Sharing Inputs and Outputs - The Nature of the Farm - page 75

Table 5.1
Summary of incentives for input sharing

Effort Land Other input Output

moral hazard moral hazard moral hazard underreporting

Inputs shared More More No Yes

Inputs not shared Less Less Yes No

is not important. For row crops (such as corn, potatoes, soybeans, and sorghum), tilling
during the growing season is much more important than it is for such small grain crops as
barley and wheat. For these crops, excessive tilling can lead to soil exploitation, which can
be curtailed by reducing the output share to the farmer. As a result, we expect the farmer's
share of output to be lower for row crops and higher for irrigated land. These implications
are consistent with the findings from chapter 4 that, ceteris paribus, irrigated land and hay
land tend to be governed by cash rent contracts (where the farmer's share is 100%) and that
row crops tend to be governed by cropshare contracts. We summarize the model trade-offs
in table 5.1.

5.3 Empirical Analysis: Dichotomous Input and Output Sharing

To test these predictions we use the Nebraska-South Dakota contract data, where the
variables used in the chapter are defined in table 5.2. The British Columbia-Louisiana
data do not contain enough information on input sharing to test these predictions. Our
model allows two sets of empirical tests. First, there are predictions about the relationship
between input shares and output shares, including implications about the choice between
pure

(q = 1 ) and input-output (q = s) cropshare contracts. Second, there are predictions

about the size of the optimal share . Both of these tests result from our model in which

the ability of the farmer to exploit soil

(s)

(r − r ) are

determining parameters. Even though these parameters are not directly observable, there are
distinct situations in which measurement costs are high and the farmer's ability to exploit
the soil is great, allowing us to test the model's implications. In addition, we are able to
observe directly input and output shares and identify situations—farmer, landowner, and
land characteristics—in which we can expect soil exploitation or input measurement to be
problematic.
The division of input costs between farmers and landowners varies across contracts (see
table 5.3). The input shares tend to take one of two forms: Either the farmer bears all input
costs other than the land costs, making his input share 100 percent, or the farmer and the
landowner share the input costs in the same proportion as the output. In other words, if the

and the costs of measuring inputs (m)