Agriculture Reference
In-Depth Information
written, and our British Columbia-Louisiana data show just 69 percent of the multiyear
contracts are written. This, by itself, indicates that courts are often not used to enforce these
contracts. Instead, farmers and landowners prefer to make oral agreements independent of
the contract length and rely on the market to enforce the agreements.
19
We test predictions 3.1a and 3.1b by using logit regression analysis to examine the effects
of selected variables on the decision to use an oral or written lease. Table 3.2 shows the
coefficient estimates for these equations. Because the data sets do not contain the same
variables, we estimate different specifications for the Nebraska-South Dakota and British
Columbia-Louisiana equations.
Irrigation assets and long-term crops like trees are specific assets in a land lease. In
the presence of these assets, there is the potential for one party to hold up the other. A
written contract is more likely to mitigate this possibility; thus prediction 3.1a implies that
IRRIGATED and TREES (only available for the British Columbia and Louisiana data) will
have negative coefficients. In both the equations the estimated coefficients for IRRIGATED
are negative and statistically significant, indicating that contracts are more likely to be
written when irrigation assets are present. In the British Columbia-Louisiana equation, the
estimated coefficient for TREES is negative and statistically significant, also as predicted.
The better the reputation of the farmer or of the landowner, the more likely the contract
will be oral, as implied by prediction 3.1b, because the simple oral contract can be relatively
easily enforced. The age of the farmer (AGE) can be used to measure reputation, and we
expect a quadratic relationship between a farmer's age and probability of an oral contract.
20
As the years go by, a farmer develops a reputation that increases the viability of an oral
contract but this reputation effect should decline and become negative because of the well-
known last period problem. If the farmer knows this will be the last period of a relationship,
it becomes his interest to cheat since he cannot be punished by an end to the relationship
(Klein and Leffler 1981). Thus, as the farmer reaches the end of his productive farm
life, the reputation effect begins to weaken. In both equations we find, as expected, the
estimated coefficients for AGE to be positive and for AGE
2
to be negative. All coefficients
are statistically significant in a one-tailed test.
Other variables are also used to measure information about reputations. The estimated
coefficients for YEARS DURATION is expected to be positive. As the duration of a contract
increases (YEARS DURATION), the parties have more information about each other. If
cheating or poor farming is detected, the contract would be terminated and so long-lived
contracts reflect satisfied parties. Therefore the more often a contract has been renewed,
the more likely it will be oral. We expect a negative coefficient for CHANGED PARTIES
because this variable indicates a recent change in contracting parties, which is interpreted as
a situation in which information is limited. New parties to a contract know less about each
other and thus the contract is less likely to be oral. All the estimated coefficients have the
