The link between DEA efficiency and return to assets

The data envelopment analysis (DEA) is a standard tool in the analysis of determinants of the efficiency of economic agents. A two-stage efficiency analysis consisting of estimating DEA efficiency and then using it as dependent variable in regression with various determinants, is often used for this purpose. In this article, we focus on the relevance of this approach using empirical agricultural data. In particular, we show that the lagged DEA efficiency of agricultural companies is not correlated with return on assets indicator, and, similarly, lagged return on assets are not correlated with DEA efficiency. Low cross-correlation values indicate that a two-stage analysis using the DEA method can lead to misleading results.