“Let them eat cash”: Some post-op reflections

Lately, my research and others have suggested that simple cash handouts might be one of the most effective anti-poverty strategies in the world. Is it time to bring it home?

Today I have an Op-Ed in the NY Times on cash for New York’s homeless.

You might also worry that the poorest of New York are different. The average person in Uganda is impoverished; it’s easy to believe he would make good decisions with cash. But a homeless person in New York is not average. Substance abuse is pervasive. Maybe panhandlers here are different from the global poor.

I used to believe this. Now I’m not sure. A few years ago, I started working in Liberia’s urban slums. My colleagues and I sought out men who were homeless or made their living dealing drugs or stealing. Many abused alcohol and drugs. We tested different programs in a randomized trial of a thousand men. One thing we tried was giving out $200 in cash.

The short story: An eccentric Chinese millionaire tried to give cash to the NYC homeless last week, and an NGO put a stop to it.

The original article is equal parts fascinating and infuriating. Mostly infuriating. First, the millionaire. He took out a full page ad to announce the $90,000 giveaway. The ad probably cost more. So it was mainly an exercise in self-aggrandizement.

Then there was this:

Mr. Chen addressed the audience and then uncorked the news the crowd had been waiting for: “I will give $300 for every participant today.”

The homeless men and women shot to their feet, whooping and applauding.

“No he won’t,” Michelle Tolson, the mission’s director of public relations, said. “The police will shut him down.”

Officials from the Rescue Mission quickly brokered a deal with Mr. Chen’s assistants, allowing him to hand $300 to several chosen homeless clients in a symbolic gesture. The clients, however, would have to return the money.

I can understand Rescue Mission’s concerns. The Op-Ed was accepted Friday night. Saturday I walked by the same two homeless people that have been on my corner since I moved here two years ago. They are not in great shape. “Really?” I thought to myself, “Did I just tell New Yorkers they should hand out big bills to these people?”

Admittedly we had our worries in Liberia too. This is why we started on a small scale–a few dozen men on the streets, with innumerable safeguards and interventions a-ready–before scaling up to the full evaluation of 1000. In the following year, most of what I believed about homeless men, drugs, and petty criminals in Liberia was turned on its head. Even though I’d already spent more than a year with many of the men. So experimentation and testing your prejudices is important.

(Aside: if you’re interested in hearing more about the Liberia project, there is this NPR Planet Money podcast. Also this Foreign Affairs piece. The working paper will be out soon.)

Even so, I worry I’m wrong, and I worry about bad consequences. At the end of the day, I advocate taking chances and experimenting because I believe (indeed I’ve seen) the cost of ignorance is greater. That doesn’t mean it’s not morally troubling.

48 Responses

  2. Sort of understand what MK is talking about but there is a deeper point. Professor Blattman has proved the effect of the US welfare state – give a man a pillow and he will sleep on it or don’t teach him how to fish just keep giving him fish, i.e., keep giving welfare. Therefore he has disproved the utility of the welfare system. So we need a radical overhaul of US welfare or its abandonment to the Monrovia model. But to be perfectly fair about it the comparison is not equivalent as many things, not remotely touched on in the post or comments, could be enumerated. Not least the women. What percentage of drug addicts in NYC are male and married to female drug addicts. Then check out the Monrovia percentage. Far less I would guess. You could extend that to what percentage of African women drink, smoke, etc. Few compared to US. So there is no equivalence in the comparison. The cultures are different.
    The irony in all of the work is that it is giving a better insight into the USA and the its problems with welfare than Liberia.

  3. Professor Blattman:

    I was happy to see your piece in the Times after reading your similar take here on the blog.

    A prior commenter may have already raised this in an elliptical way, but one aspect that seems to go unaddressed in your thinking is making the case for how proposed transfers differ from or extend the impact of the existing/traditional US welfare and government benefit systems. I imagine many (most?) homeless no longer receive full benefits under the time-bound and heavily bureaucratic system we have today, but it seems to me that you can’t really hope to move your proposal into the mainstream of domestic political debate without addressing the objection–to a person focused mainly on left-right politics I would think it would be one of the first that comes up.

    The prevalence of mental illness among the homeless might be another confounding factor (that to me has always been one of the justifications I’ve heard for “Don’t give cash”), and one I’d expect to differ between Monrovia (where with fewer safety nets and less on average to fall back on, a little bit of bad luck would seem able to put you on the street) and New York (where to be long-term homeless would seem to require more bad luck–or some would say bad decisions).

  4. It seems to me that most people spend their last $200-300 the same things: food, gas, toiletries, maybe a little pot or alcohol. We aren’t as far above the homeless as we like to think. Give them cash, they value it more.

  5. I don’t think the issue is that homeless people are qualitatively different here, but that the amount of cash you’d have to give away to someone in NYC is substantially higher than in Liberia

  6. Sorry to be back so soon. Strange conclusion from it all is that to change NYC to be as per Monrovia, at least in better use of cash, the permanent welfare has to be stopped and one off UCT given instead. Or is that just a daft conclusion but do your results imply that.

  7. A valid direct comparison is between as near as exact equivalents: homeless, drugs crime In NYC and the same in Monrovia or anywhere else; not the poor which is a different thing.
    I would say that the homeless, etc., in NYC, are different from the equivalents in Monrovia hence the benefits of UCT in NYC will be less than Monrovia. Why different is more difficult to say. Possibly: level of substance abuse in NYC is total addiction; nature of substance abuse in NYC is more harmful: crack cocaine; mental issues associated with abuse more serious in NYC. But another point is cost of substance abuse in NYC: 3 days high on 200 whereas in Monrovia maybe a month; don’t know. Finally, the Monrovia ones try to make a living while NYC not interested. The Monrovia case was a one-off; NYC is permanent welfare. This indicates a big difference, not equivalent.
    UCT to poor in this area: SSA semi-arid and poor would not result in abuse of CT in major percentage terms. Old lady here got $3000 recently; her late husbands pension lump sum; sort of a cash transfer. First thing she did with it was she built a brick house. Most would go down this type of road. Problem is poverty not abuse but it can’t be denied that some of the UCT would be abused but not any more so than other types of aid. Some sell the food aid and go to the beer hall. You can’t stop that by method of aid. But in general is UCT more likely, even with these problems of abuse, to reduce poverty, than other forms of aid. Probably, but it is not the only aid needed.