Following on from my post this morning around the Harbour bridge, I was pointed to this article from the US on trends on the amount of vehicle miles travelled each year. The post starts as

It’s now common knowledge that annual changes in the volume of driving no longer follow the old patterns.

For 60 years, the amount of vehicle miles traveled (VMT) rose steadily. Predicting more driving miles next year was like predicting that the sun would rise or that computer chips would be faster. The only direction seemed to be up.

Then, after 2004, per-capita VMT fell 6 percent, which has led to a decline in total VMT since 2007.

The most recent data are from July, traditionally America’s biggest month for driving. In July 2012, Americans clocked over 258 billion miles behind the wheel, a billion fewer miles than the previous July despite a slightly stronger economy and cheaper gasoline. In fact, you’d need to go back to 2002 to find a July when Americans drove fewer miles than July 2012.

Has America’s long increase in driving turned a corner or just taken a prolonged pause? The answer matters a lot.

They have then gone further and done some modelling of what the graph might look like based on a couple of scenarios:

But what surprised my was just how similar the graph on the numbers is to what we see with the harbour bridge so I decided to do some similar modelling. I looked at the average year on year percentage increase for four different time periods and modelled them out to 2027, 15 years from now.

  • From when we can first get data for the Harbour bridge in 1975 through to when the volumes peaked in 2006
  • From when we can first get data for the Harbour bridge in 1975 through to 2012
  • The last decade from 2002 to 2012
  • Since traffic peaked in 2006 through to 2012

Doing so gave a spooky familiar graph even though they are looking at miles travelled while we are looking at vehicles per day over the bridge. Here is what our result looks like:

As I said they are remarkably similar and just like the post we have to ask ourselves, are we really going to see a return to the old days? Sure there will probably be a bit of a pick up in the future if the economy improves but it seems hard to imagine it will be anything like what we have seen. Especially because the drop off in vehicle numbers preceded the global financial crisis, as it has done all over the developed world and because we know that on this particular route that thousands of people have chosen to use the Northern Busway once that opened. And are likely to continue to as improvements continue in Auckland’s transit systems.

The author of the US post comes to the same conclusion:

It won’t be surprising if an upturn in the economy leads to some increase in driving, especially if gas prices don’t also surge. But it’s harder to imagine that we will switch back to the sizeable increases in VMT that took place almost every year for six decades after World War II. Even if driving continues to increase at the rate of population, this would be a long-term slowdown that should correspond to major changes in transportation policy.

With the important observation that:

While we can’t yet see “the new normal,” it’s a good bet that it won’t be the same as the old normal.

Do we really want to be making $5 billion bets based on this uncertainty? Or should we pay more attention to this seven year old change, as it may well be generational, as the article urges:

Congress needs to stop trying to build out our grandparents’ transportation system.

Well certainly not unless they can guarantee the petrol prices our grandparents paid too.

Share this


  1. Pretty hard not to conclude that the safest bet, the conservative sensible action, would be to work with the trend generated by the last 10 years in each case. I understand that is standard industry practice anyway, but also in this case it looks by far the most reasonable course in a very unpredictable world economy. Unless you come to this data with a whole lot of assumptions about how the world ought to be and therefore are determined to reject what the observed numbers say. So you would have to conclude that any planner working with the pre 2002 timescales is making a highly risky and radical claim to know the future and is being not objective.

  2. WOW, a picture tells a thousand words. I’m with Patrick on this one and would go with the flat line at best. The funny thing to comtemplate is the pending actions of the industries and all the businesses that run off the back of vehicle travel. In the face of flat line at best and probably declining ‘consumption’ what are they going to do to get us driving again or will they just quietly move onto other indutries. A really interesting line of analysiswould be to data mine the WOF info where they record the distance travelled between WOF tests. Given a healthy dose of scepticism about the data accuracy it could well answer questions about who was driving less; city, urban, provincial, rural. Old cars versus new. Big cars versus small. Petrol versus diesel etc etc. What a wealth of hypotheses could be tested here to dig in behind Matt’s excellent high level analysis. Is anyone doing this sort of stuff? Cheers

    1. good point about how a considerable sector of our economy is built on personal transport, the MTA’s ads opposing changes to the the WOF regime on “safety” grounds is pretty transparently protecting the jobs of many

      a bit of a dilemma really

Leave a Reply