THE EFFECT OF CAR OWNERSHIP AND INCOME ON BUS TRAVEL

Abstract

The national travel surveys of 1964/65, 1972/3 and 1975/6 have been examined to determine the extent to which the 70 per cent increase in car ownership over the period 1964-1976 has affected the use made of stage bus services. From consideration of the frequency of trip-making by bus in households owning no cars, one car, and more than one car, and the way in which this depends on household size, it is estimated that in 1964 each additional car brought into domestic use was directly responsible (because its users travel less frequently by bus) for a loss of 380 bus trips per year, while by 1976 this loss had dropped to 300 bus trips for each additional car. Over the period 1964-76 this direct effect of increasing car ownership accounted for some 40 per cent of the observed decline in bus patronage. In addition, there is a second-round loss of patronage as bus operators respond to the reduction in revenue by cutting services and increasing fares. If the services were required to break even, then this second-round loss would be almost as great as the patronage lost as a direct result of the newly-acquired cars. However, in the period studied, bus services in the UK moved from break-even to an overall deficit approaching some 30 per cent of costs: Within a given category of household car ownership, the effect of income on bus use was also investigated. In cross-section, the use of buses clearly increases with income, but any tendency for people to use buses more as real income levels grow with time appears to be very much smaller than the cross-sectional effect with an elasticity of demand with respect to real income no greater than, and probably a good deal less than 0.3. The national trend in bus patronage was analysed to estimate elasticities of demand with respect to fares and level of service, after the reduction in patronage due to increasing car ownership had been taken into account. /TRRL/

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 10 citations exist upstream.