There’s a very long list of ways that nonprofits have tried to attack the problem of global poverty. Some give microfinance loans. Others create skills training programs. Still others give laptops to kids. Of all the methods, however, only a few have actually been proven to really help people move up the economic ladder.
One that has been one of the most successful is also incredibly easy, yet hard for many philanthropists with a pet cause to fathom. The simple trick: If people are living on less than $1 a day, just give them cash and let them decide how to spend it.
“It’s a very simple, direct, and low-cost way to help people who are poor,” says Yale University economist Dean Karlan. In 2002, he founded Innovations for Poverty Action, a nonprofit that uses randomized, controlled trials to test which poverty interventions actually work. These trials are the gold standard in science and are often used for testing the safety and effectiveness of new drugs or other medical interventions. But they are much more rarely used to evaluate social interventions, partly because they are difficult to set up.
While various governments have used cash transfers since the 1980s–Mexico, for example, gives poor families monthly stipends–it’s relatively new to the nonprofit world. In 2009, a group of grad students from Harvard University and MIT were looking for a way to give their money in a way that would have the most impact and started looking at the latest research in global development.
“Randomized controlled trials had just gotten introduced to the field, and what happened was that lots of things that we thought worked–training programs, microfinance–turned out not to work as well as we had hoped,” says Michael Faye, who was working on a doctorate in economic development at the time. “And things that we would have been surprised to work, like cash, turned out to work quite well.”
The catch: There wasn’t a place where the students could donate money directly. No charities
were working on cash transfers offered the opportunity to donate to cash transfer programs, and it’s probably not the best idea to go around the developing world handing out cash on your own. So, that year, the group decided to start their own organization, called GiveDirectly. It took advantage of the fact that it had recently become possible to transfer money via mobile phone, helping eliminate typical opportunities for corruption. All the founders had to do was find the poorest people in the world and start to hand over the money. By 2011, after testing it first with friends and family, they had opened it to the public for donations.
In Kenya, GiveDirectly uses data to identify the poorest households and then transfers them $1,000, the equivalent of a year’s salary, no strings attached. The recipients can spend the cash on whatever they need most. In many cases, in Kenya, one of the two countries GiveDirectly works, families spend money on replacing a thatched roof with a metal one that lasts longer without repairs, saving the family money over time. Others have spent money on health expenses, paying a dowry, and buying a motorbike for a business. GiveDirectly tracks spending through surveys and separately through formalized studies. In the next year, it plans to reach around 20,000 households and 100,000 recipients.
The system avoids a massive problem in international development: Charities prescribing solutions that don’t fit the real needs that poor people have (and sometimes make things worse). It also avoids wasting money on a lot of overhead.
“Cash represents a profound change in thinking from a sector that existed to allocate capital on behalf of the poor and make decisions on their behalf to a world where we trust the poor to make the best decisions for themselves,” Faye says.
Though some have worried that recipients might just blow the cash on vices like booze, studies have proven that doesn’t happen; people are most likely to invest in their homes, their work, and to save money. An independent study of GiveDirectly found that a year after families received money, they’d increased their earnings by 34%. The number of days that kids went without food dropped 42%.
When the nonprofit first launched, funders were skeptical. “When we started GiveDirectly, people asked us if we were smoking crack,” says Faye. “And talked about GiveDirectly like a nice experiment from some grad students.”
Today, attitudes are much more positive. In 2014 alone, it had given out $6.6 million and raised another $17.4 million. After some early investment from Google.org, the nonprofit just received another $25 million from Good Ventures, a foundation started by Facebook co-founder Dustin Moskovitz.
Silicon Valley loves the nonprofit’s data-driven approach, which carefully measures the best way to do everything from identifying households most in need to how to split up payments. The founders have designed an operational model that allows constant experimentation and testing–and the tech crowd adoption of the “effective altruism” movement eats it up.
Another important question is whether this idea could help the poorest of the poor in the United States and other developed countries. The relatively few experiments suggest it can: A pilot project in London that gave about $4,500 in cash to 13 homeless men–a year later, 11 lived indoors and all had spent the money in productive ways. In New York, the Opportunity NYC Family Rewards program gave $8,700 to poor families over three years, on the condition they worked to meet certain goals. After it was over, these families had more savings and less overall hardship. But GiveDirectly and others argue that we still need more evidence to know whether cash transfers can work in the U.S, since the average homeless person in New York may have deeper problems that the average slum-dweller in Kenya does not. Still, at the very least, cash-transfer advocates are now questioning the common advice that it’s better to donate to a food bank than give out money to panhandlers.
Today, GiveDirectly is still the only nonprofit to focus solely on cash transfers. Other global organizations, like the International Rescue Committee, are increasing the use of cash, though it can be a hard switch for other established organizations to make. If programs are focused specifically on food or shelter or something else, cash doesn’t really fit into the system, Faye says. If organizations have deep investments, including staff jobs, that would be eliminated if they just give away money, it’s understandable why any change might be slow.
But the model of cash transfers is also starting to have an effect in another way–establishing a baseline of success for development work. In the past, organizations might have relied on feel-good stories or some basic, self-collected data to report impact. Now, more are embracing independent studies. More importantly, their results could be compared to the return on investment from simply handing over money.
Faye compares it to the introduction of index funds to Wall Street. These gave investors a way to compare the performance of their fancier investment funds that were more actively managed by experts. “In the same way that page two of an annual report for Berkshire Hathaway will often have a comparison with an index fund, you could imagine a world where every NGO and donor is actually benchmarking their intervention against cash,” he says.
Of course, cash is not a panacea. Poor people have problems that go much deeper than not having money. “This solves the ‘I don’t have money’ problem, but what if the problem is also a matter of aspiration, a matter of hope, a matter of information about markets or trading?,” says Kaplan. And because more and more philanthropists are becoming data-driven, direct cash transfers risk become a pet cause unto themselves.
But every other needed intervention could now be studied and compared to cash. “Roads are not going to build themselves, hospitals are not going to show up out of nowhere,” Faye says. “There are lots of public goods that we will need. But we do think that there is a large fraction of it that is likely worse than cash and needs to be benchmarked against cash.”
And the long-term impact of cash transfers still needs more study. “The short answer is that we don’t have quite enough evidence on pure, unconditional cash transfers,” says Chris Blattman, a Columbia University professor who has studied cash transfers. “The GiveDirectly study is just one site and short-term data at that. It’s genuinely uncertain what happens in the long run.”
Still, out of development interventions that have been studied, cash transfers have the best evidence so far, in some cases showing benefits over six years.
Now, the question is how the work can scale. GiveDirectly has grown quickly, but is only in Kenya and Uganda so far. It also plans to begin larger studies looking at impact on a community level–how it could change whole economies–rather than just how cash can help individuals. “That’s our plan,” Faye says. “With the Good Ventures grant, that’s really the next chapter.”