Photo: Konrad Burchardi

Assistant Professor Konrad Burchardi and IIES graduate student Benedetta Lerva have had their paper "Moral Hazard: Experimental Evidence from Tenancy Contracts’’ (with Selim Gulesci (Bocconi) and Munshi Sulaiman (BRAC Institute of Governance and Development) accepted for publication in the Quarterly Journal of Economics, one of the world's top 5 publications in economics.

We wanted to learn more and sat down with Konrad for a Q&A.

Can you summarize what your paper is about?

Sure. We test an old hypothesis for why agricultural productivity might be low. Imagine a tenant farmer who rents land from a landlord, needs to pay for seeds and fertilizer himself, and at the end of the season pays 50% of the agricultural output to the landlord as compensation for the land. Economists have long pointed out that with such output sharing rules the tenant farmers incur the full costs of any investments but only get to keep a fraction of the returns to their investments, and they theorize that this leave tenant farmers with insufficient incentives to invest. For example, the farmers might simply not purchase fertilizer, and this might drive low agricultural output.

We tested this theory by conducting an experiment in collaboration with a partner organization in Uganda: we rented out land to tenant farmers, chose randomly a subset of them to be allowed to keep a higher share of output, and observed how the farmers’ behavior and agricultural output responded under those different output sharing rules.

We find that farmers whose contracts allow them to keep 75% of output produce 60% more than farmers who are only allowed to keep 50% of output. That is a surprisingly large output response. But we show that it can be explained by two things: the farmers with the higher output share invest more in inputs, and they choose more profitable, but risky, inputs and techniques.

What are the most important lessons policy makers can learn from your paper?

A large fraction of the world’s poor households rely on agricultural income to support their family, but agricultural productivity is notoriously low. Increasing agricultural productivity is hence seen as one way to improve the lives of the poor.

At the same time, farmers in developing countries are often exposed to such output sharing rules that only allow them to keep part of the output. There are many reasons for this, amongst them formal or informal taxes, insecure property rights or formal agreements with the landlords such as the one I just described. What our research shows is that any policy that has the effect of giving farmers a larger share in output likely provides them with the incentives needed to make the choices that increase output.

Where does the idea for this paper originate from?

The core idea of this paper is very old. The famous economist Alfred Marshall wrote about the incentive effects of output sharing rules when discussing late 19th century agriculture in Britain. Anybody who studied development economics will know about it. In fact, his analysis is really the main predecessor of most modern economic theories analyzing the role of incentives for decision making. What we contribute is a rigorous test and quantification of how important those factors are to understand agricultural productivity in developing countries today.

How does it feel now that your paper is accepted?

I remember the times when I started reading academic papers, and I recall that many of the papers that most impressed me where published in The Quarterly Journal of Economics. That they now publish the results of our work is very humbling. And wonderful.