Abstract

Using endogenous switching regressions to calculate transitional heterogeneity which we have used to build counterfactual scenarios to identify the treatment effects for both sellers to supermarkets and sellers to traditional markets, the results support neither a sharp inclusion nor a sharp exclusion story, but rather one of “weak inclusion” with various forms of conditionality on participation, such as a threshold land size, specialization in vegetables, and irrigation assets. Participation yields net income increases of 14% after controlling for self-selection bias.

Read the article