FORECASTING CAR OWNERSHIP

Accurate estimates of future car ownership are important because the level of car ownership is a primary determinant of personal mobility, modal split, total vehicle miles of travel and energy use in the transport sector. Different methods of forecasting car ownership are reviewed and their forecasts compared. Income and the cost of motoring are suggested to be the key explanatory variables with minor influence from saturation effects. Public transport availability may be an additional important factor for urban areas. As the unit of ownership is commonly the household, forecasting household car ownership rather than on a per capita basis may be more behaviourally consistent and allows consideration of the difference between the marginal utilities of the first and subsequent cars with a household. A summary of some earlier unpublished work presents a model, and its results, for forecasting urban car ownership using these ideas. Extensions of this model are proposed to forecast total car ownership. These extensions take into consideration the effects of differences in household structure and location and the causes and consequences of a move to smaller cars. While no completely formulated and calibrated model is put forward, a framework is presented which may suggest future research work using data from the 1976 census. /TRRL/