The praise and fertilization of folly

In this column I want to talk about one of the best presentations I attended at the Latin American Economist Association Meeting in Rio de Janeiro.

Better than the presentations of some of the 5 Nobel prize winners who were there, was Harvard's Michael Kremer. Even though the title of the paper is not the boldest, "Nudging Farmers to Use Fertilizer: Evidence from Kenya", the topic and the approach are very innovative and challenge traditional mainstream economics (Even if you are in Harvard, you are the mainstream). More importantly, there are clear policy recommendations that can greatly assist poor economies.

Should the government give subsidies for the use of fertilizers? The "Chicago Tradition" of Development (a categorisation not well accepted among the participants) starts from the presumption that farmers are rational profit maximisers, so subsidies will distort fertilizer use away from optimal levels (for those that were there, the kind of models we analyzed in Arcand's Development class last semester). On the other hand, many agricultural experts see the use of modern inputs, in particular fertilizer, as key to agricultural productivity. Pointing to the strong relationship between fertilizer use and yields in test plots, they argue that fertilizer generates high returns and that dramatic growth in agricultural yields in Asia and the stagnation of yields in Africa can largely be explained by the use of fertilizer in the former.

The answer of Kremer and friends is different. In previous studies they have observed that African farmers know that fertilizers are good and productive, but still they usually don't use them, even when expected returns can be as high as 70%. Why? Because people can be irrational!! The authors build a model where farmers are allowed to be naive about their future preferences and biased towards present pleasure, resulting in the procrastination of investment decisions (in line with the studies on hyperbolic discout rates). The on-field tests of the model seem to corroborate the predictions.

So, which policy to follow? Consistent with the model, the efficient allocation under subsidies is more a matter of when, rather than how much. A small time-limited reduction in the cost of purchasing fertilizer at the time of harvest induces substantial increases in fertilizer use, as much as considerably larger price cuts later in the season. Such small early discounts could help present-biased farmers commit to fertilizer use without substantially distorting decisions of non-procrastinating farmers and incurring other costs of heavy subsidies. In joint work with a Kenya based NGO, they evaluate the impact of this kind of policy, and find strong supportive evidence.

So the problem is not being irrational, but how to give the right incentives for productive craziness. A modern Praise to the Folly!