Production function based on 2017 agricultural census microdata

Abstract

In this paper we present the main results of the analyzes carried out in the IBGE restricted data access room based on the 2017 agricultural census microdata. A stochastic production frontier was fit to a sample of rural establishments investigated by the census. The information from the 2017 Brazilian agricultural census was explored to assess the statistical significance of the production factors’ elasticities and of other contextual variables (related to market imperfections) that affect production directly and indirectly through the technical efficiency component. A Cobb-Douglas representation was used for the frontier. Technological inputs have the highest relative elasticity, followed by labor and land. The Murphy-Topel correction was used for the variance-covariance matrix. The contextual factors that significantly and positively affect production efficiency are association to cooperatives, financing and the probability of positive net income; the practice of organic agriculture and/or livestock farming negatively affects production efficiency.