Technical efficiency analysis correcting for biases from observed and unobserved variables: an application to a natural resource management project

Abstract

This article brings together the stochastic frontier framework with impact evaluation methodology to compare technical efficiency (TE) across treatment and control groups using cross-sectional data associated with the MARENA Program in Honduras. A matched group of beneficiaries and control farmers is determined using propensity score matching techniques to mitigate biases stemming from observed variables. In addition, possible self-selection arising from unobserved variables is addressed using a selectivity correction model for stochastic frontiers recently introduced by Greene (J Prod Anal 34:15-24, 2010). The results reveal that average TE is consistently higher for beneficiary farmers than the control group while the presence of selectivity bias cannot be rejected. TE ranges from 0.67 to 0.75 for beneficiaries and from 0.40 to 0.65 for the control depending on whether biases were controlled or not. The TE gap between beneficiaries and control farmers decreases by implementing the matching technique and the sample selection framework decreases this gap even further. The analysis also suggests that beneficiaries do not only exhibit higher TE but also higher frontier output.