Escolha sob incerteza, política e desenvolvimento : uma análise da produtividade do trigo na região sul

Abstract

Ever since the classic works in economic science, Smith and Ricardo proposed that in the agricultural activity, as well as in the way each nation embraces it, lie key answers to the understanding of the differences in productivity and income between countries. This research aims to evaluate and compare the different probabilities of gains for wheat producers based over four different micro-regions in Southern Brazil, as well as assessing determinants to disparities in the observed productivity over each area. The research begins examining contributions related to the role of agriculture on the development of an economy, the importance of state policy in the process according to Development and Austrian School economists, the usage of risk analysis tools by producers and models that quantify the impact of technological change over the agricultural productivity. This research then brings two quantitative analyses: the first one by simulating scenarios with Monte Carlo method, measuring the different magnitudes of risks that pose to wheat producers in Rio Grande do Sul and Paraná; the second makes use of cross section data from the brazilian agricultural census of 2006 to assess determinants to the regional differences in productivity. Both regions present disparities in average farm productivity, even though their geographic characteristics are very close, being part of the same climatic zone. Given that regional constraint on data selection, the effect of unobserved weather-related variables is minimized. The simulation through the Monte Carlo method considered independent variables with differing probability distributions. The results allow us to conclude that the probability of gains is greater in the state of Rio Grande do Sul than in Paraná, in response to higher prices paid to the producer and lower costs. At the same time, productivity measured in tons per hectare is higher in Paraná – albeit with a greater variance as well. The cross section analysis then identifies proxy variables for the use of technology, education of the producer, average property size and the impact of non-agricultural activities as determinants to the productivity differences. The first two variables have had the greater positive impact over tons harvested per hectare, while the latter was shown to have negative impact.