Should the Hungry Be Given Cash or InKind Transfers?

Our society has reached a stage where we all agree that hunger is an unbearable form of deprivation to which no human being should be exposed. Malthusian theories have been shelved and achieving the basic nutrition needs for the whole world population is a feasible and relatively low-cost dream. However, malnutrition is still prevalent when people do not have access to jobs or fertile land to generate enough resources to feed themselves, so they must rely on social transfers.

Traditionally, there has been greater support for providing assistance to the poor through in-kind transfers rather than cash. This follows a paternalistic view that encourages consumption of the goods being transferred. Supporters argue that if we give food or vouchers to buy particular foods, recipients will consume more of these goods. On the other hand, if we give cash, recipients might spend it on vice or other less socially desirable goods.

But what does the evidence say? Policies have been based on anecdotal evidence for a long time, but a new wave of randomized control trials is bringing some robust evidence to the debate.

What empirical studies show

One of the first studies looked at conditional cash transfers in Latin America. Several field experiments show that cash transfers can work to foster health-seeking behaviors and investment in education when they are tied to strict conditions (e.g. visit a health center once per month, enroll children at school). Some evidence also shows that even when these conditions are not tightly enforced, people change their behavior in the intended direction if they are told the money should be used for a particular goal (Najy Benhassine and co-authors, for transfers labeled for education). But, evidence is still limited, and we still do not know if the same effects would be obtained without any conditionality.

There have been now several studies randomizing the way in which recipients receive transfers. Jesse Cunha studies a program that randomizes at the village level in Mexico whether transfers are received in kind, cash, or not received at all. First, he finds that the two types of transfers increase food consumption and nutrients intake, and very little of the cash transfer is spent on vice, with the majority being spent on nutritious foods. Second, in-kind transfers have small differential effects on the components of consumption compared to cash transfers, but not on total food consumption.

In Ecuador, Melissa Hidrobo and co-authors find that food transfers lead to a larger increase in calories concentrated in the goods that are transferred, but that cash or food vouchers also work relatively well and are more cost-efficient. In the Democratic Republic of Congo, Jenny Aker shows that households re-sell part of the goods transferred, with the result that food vouchers and cash transfers have similar effects. Something along these lines is found for Niger by John Hoddinott and co-authors.

The cost of transfers

The small positive effects detected for in-kind transfers have to be weighed against their costs. One cost is born by the recipient: cash transfers are usually preferred to food stamps, less stigma is attached to them and offer more freedom of choice. However, the main cost differential of food transfers vs. cash transfers is that of bringing the food to the recipients.

Taking this evidence into account, the policy debate is seeing the balance tilting towards cash transfers. This is especially the case in contexts without high inflation so that the cash transfers do not lose value fast, where food markets work relatively well, so that recipients are able to buy food easily, where corruption does not make cash hand-outs more difficult than food distribution, and, ideally, where there is a good access to the financial system so that payments can be easily allocated.

On the agenda, these are the questions for future research: how do differential effects of in-kind transfers depend on tastes, impatience levels or varying preferences within households for different goods? Is it possible to predict if and how households substitute similar products with the transferred goods or even reduce consumption levels? What are the long-term effects on health and cognitive development of an increased intake of fortified transferred products? Do they compensate the additional cost of in-kind transfers? These are key areas for research that can bring further evidence to a debate that is of essential importance in our fight against hunger.