A major goal of agricultural policy in many developing nations is the improvement of farm management. Economists have treated aspects of this issue in the literature on technical and allocative efficiency, but much of the work has focused almost entirely on devising techniques for quantifying efficiency differentials. This paper takes the next logical step and attempts to identify sources of such differentials. A simple model is presented relating technical efficiency to general modernization and agricultural information. All three variables are measured among a sample of cotton farmers in Tanzania. Correlation analysis and estimates of modified Cobb-Douglas production functions seem to indicate that general modernization is the more important causal factor and that its impact is primarily labor-augmenting. The note appended to this paper demonstrates that when management is omitted from a Cobb-Douglas production function the direction of bias in estimated returns to scale depends on the manner in which management enters the "true" function. Griliches' [1957] seminal article on this topic implicitely assumes a particular specification that leads to negative bias, whereas alternate, perhaps more appealing, specifications may yield opposite results.

The concept of technical efficiency differences—different levels of output with identical levels of input—is unsatisfactory from a production theoretic point of view. In this paper a model is developed in which differences in non‐conventional inputs and especially information obtained by managers may explain productivity differences between firms. Estimation of the underlying production structure (of a sample of California dairy farms) via a modified non‐homothetic Cobb‐Douglas production function shows the specific impact of information within the neoclassical production framework. This is conceptually and analytically superior to the methodology of frontier production functions.

Abstract:Gold production at an industrial scale developed with the discovery of gold in Australia and the USA in the middle of the 19 th century. Since then the gold production rose exponentially with a rate of approximately 2.0% thus reflecting a first-order production cycle. Within this rise, however, four individual sub-cycles can be identified. The current sub-cycle is predicted to lead from a peak in 2001 of 2,600 tons to a global production of 1,600 tons in the year 2018 or even as little as 780 tons in 2026. Further analysis of these sub-cycles, consideration of declining ore grades and energetic constraints lead us to suggest that the year 2001 indeed could have been the peak-gold year of the main 'Hubbert-style' production cycle. A cumulative achievable gold production between 230,000 and 280,000 tons is derived from the application of the so-called Hubbert Linearization. This compares well with a minimum of about 285,000 tons of combined past production and known reserves and resources.

No abstract

No abstract