Agriculture Reference
In-Depth Information
Indeed, the mean yield may be even less, only 0.8, as long as the probability of
falling below the threshold yield is diminished, farmers may still favor this variety,
particularly on lands that could not previously be cultivated.
One important implication of this, which could arise under certain circum-
stances, would be that aggregate average yields reported for a particular crop or
region may actually decrease as a result of adoption of drought- or stress-tolerant
varieties. This could arise when adoption rates are relatively high on the extensive
margin, characterized by newly cultivated marginal lands with mean yields lower
than the region's previous average but still above the threshold, making cultivation
economically-viable for poor farm households. In such areas, it may be important to
report cropping statistics by smaller aggregate units in order to disentangle effects
of switches to the new variety, at the higher-yielding intensive margin versus expan-
sion on the lower-yielding extensive margin.
2.3.4 The Distribution of Benefits to Adopters of the Technology Over Time
and Different Locations
The fourth factor affecting adoption of a new technology is the timing of its impacts
and differences in its impact in different locations. A given technology's perfor-
mance within the given production contexts—particularly in farming where there
can be significant variations in land quality, water availability, labour, and other
inputs—can vary significantly across the population of potential adopters of that
technology.
The most effective way for a farmer to arrive at a decision about adopting a
new technology is by learning from experience—such as trying the technology on
a portion of one's land or watching a neighbour try the technology, and thus expe-
riencing the results firsthand. Yet, learning-by-doing will encourage adoption only
if the technology reliably yields a perceptible benefit each season. But, depending
upon weather patterns, a drought-tolerant variety may produce a benefit only oc-
casionally, resulting in a less predictable 'stochastic' benefit stream (Lybbert and
Bell
2010
). For example, if drought occurs only 20 % of the time, the learning-by-
doing farmer does not detect any impact in four out of five years. Here, adoption is
expected to be limited if benefits are delayed and uncertain. This can be seen, for
example, in farmers' limited willingness to pay for crop insurance. Similarly, if the
drought impacts one region but not another, farmers in the affected region are more
likely to have experienced the benefits of a drought-tolerant variety and in the fol-
lowing year to plant that variety. The distribution of impacts over time and across
space, can therefore be, expected to strongly affect the rate and pattern of adoption
of this technology across markets.
Thus, the timing of commercial release of the new variety in each location will
be critical. Studies of past droughts suggest that droughts represent periods of sig-
nificant change in farming practices. It is, therefore, reasonable to expect that adop-
tion of drought-tolerant varieties may be observed in a given region only after it
experiences substantial drought.
