MEASURING AGGREGATE SUPPLY RESPONSE UNDER INSTABILITY

In spite of the strong interests that agricultural economists have had in measuring aggregate supply response in the past there are only a few studies that explicitly introduced the risk element into the models (e.g. Behrman, Just, Traill). Behrman quantified both yield and price variabilities by using three-year moving average standard deviations about a simple moving average of yield and price. Just formulated a risk response model by assuming that decision makers would form their risk expectations by geometrically weighting past observations of risk similar to the way in which price and yield expectations are formed. The risk variable was taken to be the square of the difference between the actual explanatory variable (such as gross return) and the expected values. Each lag variable was then divided into observable and unobservable parts similar to the approach suggested by Klein. Assuming the disturbance term follows a normal distribution, an iterative procedure was undertaken to derive the maximum likelihood estimates of the regression coefficients, given values of the lag parameters and the internally generated risk observations. Results obtained from Just's multivariate adapative expectations model often indicated risk to be significant in acreage response