Three studies on spatial dependence and interaction based on Brazilian data

• Grant number: 18/00043-5
• Support Opportunities: Scholarships abroad - Research
• Start date: July 01, 2018
• End date: March 31, 2019
• Field of knowledge: Applied Social Sciences - Economics - Regional and Urban Economics
• Principal Investigator: André Luis Squarize Chagas
• Grantee: André Luis Squarize Chagas
• Host Investigator: Sandy Dallerba
• Host Institution: Faculdade de Economia, Administração e Contabilidade (FEA). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Institution abroad: University of Illinois at Urbana-Champaign, United States

Abstract

This project aims to implement applications of spatial econometrics to three different fronts, within a single line of research: the application of spatial econometric methods to economic and social problems. The first study refers to the effects of the implantation of the bus and the bicycle lanes in São Paulo on real estate. The theoretical and empirical literature indicates that policies of this type have a positive impact on housing prices. However, these studies are scarce in Brazil, and the application of spatial econometrics and longitudinal data is appropriate. Specifically, I will implement a method of difference-in-differences with spatial data that I had recently developed. In the second study, I will deepen previous work of mine on the economic and social impacts of sugarcane production, focusing on the so-called Indirect Land Use Change (ILUC). I will deal with the deforestation caused indirectly if sugar cane displaces other crops to forest areas. Most ILUC measurement models do not include spatial controls. The spatial dependence is characteristic of the process of indirect land use change, with locational factors influencing the spatial distribution of the raw material employed in the activities potentially displaced by biofuels. In the ILUC case, this dependence is even more important, due to the inertial aspect of deforestation - that is, it tends to occur in areas close to areas already deforested - and diffusive. The third study will present novel evidence on environmental offenses in Brazil. The strategy of IBAMA, the official Brazilian agency, to deter violations is based on large operations and on decapitalizing offenders to signal its will to monitor and enforce the law. I deal with the following questions: do the sanctions applied by IBAMA, especially fines, deter actual and potential offenders? Are there any spatial or temporal patterns affecting violations? I will use data on offenses against flora and applied fines for Brazilian municipalities between 1998 and 2015. I will contribute to the existing literature by incorporating spatial controls in a dynamic panel. I will develop and apply a Spatial Limited Information Maximum Likelihood (SLIML) estimator that accounts for the endogeneity of sanction charges to estimate the panel models. (AU)