The main purpose of this paper was to investigate the government's preference to cause agricultural expenditure growth after World War II in Japan. Assuming that the determination process of public expenditures is modeled through the utility maximization behavior of bureaucrats, I estimate the parameters of the government's expenditure function to agricultural expenditures by three cointegrated regressions and Jahansen's ML approach, which are asymptotically equivalent.
　This paper provides some unique insights into the level of Japanese agricultural public expenditures in post-World War II periods. To begin with, there is considerable evidence of a long-run equilibrium in agricultural expenditures, and associated with this, short-run dynamics which are involved in returning to the equilibrium path. A unique long-run (cointegration) relationship can be found between government agricultural expenditures, total government expenditures and the agricultural population rate of the country. In the long run, agricultural expenditures increased stably with the growth of total expenditures. From my estimates of the parameters, a 1% increase in total expenditures would lead to a 0.95-0.97% expansion of agricultural expenditures, whereas a 1% decrease in the agricultural population rate would result in a 0.4-0.6% reduction. Next, with respect to the short-run view, I found that rapid shrinkage of the agricultural population positively affected the growth of agricultural expenditures and that the adjustment process to the long-run equilibrium level of agricultural expenditures was relatively slow. It is suggested that agricultural expenditures during the 60s and 70s in Japan were expanded temporally due to the rapid constriction of the agricultural population in the process of economic growth. However, in recent years, the agricultural expenditure level has neared the long-run equilibrium.

View full abstract

　A method to measure the index of marketing power of fruit and vegetables by production center has been developed. The index is measured by the SUR model on the inverse demand function by each brand in a market. This model is derived from specifications used with the 'cross-product repackaging' model in the context of the discrete / continuous model by Hanemann. In this case, the index of market power is the expected value evaluating characteristics of the brand.
　Taking account of the attributes of fruits and vegetables, such as the limited availability of data about characteristics from the cause of those being liable to change or the necessity of the prediction for wide range fluctuations in market share, this method is better than other characteristic approach models, such as the hedonic price approach in econometrics or the preference regression model in marketing science.
　The estimated indices of marketing power can be directly used to evaluate the effect of marketing action by a production center in marketing planning or to predict the impact of the entry of a new brand with different characteristics in the analysis of competition in the fruits and vegetables market.

View full abstract