Since I first read about unconditional transfers (in Banerjee and Duflo’s Poverty Economics book), I’ve been fascinated about this seemingly this-cant-possibly-work approach to foreign aid. Later I discovered GiveDirectly, a nonprofit organization operating in East Africa that helps families living in extreme poverty by making unconditional cash transfers to them via mobile phone.
Last week heralded the completion of (possibly) the most ambitious and expensive RCT in the field of unconditional transfers. Researchers spent 5 years and $10 million in Kenya to answer the following questions:
What happens to one of the world’s poorest places if you randomly pick more than 10,000 poor families out of an eligible pool and give them $1,000 each, no strings attached?
Furthermore, what’s the impact on their neighbors?
Does that do you any good at all? Or is your neighbor’s luck your misfortune, because local prices jump, say, leaving you worse off than before? Setting aside the direct recipients, what do cash transfers do to local economies?
And what did they find?
Cash transfers benefited the entire local economy, not just direct recipients.
I’m reminded of a talk by Dr. Trilochan Sastry where he exhorted a young group of social entrepreneurs this “To start with, just don’t make the situation worse for the impoverished segment whose lives you are attempting to improve.” This might sound jaded but it’s really a cautionary advice based on many misadventures caused by “well meaning” folks.
Against that cautionary backdrop, it was such a relief to learn that:
They could find little in the way of adverse effects from the experiment, either in villages that got the cash or in those that didn’t. Spending on temptation goods — such as cigarettes, alcohol and gambling — did not increase. People didn’t work less. Rates of domestic violence didn’t change, nor did more children drop out of school. Local income inequality levels did not change. And contrary to a common fear, the program had minimal effect on prices: Inflation increased less than 1 percent over and above Kenya’s overall rate.
The positive impact of this intervention was massive.
What made the study really path-breaking, though, is that it was huge: The money handed out amounted to more than 15 percent of the GDP in the treatment area, reaching 10,500 of the 65,385 households there. Dump that much cash into a local economy, and you would certainly expect it to grow. But by how much?
Every $100 given directly to the poorest households was generating between $250 and $270 in GDP. That’s a fiscal multiplier in the range of 2.5 to 2.7 18 months after the money was spent — a huge (fiscal multiplier) number by global standards. [PDF link]
How come? Because the very poor spend their money locally, and the shops they spend it at, in turn, spend it locally again, a chain effect that stimulates demand and lifts revenue for the tiny businesses throughout the area. The research found some evidence — though not conclusive — that local wages had risen, perhaps more strongly in villages that directly received cash than in their neighbors.
This week’s great reportage brought to you by Francisco Toro in this WaPo article.
Two other amazing related WaPo articles that you must read: