Dynamic Demand Analyses for Gasoline and Residential Electricity

HE EVENTS of this last year have made consumers everywhere aware of the need for an adjustment in energy consumption patterns. In the 1960's the price of energy appears to have fallen in real terms and as a result the growth of most forms of energy consumption accelerated. Until recently most energy resources -excluding natural gas where the price was kept down by regulation-were in a buyers' market. The emphasis was on sales promotion. In electricity, for instance, utilities offered special incentives to increase usage, such as favorable offers on appliances or free light bulbs, while continuing to rely on their long-standing practice of lower average prices for large consumption. The gasoline market was also soft, as appears from such various types of marketing actions as games and free gifts to increase sales. Excess capacity in refining strengthened the hand of independent marketers, whose emergence often led to price wars. In buying new cars most consumers paid little attention to gasoline mileage. Today, however, the problem is not to increase energy sales but to decrease them. This has raised the question as to whether market mechanisms will be able to adjust supply and demand to each other. Already there are signs that the market may not be given a chance. Public opinion polls show that a majority of the public feel gasoline rationing is needed, and one city, Los Angeles, has ordered all households to reduce electricity consumption by 10 percent or have their power turned off. Clearly there is no widespread confidence in the efficacy of higher prices in curtailing demand, although the price increases themselves are criticized on distributional grounds. Greater knowledge of the effect of prices is needed not only for short-term domestic policy, but also for anticipating developments in the world petroleum market, where a producer cartel is now firmly in charge. In this paper we present results on the demand for gasoline and residential electricity, the two largest items in the consumer's energy budget. These demand functions are estimated using pooled time series data from different states. This disaggregated approach was selected because regional variations in prices and consumption patterns provide valuable information for disentangling the effects of price changes. The data are fitted to dynamic demand functions of the flow-adjustment type; this specification corrects for the lack of adequate data on changes in the characteristics of the capital stocks involved.