The objective is to develop a forecasting framework designed to provide medium term projections. The models relate ownership and use of vehicles to a set of socio-economic variables for a large sample of the poorest countries in the world. In addition to estimation of elasticities and other relevant parameters, some sample forecasts are produced.

It is proposed that data at household level give the most suitable quantitative and qualitative variables for studies of car ownership in Kuwait.

This paper abandons the conventional reliance on household statistics, and uses as the primary dependent variable the propensity for entry to and exit from car ownership of individuals, segregated by age and sex.