“Give a man a fish, and he’ll eat for a day. Give a man a fish net, and he’ll fish for a lifetime. Give him 20 bucks, and he may buy something way better than a fish net, or he may just drink it. It’s hard to say…”
That, in a nutshell, is the state of the Cash-or-Kind debate today.
The miseries of just handing out consumption goods are well understood. Nobody wants your old sneakers. Cash is preferable in every conceivable way.
But that’s not the interesting question.
The interesting question pits cash against capital goods: investments that people could – but might not – spend their cash on. And here it’s a good deal murkier, because the normal reasons behind the “presumption in favour of cash” go wobbly in an investment context.
In the context of a critique, Matt Collin puts forward the presumption in favour of cash pretty succinctly:
Even without the growing body of empirical evidence indicating that just giving cash is an incredibly cost-effective way to increase welfare, there is an extremely compelling theoretical case to be made for cash transfers. Poor households have preferences… and no one will ever have better information on these preferences than these households. Transferring households cash allows them to best allocate these new resources to meet these preferences – otherwise, we run the risk of wasting resources onstuff that households just don’t want…
When we’re talking about consumption goods, this is entirely uncontroversial, and it’s important to fully digest the reasons behind the presumption in favour of cash in that context.
But take a step back and ask yourself why donors give in the first place.
Aid donors want to help people overcome poverty as a condition. They want to do more than just enable a one-time consumption spike, nice though that may be. They want to help people switch trajectories onto shift onto a higher consumption plane.
Is in cash aid better than in kind aid at helping people do that?
That’s an empirical question, and one that I don’t think we have a good answer to yet. (If you think that’s wrong, that’s what comments are for!)
But there are reasons to believe thatn at least in some settings, a grant of capital goods does better than cash at shifting people onto a higher consumption plane. This Ghanean RCT makes for some startling reading in that regard. And we know that in Uganda, just giving out hybrid seed to farmers one year doubles their willingness to buy it the next.
It makes good intuitive sense: when you have cash in your pocket, investment goods have to compete with any number of consumption goods, some of which may be desperately tempting right around now. Giving aid in the form of a capital good rather than cash is exactly the kind of behavioural nudge we know helps people exercise self-control. In fact, in that Fafchamps paper, it’s precisely the people with self-control issues who benefit most from capital-good rather than cash aid.
There’s a lot still to learn about this, and replicating Fafchamps’s study design in other settings really should be a priority.
The key, as always, is to keep a determined focus on incomes. Make that your metric, and the rest of the pieces fall into place.