One instrument, many outcomes

J. Edward Taylor (University of California, Davis), April 2015

Some people question whether social cash transfers (SCTs) are the most efficient way to achieve specific outcomes. For example, an expanding body of evidence from randomized control trials (RCTs) in Africa shows that SCTs positively affect school attendance. But SCT programs are expensive. Are there other, more direct interventions that can accomplish the same increase in school attendance more cheaply?
This question does not make sense unless school attendance is the only outcome that we are interested in. From an economic development perspective, it isn’t. Kids’ school attendance is a critical target, but there are many others. Child health. Extreme poverty. Food security. Food diversity. Productive capacity. Investment in non-farm production activities. The willingness to wait and the willingness to take risks—two prerequisites for making productive investments.  Confidence about the future—what social epidemiologisits and economists call “hope.”

A new generation of evaluations from Africa offer overwhelming evidence that SCTs affect all these outcomes—and more.

Take Zambia’s Child Grants Program (CGP). RCTs just completed show a 36% drop in extreme poverty in the households that participated in the CGP and similarly striking impacts on food consumption and dietary diversity, spending on health and clothing, food security, young-child feeding, and morbidity. The program focuses on three-to-five-year-olds, yet it increases school attendance for older siblings.

Results also show large impacts on crop production and input use—including a 17 percentage-point increase in the share of households spending money on crop inputs (from a base of only 21%) and a 12 percentage-point increase in the share of households selling crops. The share of households with animals rose by 21 percentage points. These are huge productive impacts.

There are still other impacts, which we see only by looking beyond the beneficiary households. Simulations using a local economy-wide impact evaluation (LEWIE) reveal that the CGP creates significant spillovers. Each Zambian kwacha transferred to beneficiary households raises income in the treated village cluster by 1.79 kwacha (90% confidence interval: 1.72-1.86). By treating the beneficiary households, the program also treats the local economy. After our presentation of findings in Zambia, the CGP program director said: “One kwacha gives you 1.78. We’ll have to engage in discussions with the Minister of Finance now!”

Interestingly, most of the spillovers resulting from SCTs are found in non-beneficiary households. That is not surprising given eligibility criteria that ensure that beneficiaries are asset and labor poor, making it less likely that they own local businesses that get cash injections through purchases by other households. The new findings suggest that SCT impacts are even more far-reaching than originally thought—an important consideration in garnering political support for SCT programs. They also suggest that complementary interventions to increase the local supply response—both in beneficiary and non-beneficiary households—could enhance the spillovers from cash transfers.

In short, evidence shows that this single instrument— an SCT program – can move the needle for a myriad of critical development indicators depending on context. These diverse impacts should be explicitly considered by governments when evaluating the cost-effectiveness of cash versus other, more focused and sector-specific interventions such as fertilizer or school subsidies.

Leave a comment

Your email address will not be published. Required fields are marked *