Abstract
This article examines the presence of political cycles inside the Portuguese governments’ aggregate expenditures by using annual data for 10 expenditure components. The results indicate that the choice of the expenditure components to be increased during election periods by Portuguese governments generally relates to more visible items such as general public services, social protection and health care.
Acknowledgements
The authors acknowledge helpful comments and suggestions from the participants of NIPE’s Seminar, University of Minho, Portugal; the 2014 Meeting of the European Public Choice Society; and the 8th Annual Meeting of the Portuguese Economic Journal.
Notes
1 The first difference of the dependent variable is used because the panel unit root tests are not clear regarding its stationarity in level. The Fisher unit root tests based on the Phillips–Perron tests do not reject the null hypothesis that ‘all panels contain unit roots’ for the CompExpd variable, but reject it when its first difference is considered. The results of those tests are available upon request.
2 The Blundell and Bond (Citation1998) estimator is used as the initial estimator where the instruments are collapsed, which avoids using invalid or too many instruments. The results do not qualitatively change with more repetitions or when other estimators are chosen initially. Those results are available upon request.
3 As the sum of all shares of the components to the government total expenditures is equal to 1, one of the components was excluded from the regression analysis.
4 In this case, DPol equals 1 when a left-wing government is in office.
5 As here the sum of the growth rates do not add up to 1 or 100%, we do not need to exclude any category from the estimations, but we keep social protection as the basis category.