Evolução do crédito consignado no estado do Rio Grande do Sul : aplicação da metodologia de Box & Jenkins

Abstract

After the implementation of the Real Plan, with the change brought about by inflationary stabilization, Brazil presented an increase in the supply of bank credit, as a strategy for banks to maintain the level of their revenues. However, the country still needs to expand its credit supply, as it has a credit-to-GDP ratio of 62.2%, lower than the average for all countries, of 131.7% (WORLD BANK, 2017). The Brazilian financial system is predominantly banking, and the provision of credit comes mainly from banking institutions. The resources obtained through bank credit return to society, in the form of consumption and investments, contributing to economic activity. In addition to this social aspect, credit lending makes banks profitable, impacting the financial performance of these institutions. Following the enactment of the law that allowed for the payroll deductible loan, which is characterized by the debit of the loan installments directly on the payroll of the debtors, the banks expanded the credit offer, since this type of loan made possible a reduction in defaults, reducing the credit risk. In view of the above, the objective of the study was to find a model that represents the behavior of the evolution of payroll loans with a financial institution, assisting in its strategy by predicting this variable. The study was conducted through a quantitative research, with secondary data provided by the financial institution. The statistical treatment of the data was performed through the analysis of time series, using the methodology proposed by Box & Jenkins, using the software R Studio. The result was an ARIMA model (0, 2, 1), which represented the behavior of the evolution of payroll loans with the financial institution, with analysis of 96 monthly observations, from January 2009 to December 2016. With the model, the forecast for six months, from January to June 2017. The maximum difference between the expected value and the actual value of the variable was 1.57%, which attests to the accuracy of the model found. Therefore, as a theoretical result, it was evidenced that the ARIMA model was adequate for the prediction of payroll loans granted by the financial institution. As an empirical result, the predictive model found can act as a strategic tool for the financial institution, since its application can contribute to the process of determining competitive strategies.