Agriculture Reference
In-Depth Information
ACRES is again used as a control variable in all equations. In each case the estimated
coefficient in not statistically significant, indicating that the size of the land plot does not
affect contract terms.
Absentee Owners, Agricultural Ladders, Capital Constraints, and Families
We use additional variables to test our model against predictions—about absentee landown-
ers, the agricultural ladder, capital constraints, and families—made elsewhere in the
literature.
32
The Nebraska-South Dakota data are most amenable because we have ex-
tra information depending on whether or not farmers or landowners were surveyed. The
British Columbia-Louisiana data do not have a detailed landowner component, so some of
these tests can be conducted only against the Great Plains contract data.
In the landowner sample, ABSENT is included to test a common prediction of landowner
behavior. In many models, landowners are assumed to provide valuable farming information
and policing along with the land; therefore, the “absentee landowner” faces higher costs of
participating in farming activities. This implies that absentee landowners will be less likely
to cropshare, so the estimated coefficient for ABSENT should be negative.
33
The negative
coefficient in the landowner sample supports this prediction. The prediction is, however,
not consistent with the estimates in table 4.4 using the British Columbia-Louisiana data.
Our model has no explicit prediction because the landowner does not monitor the farmer.
In the traditional theory of the agricultural ladder, a farmer “climbs” from wage farmer
sharecropper to cash renter to, ultimately, landowning farmer. In Spillman's (1919) words:
“The first rung of the agricultural ladder is represented by the period during which the
embryo farmer is learning the rudiments of his trade. In a majority of cases this period is
spent as an unpaid laborer on the family farm. The hired man stands on the second rung,
the tenant on the third, while the farm owners has attained the fourth or final rung of this
ladder” (171).
Although the ladder hypothesis is not derived from a well-defined economic model, it has
had a strong following among agriculturalists and it does have a clear prediction regarding
the choice between cash rent and cropshare. This implication is simply that older farmers, or
those with more farming experience, are more likely to cash rent. Thus, the variable AGE is
predicted to be negative. The evidence in the second equation in table 4.3, however, rejects
this hypothesis, because the coefficient is small, positive, and statistically insignificant. The
data in table 4.4 also reject this hypothesis using the British Columbia-Louisiana data.
In a cash rent contract, a farmer must pay the landowner prior to harvest; in a cropshare
contract, the payment comes after harvest. Thus, there is a long history in economics, from
Gray et al. (1924) to Laffont and Matoussi (1995), who argue that farmers facing capital
constraints will be more likely to choose a cropshare contract. We use the variable ACRES
OWNED—which measures the amount of land owned by a farmer—to measure such a
