Principal investigator: Viviane Pires Ribeiro
Article title: THE SHORT-TERM IMPACT OF UNCONDITIONAL CASH TRANSFERS TO THE POOR: EXPERIMENTAL EVIDENCE FROM KENYA
Article authors: Johannes Haushofer and Jeremy Shapiro
Location of the intervention: Kenya
Sample size: 120 villages
Main theme: Economic Policy and Governance
Type of Intervention: Impact of unconditional cash transfers on poverty reduction
Main Variable of Interest: Unconditional money transfer
Evaluation method: : Randomized controlled trial
Evaluation Context
GiveDirectly (GD is an international NGO founded in 2009, whose mission is to make unconditional cash transfers to low-income families in developing countries. GD began operations in Kenya in 2011. At the time of the study conducted by Haushofer and Shapiro (2016), eligibility for the program was living in a house with a thatched roof (instead of metal). Families were identified through a census conducted with the help of the village elder. After identification, families were visited by a GD representative, who asked to speak with the recipient of the transfer in order to collect some demographic data, informing them that they would receive a transfer of KES 25,200 ($404 PPP). A Safaricom SIM card was provided to the recipient to register for the M-Pesa app (M for mobile, and Pesa for money) installed on the mobile phone chip. For lump-sum recipients, an initial transfer of KES 1.200 ($19 PPP) was sent in the first month following the GD visit as an incentive to register.
Intervention Details
Haushofer and Shapiro (2016) study the impacts of unconditional cash transfers on low-income families in rural Kenya, analyzing economic and psychological outcomes. The research was conducted with the NGO GiveDirectlyDuring the years 2011 to 2013, families with a primary female and a primary male member were randomly assigned the same probability of the recipient being male or female. In all treatment families, the transfer was randomly assigned to be delivered as a lump sum or as a series of nine monthly installments. Specifically, 258 of the 503 treatment families were assigned to the monthly condition and 245 to the lump sum condition. However, the analysis only considered the 173 monthly and 193 lump sum recipients who did not receive large transfers. The total amount of each type of transfer was KES 25,200 ($404 PPP).
To study the magnitude of the transfer, 137 families from the treatment group were randomly selected and informed in January 2012 that they would receive an additional transfer of KES 70.000 ($1.121 PPP), paid in seven monthly installments of KES 10.000 ($160 PPP) each, beginning in February 2012. Thus, the transfers previously allocated to these families, whether monthly or a fixed amount, were increased by KES 10.000 from February 2012 to August 2012, and therefore the total transfer amount received by these families was KES 95,200 ($1.525 PPP).
Methodology Details
The methodology adopted by Haushofer and Shapiro (2016) was a randomized controlled trial. The authors performed a two-stage randomization, one at the village level, resulting in treatment villages and control villages, and the other at the household level, resulting in treatment families and control families. Furthermore, within the treatment group, the authors randomized the recipient of the transfer within the household (wife versus husband), the transfer period (monthly installments over nine months versus a fixed amount), and the magnitude of the transfer ($404 PPP versus $1.525 PPP).
Results
Nine months after the start of the program, the authors found a strong consumption response to the transfers, with an increase in monthly consumption from US$158 PPP to US$193 PPP. The program's effects on alcohol and tobacco spending were negative and insignificant. Regarding assets and durable goods (metal roofing), there was a significant increase in investment compared to a control group. According to the researchers, these investments translate into an increase in monthly income from agriculture, livestock farming, and businesses of US$16 PPP compared to a control group.
Transfers did not have major effects on health and educational outcomes. However, they did have a considerable effect on psychological well-being. The authors documented an increase in the standard deviation of happiness of 0,16, an increase in the standard deviation of life satisfaction of 0,17, a reduction in the standard deviation of stress of 0,26, and a significant reduction in depression (all measured by psychological questionnaires), but found no overall effect on cortisol stress hormone levels, although differences existed in some subgroups.
Lessons in Public Policy
The analysis conducted by Haushofer and Shapiro (2016) provides useful policy guidance for governments and other entities that adopt different forms of income redistribution. The results suggest that when policymakers consider the implications of different design choices for unconditional cash transfers, they may arrive at different welfare conclusions.
The three treatment types (recipient's gender, timing, and magnitude of the transfer) adopted by the authors allow us to speculate on the impacts of unconditional cash transfers on low-income households in rural Kenya. For large transfers, the outcomes are desirable across most measures, including asset holdings, consumption, food security, psychological well-being, and women's empowerment (although there are no major effects on health and education). Monthly transfers outperform lump-sum transfers in terms of their effects on food security, while lump-sum transfers show greater effects than monthly transfers on asset holdings. When the transfer recipient is the woman of the household, the effects of the transfer related to women's empowerment and psychological well-being are greater than when transfers are made to the man.
References
HAUSHOFER, Johannes; SHAPIRO, Jeremy. The short-term impact of unconditional cash transfers to the poor: experimental evidence from Kenya. The Quarterly Journal of Economics, v. 131, no. 4, p. 1973-2042, 2016.
