The Contribution of Agricultural Research to Production

Abstract

A LTHOUGH the efficient allocation of resources to and among research and development activities has become an important problem in public finance, to date little is available in the way of methodology to aid this analysis. The assessment of expected benefits from a given research effort is a key element in this analysis. While a high degree of uncertainty about the productivity of research is inherent, measurement of the output of past research effort is one basis on which expected future benefits might be assessed. This paper develops a framework for estimating the output of past research devoted to increasing agricultural production, by the state agricultural experiment stations and the United States Department of Agriculture. It is an aggregate analysis utilizing both time-series data for all United States agriculture and cross-section data for states.' In addition to estimates of the magnitude of the effects on production of this research, estimates of the time lag between dollar expenditures on research and its eventual effect on production are reported. The agricultural sector of the economy is characterized by relatively small firms. The homogeneity of agricultural products and the nature of the inventions or technological changes make it difficult for the farm firm to capture the benefits from research or inventive effort that it might undertake. The most important technological changes in agriculture are typified by plant varietal changes which can be easily reproduced or techniques which can be easily copied. In addition, the size and skill requirements for efficient research conduct means that little research of a biological and fundamental nature is likely to be undertaken by the farm firm. A recognition of this divergence between the ability of farmers to capture