An Evaluation of Farmers Price Expectation Process

Most dynamic models involve expectation of one or more variables. Existence of several expectation hypotheses, none of them being preeminent for all situations, makes it difficult to choose an expectation process while building models. In this context it would be helpful if empirical validity of different hypotheses is available for broad categories or situations. This study is aimed at evaluating farmers expectation process in terms of different hypotheses available. Here, the land allocation to chilly crop is assumed to be primarily based on expectation of subsequent prices. The resultant model from different expectation hypotheses are fitted on the acreage under chilly crop. Also, four different types of price indices for the period 1948-50 to 1988-89 are used to estimate and test the models. The best fit model is selected based on R2, Amemiya Prediction Criteria (APC) and forecast accuracy measured by Root Mean Square Error (RMSE). The actual best fit model (ideal model) for price is also identified using the above criteria and compared with the best expectation model used by farmers to assess the accuracy of the expectation form used. The results indicates that the extrapolative model proposed by Turnovsky is the best model for explaining farmers’ expectations. This model also seems to be the ideal model for forecasting prices, especially for the recent years.