Social protection programs offering income support are common anti-poverty tools in developing countries. Although these programs vary across contexts, they often take the form of either unconditional cash transfers, or government-funded employment schemes that require beneficiaries to work. A recent IFPRI project note summarizes a study authored by Kate Ambler and Susan Godlonton that compares the two approaches. The paper describes a field experiment in Malawi that examined the differential short-term effects of unconditional cash transfers and cash transfers with a work requirement on beneficiary’s expenditure and labor supply decisions. The experiment was conducted in partnership with Innovations Poverty Action (IPA) in 2015 – 2016 and the National Smallholder Farmers’ Association of Malawi (NASFAM) as part of the wider Fomento project.
The research team organized 33 day-long events in locations convenient for the members of the sample. A total of 352 participants (78 percent of the contact sample) attended. Events ranged in size from one person to 33 people, and the average event size was 10.7. Events were conducted from late January 2016 to late February 2016. When invited to the event, participants were told that they would receive a payment, but not the amount.
All participants who attended the event were paid 2,000 MWK for attendance and 8,000 MWK additional for completing the day’s activities. Participants were randomized into two groups: those for whom the activities consisted of waiting to complete a baseline survey, and those who were assigned work (classifying newspaper articles into themes and then transcribing those articles). Both groups spent roughly the same amount of time at the event, and in both cases the cash payment was one-time and unanticipated. The only difference between the two groups was the work requirement. Following the events, the study team collected information on participants’ expenditures and labor supply decisions, once a week for the next three weeks via a series of phone surveys.
The results of the study indicate that participants who were required to work for the cash transfer did not spend it differently than those who were not required to work. The study did, however, find that those who were required to work spent more overall in the first week following the events (but not in the subsequent weeks). When examining labor supply, the study found that those who were required to work, work significantly less in household economic activities, both agricultural and non-agricultural, in the week following the event. The authors hypothesize that participants in the work requirement group may have been spending more time both looking for and doing wage labor than the unconditional cash transfer group. They may have used their payment to fund further job search, as evidenced by their higher overall expenditures, and only returned to household activities when they were unsuccessful in locating work. This interpretation is, however, speculative and needs to be verified by future research.
Find the project note here. The associated working paper can be found here.
Authors
Kate Ambler is a Research Fellow at IFPRI’s Markets, Trade and Institutions Division and Susan Godlonton is an Assistant Professor at the Williams College Economics Department, December 2019.
Note
Please note that the Fomento project also involved another study with the Federation of Non-Governmental Organizations of Senegal (FONGS) in Senegal.
Further reading
IFPRI Malawi blog: Learning from Brazil: Adapting “Fomento” for Malawi, September 2014.
