FORECASTING EQUILIBRIUM MOTOR VEHICLE HOLDINGS BY MEANS OF DISAGGREGATE MODELS

This paper reports on the development of a methodology to forecast the fuel efficiency of household motor vehicle holdings in the United States. Forecasts are made by using a new partial equilibrium model of the operation of the motor vehicle market. In a break with the prevailing aggregate stock-adjustment approach, the approach described here incorporates household-level discrete choice models to explain vehicle holdings and scrap-page decisions. Given assumptions on the design and prices of future new vehicles and fuel prices, the behavior of a demographically weighted sample of households is simulated and equilibrium conditions for the vehicle market are solved each year in the forecast period. The paper presents predictions of household vehicle holdings, new-vehicle sales, used-vehicle scrappage, and the resulting average vehicle fuel efficiencies under two future scenarios. (Authors)