Abstract

The technical and allocative efficiency of broadacre farmers in a southern region of Western
Australia is investigated over a three-year period. Applying data envelopment analysis (DEA)
and stochastic frontier analysis (SFA) reveals there is some inefficiency in each year, which
decreases over time. The distributions of technical efficiency in each year are positively
skewed toward higher efficiency levels, indicating a majority of farms produce close to their
maximum technical efficiency. DEA and SFA produce similar efficiency rankings of farms
yet DEA rankings are more stable.
The relationships between farm-specific variables and the DEA and SFA efficiency scores are
investigated. There is evidence that farmers benefit from using at least a small amount of
tillage, rather than using ‘no-till’ practices. Education levels and farmer age are found to
positively influence technical efficiency.
Using a DEA profit efficiency model, the duality between the directional distance function
and the profit function allows the decomposition of economic efficiency into its technical and
allocative components. Greater gains in profitability are possible by improving allocative
rather than technical efficiency. Technically efficient farms are not necessarily allocatively
efficient. Also, Tobit regression results indicate that the variables associated with variation in
technical efficiency are different to those explaining the variation in allocative efficiency.
Key words: farm efficiency, data envelopment analysis, stochastic frontier analysis