Agriculture Reference
In-Depth Information
topic, table 1.2 points out a number of characteristics of the farmers and landowners that
are common across all of our data sets. Table 1.2 shows that 60 percent of the landowners
are or were at one time farmers. The similar social-economic background and demographic
features of farmers and landowners are inconsistent with a model that posits dichotomous
preferences and risk sharing. But there is even more. Table 1.2 shows that renters are often
landowners and in some cases (6%) rent out land simultaneously, as well as hold both share
and cash rent contracts. It is difficult to explain the coexistence of an individual being on
both sides of a market with a simple incentive versus risk-sharing trade-off. It is also difficult
to explain why a farmer would hold both types of contracts, without imposing unrealistic
assumptions on farmer preferences.
6.5
Summary
The trade-off between risk and incentives has become a mantra among economists working on agency
issues, despite the lukewarm evidence in its favor.
—Canice Prendergast, “What Trade-off of Risk and Incentives?”
We have used our data on individual contracts in modern North American agriculture—
where cropshare contracts remain an important part of farming—to test some well-known
predictions derived from the standard risk-sharing model that includes an incentive trade-
off. On a case-by-case basis, using many different empirical specifications, we find that our
evidence consistently fails to support the predictions of the traditional risk-sharing model.
Cropshare contracts are not more likely when crop riskiness increases, when landowners
are institutions, or when farmers derive more of their income from farming. Furthermore,
cropshare contracts are not less likely when farmer wealth increases, or when futures
markets exist. Risk sharing cannot explain the pattern of leasing for other assets, nor can
it easily explain simultaneous cash rent and cropshare contracting by the same farmer.
Collectively, our tests provide robust evidence that forces besides risk sharing are more
important in shaping agricultural land contracts. At the same time, we find more support
for our model that assumes risk-neutral contracting parties and stresses multiple margins
for moral hazard and enforcement costs, supporting long-time critics of risk aversion such
as Goldberg (1990) and recent critics such as Prendergast (2000).
39
While our results show almost no support for predictions derived from a classic risk-
sharing model, this does not, of course, rule out the possibility that other risk-based mod-
els could explain farmland contracts. For example, recently there have been a number of
attempts to explain why we have, along with others, found a positive or insignificant rela-
tionship between risk and incentives. Some of these attempts include models of preference
matching (for example, Ackerberg and Botticini 2002; Serfes 2001). In these models there