Poverty measurement and the targeting of social protection programs

How to cite

OConnell, Stephen; Altindag, Onur; Balcioglu, Zeynep; Sasmaz, Aytug. 2022. "Poverty measurement and the targeting of social protection programs." AEA RCT Registry. AEARCTR-0009725. https://doi.org/10.1257/rct.9725-1.2

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Stephen OConnell* , Onur Altindag , Zeynep Balcioglu , Aytug Sasmaz

* Primary investigator

Registry ID
AEARCTR-0009725
DOI
https://doi.org/10.1257/rct.9725-1.2
First registered
Last updated
Status
Completed
Trial dates
2021-05-01 – 2023-05-31
Countries
Lebanon ()
JEL codes
I38, I32, O12, D74
Registry record
http://www.socialscienceregistry.org/trials/9725

Abstract

In targeted antipoverty programs, the definition of deprivation determines who is in need, who receives program benefits, and, through them, program effects. Deprivation can be experienced in a wide range of dimensions, however, leading to tradeoffs in the determination of who is poor. We pre-specify the analysis of a nationwide field experiment in which we estimate the effects of targeting unconditional cash transfers to alternative measures of deprivation. To do this, we partnered with leading humanitarian agencies in Lebanon to experimentally vary the poverty measure used to target an at-scale year-long structured cash assistance program for refugees. Targeting among similar populations with the same amount of resources, we will estimate the program effect resulting from each of the alternative poverty targets across a range of economic well-being indicators. We further pre-specify the protocol for qualitative data collection and quantitative data analysis aimed at understanding the mechanisms that lead to differences in targeting effects.

Intervention

We target a scaled cash transfer program to alternative measures of deprivation and quantify effects on measures of household well-being.

Experimental design

Starting in 2021, we partnered with UN agencies in Lebanon to experimentally vary the targeting of their cash-based assistance program to alternative measures of deprivation. The poverty measures that form the basis for the targeting in our four treatment arms are: (i) monetary poverty, as measured by per capita expenditure, (ii) severe or moderate food insecurity, as measured by the reduced Coping Strategies Index (rCSI), (iii) having poor or borderline food consumption, as measured by the Food Consumption Score (FCS), and (iv) being moderately or severely deprived in a Multidimensional Deprivation Index (MDI). For the 2021-22 assistance cycle, the full population of Syrian refugee households were assigned to one of the targeting treatment arms by lottery. Each treatment arm received 25 percent of total program resources to distribute to a fixed number of households deemed eligible in that arm.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Households/cases
Clusters
n/a
Observations
6,500
Arms
>1,000 complier observations per treatment arm; >4,000 in primary survey sample
Primary outcomes
Poverty measures (4), child well-being, living standards and livelihoods
Secondary outcomes
property rights, social support networks, productive assets, savings behavior

Cited by

No citation data available for this registration yet.

Keywords