Africa / Development

The Efficacy of Cash Transfers

This summer, This American Life compared GiveDirectly, a young organization that provides unconditional cash transfers to villagers in Kenya through mobile phones, to Heifer International, a more traditional charity that provides similar villages with farm animals and training. Paul Niehaus, one of GiveDirectly’s founders, made an interesting proposal:

We would like to see organizations make the case that they think they can do more good for the poor with a dollar than the poor could do for themselves. That would be fantastic. And I think some may be able to make a convincing case. But if you go to the websites today, I don’t think you’re going to see that argument being made. Nobody even bothers.

Several new studies this year have provided evidence suggesting that straightforward cash transfers – giving sums of money to help people in impoverished countries meet their basic needs – can be an effective form of aid, particularly in sub-Saharan Africa.

The most studied kind of cash transfers are “conditional,” meaning that the money will only be given to the recipient if he or she uses it for the purpose the giver specifies. But greater questions surround transfers that are “unconditional,” giving recipients freedom to do with the money as they see fit. These unconditional transfers, while still relatively new and unstudied, have the potential to drastically change the way that developed countries approach humanitarian aid.

In the late 1990s, Mexico began a program called PROGRESA, tasked with providing conditional cash transfers to those in poverty for things like health care, education, and food. At this point, many were skeptical about the ability of the poor to prioritize effectively; many members of the Mexican cabinet worried that men would “beat up their wives, take the money and get drunk,” says the New York Times’ Tina Rosenberg.

But the families receiving the money used it mostly as it was intended. They spent more on food, illness in children decreased, and school enrollment increased. Since then, this has opened the door for similar programs in many other developing countries, and most agree that conditional cash transfers can work in the right contexts.

With unconditional cash transfers, the story is more complicated – the existing research is limited both by region and by the scope of the projects themselves. Broadly, unconditional cash transfer programs can fall into two categories. “Labeled” transfers come with encouragement for spending on a specific thing, like a child’s education or a vocational training program. These programs differ from conditional transfer programs because there is no enforcement mechanism – the money will keep coming whether or not the recipient spends it on the suggested purpose.

This year, several of these programs have performed positively in studies. Two studies were conducted in Uganda and were evaluated by teams led by Chris Blattman, a political scientist at Columbia.

One program, called the Youth Opportunities Program, gave money to groups of unemployed youth. Those groups submitted proposals for using the money, and then were given lump sums and left to spend it as they saw fit, regardless of whether it matched the suggested purpose. At the end of the program, participants saw their incomes increase by 41 percent, and 65 percent more of the recipients were practicing skilled trades than their counterparts in a control group. The other program, Women’s Income Generating Support, was similar operated, and its participants’ income increased 98 percent.

A third study, by Esther Duflo of MIT and Pascaline Dupas of Stanford, looked at a program for increasing school attendance in Morocco. This study compared labeled cash transfers directly against explicitly conditional ones, and found that both were equally effective. Innovations for Poverty Action (IPA), the NGO that implemented the study, wrote that “adding conditionality….did not increase the program’s impact on student attendance or school enrollment.”

GiveDirectly works with an even newer and less proven model — unlabeled, unconditional transfers — with the assumption that the recipient knows best what to put the money toward. While this form of giving is largely untested, we could know more about how families respond to unconditional transfers soon – researchers from MIT and IPA are currently performing a study to test GiveDirectly’s effectiveness in Kenya.

Though there is mounting evidence that unconditional cash transfers can work, they are still controversial. For one thing, researchers hypothesized in the aforementioned studies that the programs would increase social cohesiveness and lessen conflict, but did not find any evidence of this. Participants became richer and more skilled, but some argue that efforts to help the poor should be held to a higher standard. Sure, people have more money, but other things matter. Are they getting along better? Are they happier?

If not, is calling these programs successful a tautology? “One common test of anti-poverty interventions is whether they make people richer,” writes Rosenberg, “So giving money will always succeed – by definition.”

In a column called “The Benefits of Cash Without Conditions,” Rosenberg also makes the case that part of the controversy comes from deeper philosophical divides. “Those on the left tend to believe that the differences come from giant structural problems,” she writes. “Bad or no education, health, transport, housing, few jobs.” Meanwhile, she argues that those on the right believe “that the poor are poor because of the culture of poverty: people make bad choices, lack of discipline, look for short-term gratification.”

This, of course, harkens back to the Mexican cabinet members’ fear that men would misuse the money. That didn’t occur in that particular case, but in the case of transfers that are completely unconditional, it is easy to understand how others might have similar concerns.

Regardless, it is clear that many in the non-profit world have been trending towards a more hands-off approach to giving aid. For now, cash transfers are still, in some sense, an exception: conventional wisdom holds that structural barriers (lack of transportation, lack of nearby medical services, etc.) are too great to overcome with cash alone. But what if people in developed countries began to consider Niehaus’s proposal? Unconditional cash could become the rule rather than the exception.