Segregation in Agriculture:
How Important is it Explaining Gender Productivity Differences?

 Arturo Aguilar                      Niklas Buehren                     Markus Goldstein
 Economics, ITAM                    World Bank                           World Bank           

Abstract

Traditional decomposition methods have been recently employed to give evidence of gender productivity differentials in agriculture. However, those methods typically oversee the problem that segregation might cause when accounting for gender differences. In particular, segregation could mean that no comparable individuals of the opposite gender with similar characteristics could be realistically conceived. This is known in the decomposition methods’ literature as the common-support assumption. In this paper we employ a methodology proposed by Ñopo (2008) and data from the LSMS project to show that when relaxing the common support assumption, the segregation component could account for up to 46% of the overall gender differential. Moreover, characterizing profiles of male and female farmers might be relevant from a policy perspective to be able to design more adequate and efficient programs. We explore the use cluster analysis for such purpose.