Land Quality and Agricultural Productivity: A Distance Function Approach
better describe the land quality environment typically faced by farmers, and thus offer more
precise comparisons of productivity. This paper applies an example of this new type of land
quality indicator to a study of cross-country technical efficiency in agricultural production.
In this paper, we address two questions. Do variations in land quality between countries
affect their technical efficiency, and thus their productivity? If so, how much of the measured
inefficiency can be attributed to poor land quality? First, we describe a land quality index (LQI)
derived from new global land-cover data (generated from satellite imagery) combined with geo-
referenced data on soil qualities, temperature, and precipitation. The LQI is then used in a
distance function model to measure the impact of land quality on differences in agricultural
efficiency for a cross-section of 110 countries. Distance functions are the first step in
constructing intertemporal Malmquist productivity indices (MPIs). Computation of the
Malmquist productivity index, which is a composite of four distance functions, requires data
from two (or more) time periods. Unrestricted productivity growth and land quality limited
productivity growth are compared for the sample of countries to highlight the role of land quality
in agricultural productivity growth.
A NON-PARAMETRIC MODEL TO DECOMPOSE LAND QUALITY AND NON-LAND
QUALITY EFFECTS
Land Quality and Technical Efficiency
Figure 10.1 depicts a production system with a single input and single output. The line describes
the technology frontier for this system. Any producer whose input/output combination lies on
the production possibility curve is said to be technically efficient. Observations that lie below