Agriculture Reference
In-Depth Information
provided by the farmer. Measurement costs are likely to be much lower for inputs purchased
directly in the market than for those provided by the farmer. Inputs purchased in the market
will necessarily be measured and sold by a noninterested third party, making it difficult for
the farmer to cheat the landowner by overbilling the landowner or simply carelessly wasting
the inputs.
Thus, we expect that market inputs (fertilizer, herbicide, insecticide, and seed) are more
likely to be shared by the farmer and landowner in the same proportion as the output,
while nonmarket inputs (chemical application, drying, irrigation energy, and harvesting)
are more likely to be the complete responsibility of the farmer.
13
In other words, we
expect
1 for nonmarket inputs. The data support this
prediction. For all three major cropshare rules represented in table 5.4 (50-50, 60-40, and
67-33), the percentage of contracts for which the farmer pays all costs tends to be greatest
for market inputs.
14
For example, in 67-33 cropshare contracts the farmer pays for all
harvest costs 93 percent of the time; he pays for all fertilizer costs only 13 percent of the
time.
Prediction 5.4 implies that the fraction of contracts where
q
=
s
for market inputs and
q
=
will be greater for 60-40
contracts than for 67-33 contracts, and that it will be even greater for 50-50 contracts. This
prediction is supported by the data in table 5.4. For instance, drying costs are proportionally
shared
q
=
s
in 58 percent of the 67-33 contracts, in 77 percent of the 60-40 contracts,
and in 86 percent of the 50-50 contracts. With a few minor exceptions, this relationship
holds for all inputs. Furthermore, it is expected that input shares
(q
=
s)
(s)
will be positively correlated for all crop inputs. We calculated correlation coefficients for
all inputs and found them to be positive in all cases; the null hypothesis of zero correlation
was rejected at less than the 1 percent level.
(q)
and output share
Input Sharing within Contracts
Thus far, we have examined the data for variation in input-sharing rules across contracts. Our
model, however, has implications for the structure of input sharing within each cropshare
contract. Within each contract, market inputs should be treated the same, and likewise for
any nonmarket inputs. For instance, if seed is shared at 60-40, then so should fertilizer,
pesticide, and herbicide be shared at 60-40, and the cropshare should also be 60-40.
Nonmarket inputs (chemical application, drying, irrigation energy, and harvesting) are
predicted to be the responsibility of the farmer. Tables 5.6 and 5.7 show the distribution
of inputs shared within contracts and strongly support the implications of our model. The
data expose the dominance of the sharing rule dichotomy
and make clear
the distinction between sharing rules for market and nonmarket inputs, thus supporting
predictions 5.1 and 5.5.
(q
=
s
or
q
=
1
)
