Agriculture Reference
In-Depth Information
Table 8.1
Alternative governance structures for asset control
Own
Short-term contract
Long-term contract
Asset owner
(1) Pure family farm
(2) Custom contract
(3) Custom contract
provides labor
(e.g., custom
(e.g., year-long pest
harvesting crew)
control contract)
Asset owner
(4) Simple contract
(5) Simple contract
does not
(e.g., most equipment
(e.g., standard land
provide labor
NA
leases)
leases)
Note:
NA = not applicable.
The Basic Model
The model in this chapter is a more general version of our model of contract choice. As
before, we consider an asset to be a collection of attributes that are both variable and
alterable. Typically, just one or a few attributes are priced and the rest remain unpriced
and unspecified in a contract. When an attribute is not directly priced moral hazard occurs,
the attribute becomes exploited beyond its first-best level, and the moral hazard costs are
incorporated into the priced attribute of the asset. For example, when a farmer rents land
by the acre, the competitive price per acre reflects the extra nutrients and soil moisture that
are expected to be used by the lessor since these margins are not priced directly.
11
We incorporate this into our basic model by maintaining the original notation, but slightly
reinterpreting the variables such that our production function has both a priced and an
unpriced attribute of each asset used. The full production and information structure is
given by
q
=
h(e
l
k)
+
θ
q
,
,
, where
is the crop output (with price still normalized to 1),
e
is the amount of
standardized
farmer effort and measured in terms of hours;
l
is the
priced
or contractually specified standardized attribute of the asset; and
k
is the
unpriced
or unspecified attribute of the asset. For example, if the asset is land,
l
would be acres and
k
might be the soil nutrients; if the asset is a building,
l
would be square feet and
k
might
be durability. The ratio
is a measure of the quality of the asset; the greater the ratio, the
higher the quality. Finally, nature's input,
k/l
θ
, is distributed with probability density function
x(θ)
. As always, we assume that all marginal products
are positive and diminishing, all inputs are independent of one another, and that
and cumulative density function
X(θ)
0.
Increases in specialization are incorporated into the model as decreases in marginal costs
for labor and the asset, which implies that competitive rental markets exist for all inputs.
An alternative approach is to introduce specialization effects directly into the production
function as we do in chapter 9. Labor specialization occurs when the human capital of the
operator increases with the use of the asset; that is, learning by doing occurs as the farmer
E(θ)
=
