کشش تقاضای توان الکتریکی منطقه ای بخش های صنعتی و تجاری ژاپن

In the assessment and review of regulatory reforms in the electric power market, price elasticity is one of the most important parameters that characterize the market. However, price elasticity has seldom been estimated in Japan; instead, it has been assumed to be as small as 0.1 or 0 without proper examination of the empirical validity of such a priori assumptions. We estimated the regional power demand functions for nine regions, in order to quantify the elasticity, and found the short-run price elasticity to be 0.09–0.30 and the long-run price elasticity to be 0.12–0.56. Inter-regional comparison of our estimation results suggests that price elasticity in rural regions is larger than that in urban regions. Popular assumptions of small elasticity of 0.1, for example, could be suitable for examining Japan's aggregate power demand but not power demand functions that focus on respective regions. Furthermore, assumptions about smaller elasticity values such as 0.01 and 0 could not be supported statistically by this study.

Understanding the power demand of industrial and commercial users is very significant in Japan from a regulatory policy viewpoint. This is partly because such users consume a large part (about 70%) of power served by power companies and partly because the recent power market reforms in Japan have been particularly targeted at these users, who are supposed to actively respond to changes in power charges induced by the reforms. In this study, we estimate price elasticity to capture the characteristics of their power demand, considering the differences of demand patterns in nine service areas in Japan.

Discussions and debates over regulatory reforms for the power market in Japan have often been based on the findings of, and conclusions drawn from studies on the power markets in Europe and the US. Since these studies were conducted on reforms in Europe and the US, their implications might not be valid in a Japanese context. Indeed, it is ideal for us to employ views on the unique features of Japan's power market in discussions about its future reforms. However, to date, only a few studies are available on this topic. Even price elasticity of power demand, which is one of the key parameters that characterize the market, has rarely been estimated empirically. Instead, it has often been assumed, in an ad hoc manner, to be a small value such as 0.1. In our study, we econometrically estimated the regional electric power demand functions for industrial and commercial users and derived the price elasticities. Price elasticity was found to be 0.09–0.30 for the short run and 0.12–0.56 for the long run. Power demand was found to be relatively more price elastic in the rural areas than in the urban areas. This result suggests that: (1) there is some validity in an a priori assumption of 0.1 for the price elasticity of power demand in analyses of the nationwide market in Japan, but not in analyses focusing on the characteristics of regional power markets; and, (2) there is weaker empirical support for a price elasticity of 0.01 or 0. This implies that (reform induced) charge falls would lead to more significant increases in power demand than we have expected so far. Thus, this would bring more benefits, in terms of declining charges, particularly to users in the rural areas. It should be noted that while our findings were derived using an annual dataset, price elasticity of a different magnitude can be found by using a dataset for another market setup, such as an hourly power market in peak load times, as analyzed in existing literature referred to in 1.1. In this sense, it is necessary to exercise some caution in accepting our findings and their implications. We attempted to control for the shocks caused by the reforms using dummy variables in our regional power demand functions. However, we did not seriously consider the endogeneity of explanatory variables in our model. When the reforms are widespread and have created intense competition in the power market, we need to conduct a simultaneous estimation of power demand and supply functions. Moreover, our analysis, which focused on power demand by industrial and commercial users, can be extended to an analysis of power demand by residential users.