• logo linkedin
  • logo email
Investigating the distributional impact of social  protection: short- and longer-run inequalities in Ethiopia, India and Peru
The project aims to establish the causal impact of three large-scale social-protection schemes in India, Ethiopia and Peru on both the level of economic resources and a number of dimensions of inequalities. The researchers involved in this project use a novel longitudinal dataset to carry out comparative analysis of these phenomena and draw conclusions that are relevant for other countries with similar circumstances.
Context

Social-protection schemes have become a popular form of intervention in developing countries both by governments and within the international-development community, as they are seen as a tool to combat the adverse impacts of natural and economic crises. However, the empirical evidence on the effectiveness of these programs remains mixed. Notwithstanding the wealth of interest from policy makers, donors, and researchers, there is a paucity of evidence about the distributional incidence of these programs. Existing work has often not been able to establish longer-run effects of public-works programs on poverty and inequality. In addition, there is a lack of focus on horizontal and spatial inequalities. This research project intends to fill this evidence gap.


This project is part of a European facility for a research program on inequalities in developing and emerging countries which is coordinated by the AFD. Financed by the Development Cooperation Instrument of the European Union, this facility enables to implement 20 research projects over the period 2017-2020, in partnership with donors and research centers from the South to the North.


 

Goals

This research project proposes to investigate the distributional impact of three large-scale social-protection schemes - the Productive Safety Net Program (PSNP) in Ethiopia, the Mahatma Gandhi National Rural Employment Guarantee Scheme (NREGS) in India, and the Juntos conditional cash-transfer program in Peru. These programs were chosen because they are large-scale projects involving a coordinated effort by governments, donors, local authorities and individual households. The programs cover three countries with diverse social, cultural, political, and economic contexts from which to draw lessons for future policy initiatives.

The researchers propose to go beyond measuring the direct intended impacts of the programs and rather focus on their indirect, and not necessarily intended, consequences. In particular, they first consider the effect of these programs on the income and wealth of program participant households over both the shorter- and longer-run. They then turn to the programs’ impact on horizontal and spatial inequalities, as well as their effect on household social relations and the distribution of resources within households.

Method

Unlike traditional benefit-incidence studies, the researchers exploit policy differences across time and space, both within and across the countries, to provide causal estimates of the social-policy impacts. They use the Young Lives cohort study that collected data both pre-and post-program implementation between 2002 and 2014 in all three countries. They exploit a number of aspects of the Young Lives cohort study and the roll-out of the social-protection schemes in each country to produce estimates that deal with non-random program placement.

The researchers planned to begin by conducting an individual-specific pre- and post-program comparison of income and wealth. Then to exploit the staggered rollout of the social-protection programs across districts to causally identify the impact of the schemes on a set of indicators. They intend to compare changes in districts that received the program earlier to changes in districts that received the program later in a difference-in-differences approach. Last but not least, they will further assess the impact of the program on individual outcomes, taking advantage of administrative boundaries to separate treated and control areas in a geographic regression discontinuity.

Results
Teachings
01/12/2018
Project start date
01/05/2020
Project end date
Ethiopia, India, Peru
Location
125 000
Financing amount
Ongoing
Status

Contact:
Conchita d’Ambrosio, Professor of Economics, Université du Luxembourg
Cecilia Poggi, Research Officer, AFD
 

Le contenu de cette fiche projet relève de la seule responsabilité de l’AFD et ne reflète pas nécessairement les opinions de l’Union européenne.