Abstract
A model for motor gasoline demand in Canada is developed by household. The model identifies and separates effects of several responses by the household to a change in gasoline prices such as driving fewer miles, purchasing fewer cars, and buying more fuel efficient cars. It also estimates the manufacturers' response of improving the technology of new automobiles. The size and the composition of the fleet according to the interior volume of four classes of automobiles rather than their natural weight is used. Furthermore, two components of the average fuel efficiency of new cars were identified and estimated. The first is the technical fuel efficiency set by the car manufacturers and the second is the sales ratio of four classes of new automobiles. The use of household expenditure survey data make it possible to experiment with some socio-economic variables such as the percentage of households living in urban areas, number of cars per household and the number of persons in the household who can drive. The relatively new technique of cointegration is also utilized. The results indicate that there are certain advantages associated with the elaborate treatment of the stock adjustment and the fuel economy of the fleet. In general, the estimated coefficients suggest that most of the adjustment after a gasoline price increase comes from miles driven in the short run and from miles per gallon, hence fuel efficiency improvements in the long run. The model gave the total short run (one year) price elasticity of gasoline consumption between 0.312 - 0.313 for the different provinces which is relatively small range. One of the more interesting results is that approximately 10 percent of the household response to a price change in the first year was due to a change in the composition of the fleet to a more fuel efficient vehicle. Approximately 75 percent was due to driving fewer miles while the remaining 15 percent was attributed to a change in the size of the fleet. The intermediate run (five year) price elasticities range from 0.689 to 0.7 09 and the long run price elasticities (ten year) range from 0.97 5 to 1.059. The long run price elasticities exceed unity which does not lend support to the belief that long run gasoline price elasticities are also inelastic. The short term household income elasticities range from 0.301 to 0.306, the intermediate term range from 0.655 to 0.679, and the long term range from 0.868 - 0.949.