81319
                  enGender Impact: The World Bank’s Gender Impact Evaluation Database



Estimating the Benefit Incidence of an Antipoverty Program by Propensity Score Matching

      Author(s)            Jyotsna Jalan and Martin Ravallion
      Contact              jjalan@isid.ac.in

      Country              Argentina

  Organizing Theme         Economic Opportunities and Access to Assets

       Status              Completed

Intervention Category      Cash Transfer

       Sector              Social Protection
                           We apply recent advances in propensity-score matching (PSM) to the problem of
                           estimating the distribution of net income gains from an Argentinean workfare program.
                           PSM has a number of attractive features in this context, including the need to allow for
                           heterogeneous impacts, while optimally weighting observed characteristics when forming a
      Abstract
                           comparison group. The average direct gain to the participant is found to be about half the
                           gross wage. Over half of the beneficiaries are in the poorest decile nationally, and 80% are
                           in the poorest quintile. Our PSM estimator is reasonably robust to a number of changes in
                           methodology.
 Gender Connection         Gender Informed Analysis

  Gender Outcomes          Gender disaggregated income

     IE Design             Propensity Score Matching
                           The program provided short-term work at relatively low wages in attempt to self-selected
                           unemployed workers from poor families. Additionally, the scheme attempted to locate
    Intervention
                           socially useful projects in poor areas to help repair local infrastructure. The wage is set at
                           maximum of $200 per month.
 Intervention Period       1997-1998
                           The study focuses on current Trabajar participants in the reference week, fixed at the first
  Sample population
                           week of September 1997. The total participants in the sample is 2802.
                           The study uses Propensity Scores to match participants with non-participants who have
Comparison conditions
                           similar observable characteristics
   Unit of analysis        Individual Level

  Evaluation Period        1997-1998
                           Participants are more likely to be poor than non-participants. Participants are less well-
                           educated, live in poorer neighborhoods and are more likely to be members of
       Results
                           neighborhood associations and political parties. Average gains are similar between men and
                           women, but are higher for younger workers.


   Last updated: 14 August 2013                                                                         1
                   enGender Impact: The World Bank’s Gender Impact Evaluation Database


                            The accuracy of the PSM methodology will depend on the similarity in the economic
Primary study limitations
                            conditions of treatment and control groups.
    Funding Source          The World Bank Research Committee.
                            Jalan, J., & Ravallion, M. (2003). Estimating the benefit incidence of an antipoverty
      Reference(s)          program by propensity-score matching. Journal of Business & Economic Statistics, 21(1),
                            19-30.
    Link to Studies
                            http://www.tandfonline.com.libproxy-wb.imf.org/doi/pdf/10.1198/073500102288618720
       Microdata




    Last updated: 14 August 2013                                                                    2