Chris Blattman

Search
Close this search box.

Dung as development

A non sequitur on Kenya, thinking past the current crisis to long-term development.

I’m sitting in a session on psychology and development at ‘The Meetings’, and the current presenter, Esther Duflo, asks why do just a fifth of farmers in western Kenya use fertilizer?

This may seem like a relatively unimportant question, but in fact a great deal relies on the answer.

Effecting a green revolution in Africa (of which fertilizer is an essential component) is crucial to the continent’s human and economic development. Increased agricultural productivity will raise rural incomes, reduce the likelihood of famine, liberate children and women from back-breaking work, improve family health, and free people to work in other more productive sectors.

Moreover, the question is part of a larger puzzle: why aren’t simple, inexpensive and available technologies more widely adopted in developing countries?

So understanding why the use of an old, reliable and simple technology like fertilizer is the exception rather than the rule is a crucial area of study.

Studies of model farms suggest that the returns to fertilizer use are not only large, but can help reduce risk as well. So what’s the problem?

Well, it might be that your farm needs to be a model one to get these results, requiring complementary inputs and knowledge of how to use the technology properly. Or it may be that farmers face credit constraints–the cash comes from future sales, but you can’t borrow against that future return because loans are difficult to obtain or expensive.

Esther Duflo, Michael Kremer, and Jon Robinson have been running a number of field experiments in Busia, Kenya, to figure out this puzzle. As it turns out, there is a sweet spot of fertilizer usage that is extremely profitable. Unfortunately, the returns are fairly sensitive—a little too much, or a little too little, and the profitability erodes. It also turns out that the officially recommended mix is unprofitable in the area they study. This could be a simple error on the Government’s part, or it could be that the official recommendations are based on different land types, and not locally tailored.

The optimal amount of fertilizer is difficult to pinpoint, and as a result, profitability only comes with learning-by-doing — a fixed cost that not all farmers can afford.

The authors go one step further, however, and ask whether psychology plays a role as well.

Fertilizer is easy to buy in small quantities, and (if informed about how to use it best) these small quantities can be very helpful. Such small amounts of inputs should theoretically be easy to save for from the last harvest.

The authors propose that we may see a savings problems rather than a credit problem: like retirement savers in the US, Kenyan farmers display too little saving discipline. These are findings that don’t surprise psychologists, but do tend to surprise and excite economists, since they suggest our fundamental workhorse models of human behavior and the economy might have some fundamental problems.

So Duflo, Kremer and Robinson helped some Busia farmers set up voluntary savings programs. Shortly after harvest, the farmers were visited and ask if they want to pre-purchase fertilizer for the next season. Take up of the program was high, and resulted in an increase in fertilizer usage of about a sixth. Take-up is significantly lower if these people are asked later on, such as the beginning of the planting solution. Providing them with a commitment device–the ability to commit to fertilizer before they spent their earnings–significantly increases income over the longer term.

What do these findings tell us? First, they illustrate the need for a strong local agricultural extension service—that is, an effective local state (or private) capacity to help farmers learn how to optimize use of their land.

The findings also illustrate the importance of local research institutions, such as agricultural colleges. Years ago I worked in rural India, and it was remarkable that locally-grown rice varieties were often ones developed by local agricultural research stations to be optimal for local conditions.

But the findings also suggest that making financial products and services available to farmers that allow them to save and pre-commit to future investments—probably a private sector intervention—should increase welfare. Much like we in the developed world are told to ‘pay ourselves first’ (that is, save our paychecks when they come in, rather than after we’ve spent it on consumption) poor agricultural households can also benefit from commitment savings devices. Significant innovation and development of local finance and savings mechanisms could have a huge aggregate impact on development.

Why We Fight - Book Cover
Subscribe to Blog