Agriculture Reference
In-Depth Information
In addition, the farmer could deplete the water resource or may contribute to soil salinization.
Pastureland creates an incentive for overgrazing. Overgrazing might be mitigated with
a share contract, but sharing live animals is difficult and infrequent given the costs of
measuring the livestock characteristics. Pasture is distinct from hay land in that pasture is not
cultivated and harvested like alfalfa and other hay crops. We use the dummy variables ROW
CROP, PASTURE, and IRRIGATION to identify crops with asset exploitation incentives.
Grain crops generate incentive effects distinct from those mentioned above. Grain crops
are less susceptible to moral hazard because limited cultivation reduces the potential for soil
exploitation compared to row crops. Also, because grains are sold in markets where third
party measurement occurs, there is a reduced potential for crop theft compared to grasses
and tree crops. We thus predict that land used for grain crops is more likely leased than land
used for nongrain crops. In our regressions we include a dummy variable (GRAINS) that
identifies these crops.
Finally, effort specialization can influence the ownership—contract decision for land. The
lower the amount of specialized human capital (farming skills), the more likely a landowner
will lease out his land. We use a dummy variable—LANDOWNER HUMAN CAPITAL—
which identifies landowners that have low human capital.
Table 8.4 presents the logit coefficient estimates from two separate equations. Each
of the 1,573 observations is a single plot of land, either leased or owned by the farmer.
The variables are organized into those measuring capital constraints, asset moral hazard,
effort specialization, and other controls. The coefficients estimates for WEALTH and
NET WEALTH are both negative and statistically significant. These estimates provide
evidence that capital constraints lead to more leasing of land. The coefficient estimates
for IRRIGATED, ROW CROP, HAY, PASTURE, and TREES all have the expected sign
and are statistically significant. Land with multiple moral hazard costs or single but severe
moral hazard costs is less likely to be leased. The estimated coefficient for LANDOWNER
HUMAN CAPITAL is positive and statistically significant, indicating that land leasing is
more likely when the landowner lacks specialized farming skills.
21
The estimated coefficient
for GRAINS is not significantly different from zero, so we find no support for the prediction
that grain crops are more likely to be associated with leased land.
Table 8.4 also shows the coefficient estimates for several control variables. ACRES
and CROP VALUE control for the size and value of the farmland plot. The estimated
coefficients indicate that the size of the land plot has little impact on the decision to lease
or own land. The CROP VALUE estimates, however, indicate that the higher valued crops
are more likely to be leased. EDUCATION measures the farmer's formal education. The
estimated coefficients indicate that increases in farmer education increase the probability of
leasing land. WORKERS indicates the number of laborers on the entire farm. The estimated
coefficients indicate that this has no impact on the lease-own decision for land.
