POLICY DIALOGUES Social Protection: The Impact on Inequality in Ethiopia, India, and Peru October 2020 – No. 1 | EU-AFD Research Facility on Inequalities
MESSAGE
to the debate. ‘Horizontal inequality’
households’ assets and amenities,
Three large-scale social protection
between ethnic groups has major socio-economic implications,
including materials used for housing
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 – contribute to reduced consumption inequality between and within socially and culturally constructed groups. New research that analyzes data from Young Lives, finds a strong correlation between living standards and the coverage of these social protection programs, suggesting that the schemes are generally well targeted at the most needy. Further analysis shows that within-group inequality accounts for a large proportion of total inequality. CONTEXT & MOTIVATION A growing body of research is documenting the effects of public works programs like NREGS on labor market responses, risk-sharing networks, household consumption or risk management. Similarly, inequalities along ethnic, gender, and other social and cultural lines, and their determinants are also relevant
potentially leading to conflicts and
construction and access to services.
under-provision of public goods.
The study focuses on five dimensions of horizontal inequality that capture
Few social protection policies evaluations however have used a
the most relevant aspects of between-group inequality: gender,
distributive lens, maybe due to the fact that social protection programs
ethnicity, region, place of residence, and religion. These group types
are, by design, targeted at the poor. This restricts findings from
constitute a relevant subset of all potential circumstances available in
distributional analysis to the lower
the data. Ethnicity is at the centre of Ethiopia’s political structure under the
end of the consumption distribution. This study fills some evidence gap by evaluating the role of social protection programs on horizontal inequality. It examines changes in
country's ethnic federalism system. In India, the caste system is still very important in various social, cultural, and political spheres.
consumption inequality between and within culturally and socially defined
Different measures of inequality are estimated to determine whether
groups in Ethiopia, India, and Peru over the period from 2006 to 2016,
there is unequal access to basic services across different groups
analyzing the impact of participation in PSNP, NREGS, and Juntos.1
(‘between-group’) and across members of each group (‘within-
METHODS
group’). Decomposing inequality by components can help to shape
The data used come from four survey waves tracking the lives of young people between 2006 and 2016. The outcome variables used are household consumption data and a wealth
index
composed
of
economic policies that will be effective in reducing inequality and poverty. The researchers consider different approaches and decompose overall inequality into betweenand within-group components.
Authors Liyousew BORGA, Raphaël COTTIN, Conchita
Key words Social protection; Horizontal inequality;
D’AMBROSIO.
Decomposition analysis.
Geography Ethiopia; India; Peru
Themes Economics, Inequality, Finance.
Find out more about this project: https://www.afd.fr/en/carte-des-projets/investigating-distributional-impact-social protection-short-and-longer-run-inequalities
RESULTS
Trends in consumption inequality by year
The study documents the evolution of between-group and within-group inequality among ethnic, religious, gender, and regional groupings. There is a strong correlation between living standards and social protection indicating
program positive
coverage, targeting
efficiency. Rough calculations indicate that the ‘exclusion error’ (the proportion of people eligible for a social protection program who are excluded) is high in the sample. This is partly due to the oversampling of poor households in the survey design. At the same time, the ‘inclusion error’ (the proportion of people receiving transfers despite not being poor) is very low in the sample. In India, for example, the coverage rate is favorable to disenfranchised
the historically groups of
‘scheduled castes’ and ‘scheduled tribes’. Inequality among program participants is also much lower than among non-participants. A decomposition of total inequality shows that within-group inequality consistently accounts for a larger share of total inequality for all groups considered. But the lower contribution
of
between-group
inequality
does
redistribution disadvantaged
not
imply
that
in favor groups is
of not
implying the need for dedicated policy initiatives marginalized
in
support of communities.
important. The policy significance of this finding should be qualified with a
Restrictive targeting of Ethiopia’s program appears to be one factor
closer examination of the patterns of inequality, the policy instruments at
limiting the ability of PNSP to play a
hand, their impacts, and their costs.
larger role in reducing inequality.
The results from decompositions of
The decomposition exercise quantifies the contribution to
consumption gaps between groups show that the social safety nets of
inequality of a set of factors such as household demographics (the
these programs did not lead to an automatic reduction in the average
education level of the head, the proportion of children), access to
differences between groups.
basic services (electricity, sanitation) contributing towards the explained
In India, factors other than the program are far more powerful in driving differences between groups,
part of within-group inequality.
RECOMMENDATIONS
Evaluation of local inequality based on socially and culturally constructed boundaries should inform policymakers on the potential of social protection programs for outreach to marginalized groups.
Despite the success of public works programs in Ethiopia and India in fighting extreme poverty, it is important to note that they have only had a modest impact in reducing inequality.
In Ethiopia, differences in participation in PSNP explain only a minor part of the total consumption gap due to the restrictiveness of program targeting. Horizontal expansion of the outreach of the program might be warranted.
Addressing inequality within and between groups through policy calls for a closer examination of the patterns of inequality, the policy instruments at hand, their impacts, and their costs.
1
Borga L., Cottin, R. & D’Ambrosio, C. (2020). Social Protection and Inequality: Evidence from Ethiopia, India and Peru, AFD Research Papers N°139.
Publishing Director Rémy Rioux Editor-in-Chief Thomas Melonio
Agence française de développement (AFD) 5, rue Roland Barthes | 75012 PARIS | France
For other publications of the Policy dialogues collection: https://www.afd.fr/en/collection/policy-dialogues
Legal deposit 4th quarter 2020 ISSN in process | © AFD