Ensaios em avaliação de políticas públicas de infraestrutura rodoviária e segurança pública no estado da Paraíba

• Autor(a) principal: Anjos Junior, Otoniel Rodrigues dos
• Data de Publicação: 2021
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFPB
• Texto Completo: https://repositorio.ufpb.br/jspui/handle/123456789/22705
• topic: Políticas públicas; Eficiência técnica; Infraestrutura rodoviária; Caminhos da Paraíba; Paraíba unida pela paz; Public policy; Technical efficiency; Road infrastructure; CNPQ::CIENCIAS SOCIAIS APLICADAS::ECONOMIA

This Thesis consists of two essays aimed at evaluating public policies implemented in the State of Paraíba in the last decade. Chapter 2, corresponding to the first essay, discusses the effects of investments in road infrastructure on the local development of Paraíba’s municipalities. For this purpose, the Difference-in Difference method reweighted by the weights of entropy balancing was employed. The database was obtained from the Departamento de Estradas de Rodagens (DER) of Paraíba, covering the period from 2011 to 2017. The results show the hypothesis of a positive relationship between investments in road infrastructure and economic growth cannot be rejected. The most significant impacts are observed on the employment balance of the benefited municipalities, especially in those that received paving works and were removed from isolation. In this case, the municipalities created, on average, approximately 87 new job openings in the formal job market. In the municipalities benefiting from restoration and supplementation, there was an increase in the employment balance of 82 and 85 vacancies, respectively. In addition, there was growth in real GDP (13%) and in GVA service sector (16%), both restricted to the municipalities that received supplementation works (duplication, viaducts, accesses, clovers, contours, among others). In turn, Chapter 3, referring to the second essay, aims to analyze the technical efficiency of the "Paraíba Unida pela Paz" program from a set of selected inputs and outputs. For this purpose, the MDEA-S method and the DBSCAN Machine Learning algorithm were adopted. The database used consists of the record of 429.699 police cases from the years 2017, 2018 and 2019, and were obtained from the Secretaria de Estado da Segurança e da Defesa Social do Estado da Paraíba (SESDSPB). The results show that about 40% of the units covered by the "Prêmio Paraíba Unida pela Paz" should not have received the bonus. When simulating the adjustment by the efficiency scores of the MDEA-S method, it was found that it would have been possible to have saved approximately 58% of the resources invested in the program between 2017 and 2019. Together, situations in which the policeman had received the premium, but should have not, with the situations in which he should have receive, but did not, inform the total error in the payment of the premium: 46,04% of the situations. Therefore, there are indications that the criteria employed by the program fail to distinguish fairly and technically who should be considered for the prize and who should not. Furthermore, the AISPs’ efficiency scores are very concentrated in the lower efficiency classes, mainly up to the value of 0,600. It was also evident that the efficiency of AISPs could be improved by reducing the number of police officers (20,43%) and the number of vehicles (26,27%). The results of the DBSCAN, on the other hand, allowed the identification of crimes hotspots that revealed that João Pessoa’s downtown region stands out for having a much higher rate of car theft and robbery than that registered in other locations. In addition, the night time (from 18:00 to 23:59) showed greater prominence, with 1.640 occurrences recorded at this time. The largest observed hotspot is located in the region comprised by the neighborhoods of Ernesto Geisel, Costa e Silva, Funcionários, Grotão and João Paulo II (228 observations or 17,56%), while the smallest is located in the Cabo Branco region (21 points or 1,28 %).