1 Standard errors are in parentheses.