Peak car: evidence indicates that private car use may have peaked and be on the decline

By Phil Goodwin

‘Peak car’ is the idea that car use may not saturate, but turn down. The suggestion was that ‘approach to saturation’, when disaggregated, would show that different places and groups only differ in the pace and level at which they approach saturation, whereas a curve about to turn down, when disaggregated, would show that some places and people have already made the turn. So now to see if such evidence exists.

It is clear that the planner and forecaster working in 1970 would have no more evidence to doubt the possibility of continued uninterrupted growth in car demand than their parents to doubt the continued growth of road public transport use in 1950 or their grandparents to doubt the continued growth of rail use in 1918. But will they be proved right even though their parents and grandparents were proved wrong?

Yes but, you will say – or at least I hope you will, because there is an evident flaw in my comparison. Such curves and trends are caused by real world forces, and demonstrated by real world data. What is there in the real world that could make the car curve of the future look like the rail and road public transport curves of the past? This is a question we have to address without any benefit of hindsight at all, and I would like to suggest that we should consider three contending hypotheses, not two – continued growth, saturated steady-state, and the ‘peak car’ hypothesis of trend reversal leading to decline. I’ll work on it, and I hope some of you will as well.

Thanks to LTT readers for letters and advice on this question, which indicate that the same issues have been occupying many minds, and for a considerable time. I won’t say that the following points add up to a coherent and completed research programme, since people have used different methods and definitions, and a different focus, but they surely add up to a prima facie case that the idea of peak car cannot be dismissed as silly. Something important is happening, and has been for years. Of the many different sources of evidence, I will just mention five at the moment:
First. My colleagues Kiron Chatterjee and Geoff Dudley a year ago gave a seminar at the University of Manchester where they reported that between 1992 and 2007 driving license holding had decreased from 48% to 38% for 17-20 year olds, and from 75% to 66% for 21-29 year olds. Their proportion of trips as car driver had correspondingly reduced.

Second. David Metz, in his letter to LTT of 9 July, wrote that in London 'peak car use came and went at least 15 years ago, when none of us noticed'. There has been a decline in private transport’s share of trips since at least 1993, when it was 50%, to 41% in 2008. He pointed out that 'historically, car use has invariably increased as incomes have risen. So it is remarkable that this trend has gone into reverse in London, a prosperous world city with a growing population.'

Third. In the Sustainable Travel Towns report by Lynn Sloman and colleagues, car driver trips per person went down by 9%, and car driver distance per person by 5% to 7%, between 2004 and 2008. But, interestingly, when Sally Cairns compared these results with the National Travel Survey results for other towns of similar size, she found thatcar use had gone down there as well, though not by as much: car driver trips per person were down by 1.2% and car driver distance per person fell by 0.9%.

Fourth. Studies by Carmen Hass-Klau of the impact of building new urban tram systems in European cities found that car ownership was reduced in the neighbourhood of the trams, by an average of 13%, even though these areas were also affected by gentrification and increased property values as a result of the same improvements: they became richer, but with lower car ownership.

And fifth. The new National Travel Survey, published last month, enables a longer series to be put together (by chaining the 1999, 2004 and 2010 reports) with a little juggling to allow for changes in presentation and definition). The results are shown above, for car trips, and car mileage travelled, including drivers and passengers. These are compared with the same information for the total of the main ‘alternative modes’, walking, cycling, local bus and rail travel.

The turning point in the growth curve seems to date from about 1992 – i.e. during the previous Conservative administration, at about the time that it was abandoning ‘predict and provide’ as its policy intention – and was re-inforced during the New Labour period in spite of the fact that most people (including me) did not really consider that the policy measures to reduce car use were substantial enough to have an effect. That frightening phrase ‘when none of us noticed’ may come back to haunt us.

By now there is some sign that the very long downward trend in walking, cycling and public transport use has bottomed out, and just started to increase, though the turn was later, and smaller, than the reversal in the car trend. Between 1999 and 2009 the miles travelled by car per person reduced by 500 miles a year, while the miles travelled by walking, cycling, local bus and rail only increased by 133 miles a year, suggesting that a little over a quarter of the decline in car use could have been accounted for by a like-for-like mode transfer of journeys, the rest being accounted by a shortening of journey distance and the abandonment of some car trips altogether. So people were changing their destination choice and propensity to make car trips, not only their modes of travel.

We must assume that the very latest figures are influenced by recession and therefore may have exaggerated the trend. But most of the results above definitely precede the recession: this is a real effect, not a temporary blip. The evidence that car use may have passed its peak and be on the way down is especially significant since the cases cited – young people, the capital, towns with the most enlightened ‘smart choices’ policies and towns improving public transport most dramatically – are associated with positive messages for the future, not negative ones.

New land-use trends, particularly in the increased popularity of central and inner cities, and possibly internet use, would also seem to be important. I don’t think that this evidence is yet conclusive – very much more substantial study is required than can be done in the preparation of a series of LTT articles – but surely it is strong enough to justify such a study?

Stephen Plowden in the last issue wrote that it does not really matter what has caused this shift of trend: the important thing is to recognise that it has happened, and change plans accordingly. In one way he is right: a traditional blinkered official view 'we will assume it has not happened until someone has completely explained it' would not do at all.

But it is important to know why, all the same, not least because if the shift in trend has been partly or fully caused by transport policies and prices, then the sensitivity of choices to such factors must be bigger than has previously been thought, and that would affect the confidence with which new policies can now be adopted. But what are those policy implications? They may not be straightforward or unambiguous. That’s for next time.