Agriculture Reference
In-Depth Information
second-best alternative, we feel that some proxies are better than others. We have attempted
two things to alleviate this problem. First, we have sought predictions from our models that
did not require a measurement of transaction costs. For example, in our prediction regarding
the input-output sharing dichotomy, nothing in the prediction requires such a variable, even
though the prediction falls from a transaction cost model. The input and output shares are
completely observable. Second, we have sought to find practical and effective proxies. In
the empirical risk-sharing literature, the proxies for risk are often self-reported indices of
risk preferences or ad hoc dummy variables (for example, gender). We have never relied
on survey data over preferences. All of our data come from questions regarding the facts
of the farm, and our measures of risk (exogenous variability) come from the actual output
data from the region.
Finally, perhaps because of the empirical focus, transaction cost economics often suffers
from a lack of a generally accepted, well-defined body of theory. Even the very definition of
transaction costs is still not generally accepted. Many economists simply view it as “the cost
of trading” or the “costs of a market transaction.” With such definitions it is easy to see that
the Coase Theorem has little relevance to the study of organization.
4
This limited formal
theory is perhaps another reason for the second-class status of transaction cost economics,
but even this shortcoming is changing over time. As mentioned above, much of modern
contract theory is adopting the transaction cost approach, and there is a movement away
from the simplistic view of transaction costs just mentioned. This topic, with our explicit
definition, consistent modeling, and empirical focus is simply another step forward in the
formal development of transaction cost economics.
Final Remarks
We have found that the pattern of contracts and organization in modern and historical
agriculture is consistent with an effort to mitigate transaction costs. These costs, in turn,
arise from attempts to establish and maintain property rights to complex assets, such as land,
equipment, crop and livestock output, and human capital. Ownership of assets and tasks and
output is ideal in terms of incentives but forgoes the potential gains from specialization.
Thus people contract with each other and transaction costs ensue. Although our topic has
focused on several specific transaction cost issues on the farm, we hope that it has become
obvious that the number of such related issues in agriculture is virtually endless. Whether
the issue is how farmers and custom operators control moral hazard in the presence of great
timeliness costs, or how land rental creates potential for soil exploitation by renting farmers,
transaction costs are ubiquitous.
The theory of organizations has developed a great deal over the past thirty years, and
we want to suggest that agricultural economists, with their knowledge of farming, are well
positioned to take advantage of the fertile ground of economic organization. A number
