Impacto socioeconômico da cultura da soja nos municípios brasileiros.

Abstract

The objective of the present study was to analyze the impact of soybean cultivation on the level of development and economic growth of Brazilian municipalities between the census years of 1991 and 2010, in addition to evaluating its impact on the components of socioeconomic development. To achieve the research objectives, the Propensity Score Matching (PSM) and Differences in Differences with Propensity Score (DDM) models were estimated. The municipalities were classified by Minimum Comparable Areas (MCA) in order to ensure the visualization of identical territories in each year of analysis, due to the creation of new municipalities between the time period considered. The use of pairing in the modeling aims to ensure common support in the evaluation of effects, so it is necessary to separate the AMC's between two groups: the treatment group, composed of the AMC's that did not grow soybeans in 1991 and started to produce there in 2010 and; the control, encompassing those that did not produce the grain in any of the years considered in the research. In this way, the pairing by two nearest neighbors ensured the homogenization of the observable characteristics between the groups through the calculation of the propensity score by the PSM. On the other hand, the use of DDM enabled the mitigation of biases arising from the distribution of observable characteristics, the absence of common support and reduced the selection bias for unobservable characteristics. The results showed that soybean production did not have a direct impact on the level of development, with only occasional effects on the improvement of the level of MHDI - Education (positive effect) due to the guarantee of increased infrastructure, especially in new agricultural frontiers and on the worsening of the IDHM - Health, due to the intensive use of pesticides in the production process that ends up impacting rural workers and residents of nearby locations. In the case of economic growth, its effect was positive and statistically significant, inferring that “being a soy producer” tends, on average, to intensify the per capita GDP of an AMC.