É dando que se recebe? Evidência empírica do efeito da despesa de pessoal para o sucesso eleitoral do executivo estadual brasileiro

• Main Author: Torres, Rayza de Lima
• Publication Date: 2018
• Format: Master thesis
• Language: por
• Source: Biblioteca Digital de Teses e Dissertações da UFPB
• Download full: https://repositorio.ufpb.br/jspui/handle/123456789/13202
• topic: Ciclos políticos; Despesas de pessoal; Reeleição; Political cycles; Personnel expenditures; Re-election; CNPQ::CIENCIAS HUMANAS::CIENCIA POLITICA

In the brazillian states, personnel expenditures reflect a greater absorption of responsabilties with the provision of public good and services, removing this burden from the federal jurisdiction. On the other hand, the expansion of civil service expenditures, even contradicting the norms of the Administrative Reform or limits of the Law of Responsibility, may be associated with the political cycles formation, which leads the governors to increase public outlays or manipulate the fiscal variables during the election period, in order to succeed at the polls. In this regard, this work aimed to mesure how the personnel expenditure could influence the reelection chances of the state governor or of the political party in power, in the period between 1996 and 2014. To carry out this work, fiscal (health expenses, education), political (political parties, election year) and socioeconomic (GDP per capita) data were used as control variables. Concerning the variable of interest, a proxy was proposed in which the personnel expenditures correspond to the active servers personnel payroll of the state and Federal District administrations. The Maximum Likelihood (logit) probabilistic method was chosen for the hypothesis test, and the Hausman and Wald tests, for a better adjustment of the econometric model. The results confirmed that the personnel expenditure positively influenced the chances of state reinstatement in the period evaluated (1996-2014), that is, The growth of the personnel sheet increases the probability of re-election of the governor / political group by 37%. The coefficients absolute variation of GDP per capita and absolute variation of education per capita, corresponded to the expected signal (positive). Regarding absolute health variation per capita, this control, besides not being statistically significant, showed a sign of the causal relationship different from that expected. In this way, it can be concluded that the expansion of personnel expenditures in the states was interesting in electoral terms.