COMPARING METHODS FOR TOTAL FACTOR PRODUCTIVITY MEASUREMENT

Abstract

Productivity measures the level of efficiency, which a particular economy uses its resources to produce goods and consumer services. Increasing productivity is the fastest route to get economic growth and social well-being since such production improvement reflects all effectiveness of the production sector, as well as the degree of development of a company. This article aims to compare four models to estimate the Total Factor Productivity (TFP). The tested models were the Olley and Pakes - OP (1996); Levinsohn and Petrin – LP (2003); Wooldridge - Wool (2009); and Ackerberg, Caves and Frazer - ACF (2015). As an intermediate input, we used the per capita energy consumption. We found that the model ACF (2015) proposes an improvement of the OP and LP models, in addition to present results with statistical meaning, the model Wool (2009) also is about an improved model of previous estimations, beyond it presented results near the same. However, the ACF model presented high dispersion around the models average. For this reason, according to our analysis, the Wool model was preferred than others.