Combining Longitudinal Household and Community Surveys for Evaluation of Social Transfers: infrastructure rehabilitation projects in rural Georgia

Abstract

This paper combines longitudinal household and community level survey data to evaluate the effect of infrastructure rehabilitation projects on household well-being in rural Georgia. The panel structure of the data is utilized in an empirical approach to control for time-invariant unobservable factors at the community level by applying propensity score-matched double difference comparison. The results indicate that improvements in school and road infrastructure produce non-trivial gains on village and country levels. School rehabilitation projects produce the largest gains for the poor, while the road projects benefit the poor and non-poor in different aspects of well- being. From a methodological point of departure it is concluded that ad hoc community surveys matched with ongoing nationally representative longitudinal household surveys can provide a feasible and low-cost tool for evaluation of the effectiveness of social transfers.