Grow, peak or plateau - the outlook for car travel

Abstract

At the request of the New Zealand Ministry of Transport, a meeting of UK researchers and commentators was convened in London on ‘peak car’, a hypothesis that per capita car use is close to its maximum level, and may stabilise or turn down. In both countries car traffic forecasts have been overestimated, though current official thinking is that recent levelling off and decline in car use is a temporary result of poor economic conditions, and that a resumption of strong growth is expected in future. Research findings include features that may show evidence of structural changes in the influence of economic and other drivers of car travel, and/or demand reaching natural saturation levels where further growth gives little benefit to travellers. The discussion reflected diverse experience and judgements, not all in agreement, and mostly related to UK experience, though influenced by a growing international research literature, mostly empirical in focus, showing some important similarities in many other countries.

Key features included:

• There have been strongly divergent trends in different locations (the main areas of decline being in cities, with growth continuing in many lower density areas) and among different groups (young men showing a decline, women an increase), with noticeable differences by journey purpose, length and mode (rail use has grown strongly at the same time as car use has fallen). As a result the trends in national aggregate totals have been damped, not fully revealing underlying causes, and disaggregate experience has varied not only in size but also in direction of change.

• At the aggregate level, evidence of shifts in demand predates recent economic difficulties, including decoupling of traffic growth and economic growth, reductions in propensity to learn to drive, changing land use and migration trends (which had previously been dominated by movements to areas encouraging higher car use, and in the 2000s reversed), growth of internet use, and decline in traffic levels in London and some other urban areas. These shifts have mostly been observed in the 1990s and early 2000s, including periods of strong economic growth.

• Over the same period, there have been changes in transport policy and travel conditions favouring public transport, walking and cycling in some places, changes in tax effects on prices (especially company car use), parking control, congestion levels, and fuel prices. It has often been the case that policy changes, though worthy, are deemed to have rather small effects on travel demand, but there is a view that the cumulative effects may have been larger than expected, the resulting car use levels representing a new equilibrium to the prevailing conditions.

Taken together, these features would imply that per capita car use is influenced by both economic and other structural factors, cannot be converted to total traffic levels simply by multiplying by population, and demands serious re-examination. Although all these features are researchable, there is not currently a professional consensus backing either the official forecasts or any specific alternative, and a strong implication is that methods of policy formulation and project design should test robustness to a much wider span of feasible futures than is reflected by traditional methods involving rather narrow bands of statistical uncertainty.