Agriculture Reference
In-Depth Information
In part III of the topic the predictable aspects of nature—its seasonality—become very
important because they limit the degree to which farmers can specialize in production. Most
types of farming are greatly restricted by nature. Both plant and animal crops have “growing
seasons” that restrict the nature of farm production. As a result, farmers are seldom able to
exploit many forms of economies of size and tend not to develop into large corporate farms.
Finally, throughout our topic we attempt to explain farming contracts and organization
in the context of
risk neutrality
. As we note below, this is a significant departure from most
attempts to explain such matters. Risk aversion is an assumption about preferences that
we do not make. There are several reasons for this. First, in modern agriculture, where
the ability to avoid risk though insurance and asset markets is so well developed, it seems
implausible that farmers and landowners would use their land contracts to further avoid risk.
Second, as we show in chapter 6, empirical implementation of even the simplest risk-sharing
hypotheses is difficult, and often impossible, because of the stringent data requirements.
Third, by avoiding the complexity of preference and uncertainty modeling, we are able to
develop models that yield clear and testable predictions. Ultimately, the importance of risk
sharing is an empirical matter. Though we push our risk-neutral model in all directions, in
part II we pause to compare our results with those based on risk sharing, where we find no
compelling grounds to abandon our assumption of risk neutrality.
Modeling Transaction Costs
To summarize, we use a series of transaction cost models in which all parties are risk neutral,
in which all assets are complex, and where nature is both an uncertain and seasonal force.
These conditions make it costly for the contracting parties to identify exactly the input
contributions of their counterparts and, similarly, make it costly to identify the quantity
and quality of the output. In agriculture, nature's seasonal forces limit the gains from
specialization and the ability of parties to monitor each other.
Although the specific models we use vary from chapter to chapter with the details of the
questions we address, we outline the basic framework and its characteristics in this first
chapter. In our models output takes the following general form:
Q
=
h(
land, labor, capital
)
+
θ
,
(1.1)
2
where
)
is the randomly distributed composite input of nature. Exactly how each input is defined
depends on the question at hand. For example, when analyzing the choice of contract
between cash rent and cropshare, the critical inputs are the unpriced land attributes supplied
by the landowner (such as fertility and moisture content) and the labor effort of the farmer.
When analyzing the decision to rent or buy an asset such as land, we make a distinction
between the priced attributes of the land (such as size measured in acres) and the unpriced
Q
is the observed harvested output that is assumed to have a unit price and
θ
∼
(
0,
σ
