Technical efficiency of peasant farmers in northern Ethiopia: a stochastic frontier approach

Empirical works on efficiency of small farmers has been triggered by Schultz's (1964) popular 'poor-but-efficient hypothesis'. Peasant farmers in traditional agricultural settings are reasonably efficient in allocating their resources and they respond positively to price incentives. If farmers are reasonably efficient, as hypothesized by Schultz, then increases in productivity require new inputs and technology to shift the production frontier upward. If, on the other hand, there are significant opportunities to increase productivity through more efficient use of farmers' resources and inputs with current technology, then a better allocation might be essential. But how to measure and compare their efficiency? The concept of efficiency measurement by means of a frontier method has its origin with Farrell (1957). Several different approaches could be applied (see e.g. Fried et al. .... for an overview). We have chosen for the output-oriented or primal approach, where the central issue is by how much output could be expanded from a given level of inputs. The empirical analysis uses a stratified sample of farm dataset refers for the 1996 and 1997 production years. Farmers in the sample are located particularly in Enderta and Hintalo-Wajerat districts of the Tigrai region. A preliminary analysis showed that productivity differences among farmers are rather small (compared to other studies). There appears to be increasing emphasis by policy-makers on investments in new technologies and inputs rather than efforts aimed at improving the efficiency of less efficient farmers. Obviously, the level of efficiency of peasant (small) farmers has important implications for choice of development strategy. As the choice of development strategy, at least partly rests on the policy makers' conceptions of farm/ farmer-level performances. This analysis is intended to contribute to such strategic choices