An Incentive to Learn

In global development circles, it’s long been understood that providing money to poor households when they do things like send children to school or get children vaccinated can fight poverty and meet broader social goals in the short term. But neither educators nor economists have known much about whether these conditional cash transfer (CCT) programs can make a difference over time in high-poverty communities.

Real Results

Now, some evidence is emerging that they can. Building upon his past analysis of CCT programs in Bogota, Colombia, Harvard Graduate School of Education economist Felipe Barrera-Osorio has found that these programs — depending on their structure — can increase educational attainment for as many as eight years after the incentives were provided.

Tracking long-term outcomes for enrollees in his original study, which began in 2005 and was first reported in 2011, Barrera-Osorio has now found improvements years after families’ participation in the CCT programs had ended. He found that the programs — which had been designed to spur secondary enrollment — significantly increased the tertiary enrollment of low-income students, the first study to document such an effect.

Structural Changes

CCT programs are used around the world — in both the developing world and in wealthy nations — to reduce poverty and encourage a variety of goals, including school attendance and achievement. They’ve been the subject of considerable attention from researchers and policymakers, but most of the assessment thus far has focused on short-term impacts. By drawing on his earlier study, which implemented and compared varyingly structured CCT programs, Barrera-Osorio’s new work provides rare evidence about long-term educational effects of CCTs. The findings are detailed in a paper currently under review for publication.

Both the short- and long-term analyses support one of Barrera-Osorio’s key experimental innovations — that changing the payment structure of the CCT can help policymakers generate more significant results. He compared a standard CCT model — providing a fixed bimonthly transfer conditional on secondary school enrollment and continued attendance — with one that forces families to save almost one-third of the stipend each month until the time at which they have to make enrollment decisions for the following academic year. Even though the total amount of the stipend didn’t change, the structural change — withholding some of the money — sharply boosted enrollment.

In a brief mention of the work in a recent piece on the World Bank’s new behavioral approaches to global development, The Economist took note of how such small policy tweaks — rooted in observation of how people actually behave, rather than in an expectation that they’ll behave rationally — may change the way development policy is implemented and assessed.

And yet more research is needed to analyze the effectiveness of these and similar educational incentives, Barrera-Osorio says. “These programs are very important in the developing world, but there are still open questions,” he says. “The effects on achievement from these programs are quite mixed; some programs rendered positive results, others null results. So the question is, why this differential effect? Does it depend on the complementarity of the supply of education? Does it depend solely on the effort of the families?”

***

Get Usable Knowledge — Delivered
Our free monthly newsletter sends you tips, tools, and ideas from research and practice leaders at the Harvard Graduate School of Education. Sign up now.