The effects of research on yields and costs for five classes of wheat

Effects of US wheat research were investigated using both production function and cost function approaches. A Just-Pope production function was applied to model the effects of research on mean yield and yield variance of five major classes of wheat produced in the United States. Both class effects and time effects are taken into consideration. Fixed-effect and random-effect models are estimated. The random-effect model is found to be more appropriate than the fixed-effect model.;Results show that research during the period studied improves yield stability with no significant effect on mean yields. The findings are consistent with other recent studies on similar issues. More and more objectives, such as quality improvement, are added to agronomic research projects. This may be the reason why the effect of research on mean yield is not statistically significant at the commonly accepted levels.;Translog models are used in the cost function approach. Cost functions are estimated simultaneously with factor share equations using the maximum likelihood method.;Results provided statistical evidence for the existence of spillover effects among different classes of wheat. Hard red winter wheat is the only variety which does not benefit significantly from other wheat research, while soft red winter wheat is the only one which is not a significant spillover source for other classes of wheat. Durum and white wheat are found to be mutually beneficial.;Compared with earlier studies, rates of return to research investments are in the middle range for hard red winter, hard red spring and durum wheat. Rate of return to research investment for soft red winter wheat is rather low. One possible reason is that soft red winter wheat is produced in more than twenty states. Average research expenditures for each state are relatively small. As a result, economies of scale could not be achieved.;The social rates of return to research investments are substantially higher than the private ones. However, compared with studies of high-tech industries, the differences are not large.;Except for hard red spring wheat, research is found to have a significant impact on factor demands. A reasonable explanation is that research activities result in new technology and new varieties. As a result, input structure changes to meet the needs of the new technology and varieties.