Eficiência das usinas de açúcar e álcool de cana no Brasil: uma aplicação da análise envoltória de dados (DEA)

Abstract

The sugar and alcohol sector historically has a relevant role in the Brazilian economy as it contributes to food production and is part of the Brazilian energy matrix. It is currently the largest producer of sugarcane and in agribusiness is side by side with soybeans in terms of representativeness. This study aimed to analyze the technical efficiency of the sector through the methodology known as DEA, which consists of a linear programming technique known as data envelopment analysis (Data Envelopment Analysis), initially developed by Charnes, Cooper and Rhodes (1978). The mills were then grouped according to their location in the five geographical regions and also grouped by size of mills following the UNICA classification. Concomitant to the technical efficiency analysis to study the scale efficiency analysis, compararando to the size of the plants to see if the results are within what is expected according to the literature. Analysis parameters of each group were studied their respective averages of efficiency, which were compared to average global efficiency (Brazil) and applied the statistical t-test mean differences to give robustness analysis. The study results show that the South has an average lower efficiency compared to the global average efficiencies. In relation to the scale, small plants that operate in the range of growing and steady returns, midsize have plants in all the returns to scale ranges and large ones with plants only in the range of diminishing returns and is as expected for the same. Therefore, it is concluded that the size of the plant showed a tendency to efficiency of the same and, with respect to location, average efficiency differences were observed for most of the regions analyzed, compared with the average efficiency for Brazil.