Agriculture Reference
In-Depth Information
Panel A
Panel B
Panel C
Variable input ( k
Farmer's effort ( e
Land attributes ( l
h
k
h
e
h
l
h
e
h
k
h
l
s*
s*
h
h
l
s*
h
k
e
A
E
C
B
F
r
c
w
D
r
s
k
s
s
e*
e
e
l*
l
l
k*
k
Note:
s
* is chosen so that
B FCD
+
=
.
Figure 5.2
Output shares and input use:
q
∗
=
1
(s
∗
)
PREDICTION 5.2
The optimal output share
will be higher when inputs are not shared
(q
∗
=
(q
∗
=
s
∗
)
.
9
1
)
than when inputs are shared
(q
∗
=
s
∗
will rise
PREDICTION 5.3
When inputs are not shared
1
)
, the optimal output share
as the number of other
k
inputs increase.
Prediction 5.2 holds because the distortions are spread over more inputs when inputs are
not shared. This can be seen by comparing figures 5.1 and 5.2. Prediction 5.3 holds because
the distortions will be spread across more margins. Furthermore, this analysis refutes the
argument that proportional input sharing is required for efficient cropshare contracts. The
model also implies a relationship between the fraction of contracts for which input costs
are shared and the farmer's share of the output.
PREDICTION 5.4 The fraction of input costs that are shared will increase as the farmer's
share of the output decreases.
, the input distortion increases as the output
share to the farmer falls. This occurs because the farmer's incentives are increasingly at odds
with those of a complete owner of the crop. This distortion can be viewed as an increase in
the potential benefit of having the farmer pay the same portion of the input costs as his share
of the output. Sharing input costs creates incentives that move the contract toward the first-
best input use, but requires that inputs be measured. The model implies that the costs of
Prediction 5.4 holds because, for any input
k
