Four essays about the link between improvements of urban transports and criminality in big cities : the case of Bogota.

Summary This thesis seeks to study and determine the impact of public transport improvements on the crime pattern in a large city like Bogotá. In a first step, this research seeks to specify the number of jobs reachable by the city's inhabitants in three different time intervals. The effective size of the labor market is thus defined in order to determine whether the inhabitants of the different areas that make up the city have the same degree of accessibility to jobs. Once the accessibility is defined, this study makes a comparative study in relation to the socio-economic characteristics of the inhabitants per area as well as in relation to the presence or lack of transport system improvements in each area that make up the city of Bogotá. The presence of transport system improvements is defined by the passage of the transport system called Transmilenio (TM) in each zone. In order to deepen on the link between the socio-economic characteristics of the inhabitants and the presence of TM in each zone of the city, we focus, in a second step, on the existence of a possible endogenous relationship of the presence of TM in the concentration of jobs and high incomes in Bogotá. The objective of this analysis is to determine whether public transport improvements have a causal relationship in the location of jobs and high incomes in each area of the city. We then adopt the hypothesis that criminals prefer to commit their crimes in employment areas and in areas with a high concentration of high incomes (independently of the location of residential areas). In the same vein, and in order to establish a causal relationship on the evolution of five different types of crime in each area of the city, this study performs an ex-ante and ex-post analysis of the implementation of Transmilenio. Due to the weakness of the data base for the different periods, the results of this step are not homogeneous and this makes them unconvincing. However, they give us a relevant approximation of the impact that Transmilenio can have in the configuration of crime in the city. These results lead us to make, in a final step, a causality analysis for the period for which the available data are complete for all areas of the city and do not show reliability problems. After this stage of analysis, the results are convincing. They suggest that there is indeed a causal relationship of the presence of Transmilenio on the evolution of three of the five types of crimes subject to our study. The results also identify a clear spatial dependence of the concentration of crimes in the city. Thus, it seems that, despite the multiple positive effects that improved public transport can have for the inhabitants of a city, it can also stimulate an increase in certain types of crime in the areas served by Transmilenio. Despite the limitations of this study that will need to be resolved in future research, the results obtained, as well as the way the topic is approached, represent an innovative analytical perspective for a better understanding of the possible negative consequences that can thwart the objectives of urban transport policies in large cities. We believe that this thesis contributes to the complementary nature of studies on the effects of urban transport.