Land Quality and Agricultural Productivity: A Distance Function Approach
12
be underestimated relative to the productivity level that they can actually achieve. For the true
productivity potential of a country to be measured, it is necessary to separate the share of
inefficiency that is attributable to land quality from that which is due to inefficient use of inputs.
Productivity measures that do not account for differences in land quality will thus overestimate
the potential of countries with poor land quality, and possibly lead to unsound policy decisions.
For example, because low levels of land quality restrict production possibilities, increases in
input use in low LQI countries will have a smaller impact on agricultural productivity than in
countries with a high LQI. The reduced marginal return further limits the use of fertilizers and
other inputs. For these resource constrained environments, suitable technologies and policies are
needed, such as drought tolerant and nutrient efficient crops.
In this study, we have assumed that all differences between the unrestricted and limited
frontiers are attributed to land quality. In reality, there are likely to be other factors that
contribute to such differences as well. Therefore, the land quality-constrained measures
developed in this work must be viewed as upper bounds on the contribution of land quality to
efficiency/productivity differences. Future research is needed to disentangle these multiple
factors—including the influence of non-conventional inputs such as research and education—
that contribute to differences in efficiency and productivity. Development of time-series data on
land quality would also permit improved analysis of changes in productivity over time.
REFERENCES
Benjamin, D. (1995), ‘Can unobserved land quality explain the inverse productivity
relationship’, Journal of Development Economics, 46: 51-84