Ought we to help hungry households with cash or with food? Evidence on consumption and nutrition from a randomized control trial in Mexico:
households do not indulge in the consumption of vices when handed cash. Furthermore, there is little evidence that the in-kind food transfer induced more food to be consumed than did an equal-valued cash transfer.
the in-kind basket contained 10 individual items, and these transfers indeed altered the types of food consumed for some households. While this distorting effect of in-kind transfers must be a motivation for paternalism, households receiving cash consumed different, but equally nutritious foods.
Finally, there were few differences in child nutritional intakes, and no differences in child height, weight, sickness, or anemia prevalence.
Full paper here.
There might be other rationales for sending food over cash. I remember Lant Pritchett describing the decision to distribute rice not cash after Indonesia’s 1997 crisis. Rice bags are huge, heavy and inefficient. This was exactly what they wanted–for it is much more difficult for middle men to steal huge, heavy, inefficient things.
I feel these are the type of questions that just scream “observation bias”. What people do when handed a wad of cash knowing they are being observed and what would do if such experiments were scaled up beyond the capacity of direct oversight may be significantly different.
The real issue is that moving toward giving cash would mean fewer employment opportunities for development workers. Efforts like this will be resisted tooth and nail by the “development community”.
Chris: one reason to prefer food to cash is based on gender concerns. In countries such as India, men control cash while women control food. Right to food activists in India argue that women take better spending decisions than men, though I have not seen any empirical study on this as yet.
Will WFP manage the transition from bulk food delivery, to a nimble, flexible cash-based aid?
Interesting stuff. Of course cultural differences may mean that the results are not universally applicable, but it is still good to knock out at the cynics.
One contrarian thought did occur to me. Is it possible, a la quantum mechanics, for the act of observation to change behaviour? Not being an economist I found much of the paper a bit bewildering, and so I might have missed something, but I couldn’t see any consideration that since the cash recipients knew they were being monitored (they were visited before the cash handouts were made), maybe that influenced their spending decisions.
It is difficult to know how one could get around that problem in research design, but perhaps a longer term monitoring programme would allow time for panel members to become accustomed to the research, and thus to cease to consider it when deciding how to spend their extra cash.
On a similar note, The Economist just ran a brief story about cash handouts to homeless people in London: http://www.economist.com/node/17420321 . They also seemed to have handled the money very wisely and not squandered it.