O comportamento dos custos de produção do café arábica em relação a venda de máquinas agrícolas no estado de Minas Gerais

Abstract

Brazil stands out on the world stage with regard to the marketing of agricultural products, ranking third in the world rankings when it comes to exports, according to the World Trade Organization - WTO (2011). The agricultural sector is the second-level employment in the country, according to the Brazilian Institute publishes Economic Geography - IBGE (2011), which shows the importance of this economic sector for the country. In this sector, there to emphasize the coffee culture in the country is a leader in production and ranks second worldwide in the item consumption. The state of Minas Gerais is responsible for approximately 50% of total production of coffee, with emphasis on type Arabica coffee. Although increasing the introduction of agricultural machinery, in search of lower costs of production and, consequently, improved results of the project, the consensus among researchers that there is no ideal level of mechanization. Thus, to decide mechanize the production process, or other change in cultural tract, it is prudent to study in order to analyze which best cost and benefit. Given the above, this study seeks to answer the questions: What has been the trend in costs of production of coffee (\"size of the pie\") in relation to the increased use of farm machinery? What is the relationship of increased introduction of agricultural machinery and various cost elements (\"slices of cake\")? This research used data relating to the costs of production of Arabica coffee, released by the National Supply Company - CONAB - and restated by the IGP-DI. The years were analyzed from 2003 to 2010, relative to cities Guaxupé, Sponsorship and São Sebastião do Paraíso. The data were extracted from the mechanization yearbook of the National Association of Automobile Manufacturers - ANFAVEA - that was adopted as a proxy of mechanization, the number of wheeled tractors sold in the state of Minas Gerais. After reviewing the estimates of mean, standard deviation and variance, we used the Shapiro-Wilk test for normality of the data. We calculated the Pearson correlation coefficient for the data whose distribution was seen as normal and not normal for the Spearman correlation coefficient, noting the significance level. Upon obtaining the linear regression equation, used for checking the level of dependence between variables, we calculated the coefficient of determination (R2) in order to measure the explanatory power of the regression model adopted. Finally, research has used the technique of canonical correlation, separating variable cost by criteria CONAB, and associating the variable amount of these tractors sold. The result showed that all groups showed good levels of correlation, and a descriptive level below the nominal 0.05, and thus proving the influence between the sale of tractors and the cost elements. The group formed by variable costs stood out due to its higher correlation with variable selling tractors. After the results of the analysis of normal errors, the coefficient of determination and descriptive level, presenting considerably positive, one can associate the behavior of the variables studied by means of regression models used. It was evident that although research, lead to believe that the introduction of machines have focused on the reduction of costs in the cities analyzed, there was an increase of total costs.