Dear governments: Want to help the poor and transform your economy? Give people cash.

I’ve just finished a new paper with a clunky title (the kind that economics referees hopefully love), Credit Constraints, Occupational Choice, and the Process of Development: Long Run Evidence from Cash Transfers in Uganda. It’s with Nathan Fiala and Sebastian Martinez, run along with Innovations for Poverty Action (IPA).

We tackle one of the big questions in development: How to create jobs and speed up the shift from agriculture to industry in developing countries? We look for answers not at the macro level, but with a field experiment and micro-level data in Uganda.

Countries like Uganda have mostly young, poor, populations. No one is unemployed. If you have no income, you’re in trouble. So you scrape by doing odd jobs and low return activities. The real problems are underemployment (not enough hours) and low productivity employment (low return, low wage work).

So how do you create “good” jobs and productive work? Another way of asking this question is “what is holding young people back?” or “what constrains them?”

Every government or NGO program has an answer to this question, even when they don’t know it. From the vast, vast number of training programs–financial literacy, trade skills, life skills–the default answer seems to be “skills”. If you think these programs are worth doing, presumably it’s because you think (1) youth lack these skills, (2) they can’t otherwise get them, and (3) giving them these skills will produce high returns.

Development economics has a slightly different answer. The evidence is pretty pessimistic about job training programs and financial literacy in poor countries. It’s more optimistic about returns to primary and secondary school in poor countries–wages go up maybe 10-15% with every extra year of schooling. Given how much time and money school takes, though, that’s not always the best return.

More and more, economists think that the real constraint is capital. Studies show that the poor, on average, have high-earning opportunities if they get a little cash or equipment. Studies with existing farmers or businesspeople have seen returns of 40 to 80% a year on cash grants.

This gels with economic theory, which says that infusions of capital should expand people’s choice of occupations, self-employment, and earnings. People can’t get access to that capital through loans because credit markets are so broken and expensive. This can be a development trap, or at the very least a drag on growth.

The studies we have, however, overstate what we know. Most of it comes from Asia. Most of it looks at existing businesspeople and farmers only. So we don’t know a lot about giving cash to the very poor and unemployed, or how to help people shift from agriculture to cottage industry–the structural change so fundamental to modern economic growth.

Enter our study. We look at a large, randomized, relatively unconditional cash transfer program in Uganda, one the government designed to stimulate this kind of job growth and structural change.

The Ugandan government did what dozens of African governments are doing under the guise of “Social Action Funds” and “Community Driven Development”: they sent $10,000 to a group of 20 or so young people who applied for it. This is about $400 a person, equal to their annual incomes.

To many people, this sounds like a crazy development strategy. We don’t trust the poor (let alone a bunch of rural 25-year olds) to spend that kind of money responsibly. We want to tie their hands, or make the decisions for them, or at least make them dig useless ditches for three months in exchange for cash.

We wanted to know. So we worked with the government and World Bank to randomize the grants, and followed nearly 2500 people two and four years afterwards.

Here’s the “surprise”: Most start new skilled trades like metalworking or tailoring. They increase their employment hours about 17%. Those new hours are spent in high-return activities, and so earnings rise nearly 50%, especially women’s.

The people who do the best are those who had the least capital and credit to begin with–consistent with the idea that credit constraints are holding back youth. The more tightly coiled the spring, the bigger the bounce on release.

What’s more, credit constraints seem to be less binding on men, since men in the control group start to catch up over time. Female controls do not, partly because they have worse access to starting capital. With the grant they take off, further even than men. Without it, they stagnate, even more so than men.

This is not a unique result. Two weeks ago I put out a report with IPA on a different program in Uganda, with poor women only. their incomes doubled after getting small grants.

Last, we go beyond economic returns and look for broader social effects. Why? The other belief many people hold dearly: Poor, unemployed men are more likely to fight, riot or rebel.

Because of this, governments and aid agencies routinely justify their employment programs on reducing social instability, or promoting cohesion. Indeed, that was the express goal of the Ugandan program.

Even though we see huge economic effects, we see almost no impact on cohesion, aggression, and collective action (peaceful or violent). If that’s true more broadly, we probably can’t justify all this public spending on the grounds of social stability. But the impacts on poverty and structural change alone probably justify big public investment.

So is it time to stop giving people skills? Not entirely. Part of the reason these Ugandan youth did well is that they invested some of their grants–maybe a third–in skills training. But mostly they invested the grant in tools and inventory and inputs. It was their choice.

I used to think skills and capital were like right and left shoes: one’s not so useful without the other. Now I think of capital like the shoes and skills like the laces: if I have capital, i can jog a good pace, but I can’t really run unless I have the skills. But first I need the shoes. (And cash can buy me both.)

The problem is: too many programs just hand out laces. Old, ratty laces that don’t even fit people’s shoes. I don’t know why we do that. Maybe because we academics and NGO workers and elite government officials all live in a world where we ourselves invest in skills because there are things out there called firms and bureaucracies that have capital, and will pay us to use it.

The poorest don’t have firms ready to hire them. Perhaps we need to stop projecting our own labor markets and biases and low opinions of our own self-control onto the poor, and show them the money.

Read the full paper, which has some of the backup for my claims and references above, here. See the project summary and policy brief on the IPA site.