Sunday, August 17, 2008

What matters to the US driver: gas prices and how much they drove yesterday

Gas prices have finally changed the behavior of the insatiable American driver. As soon as gasoline hit $4 in July, Craigslist was flooded with the “dinosaur” SUV and lines were forming outside of the Toyota dealership with hopeful buyers of the now-it's-not-so-ugly Prius. I am from Houston and let me tell you that once those crazy Houstonians start selling off their shiny Hummers (yes, they love their Hummers), I knew that something was afoot.
Since January, gas prices have risen 33% and total vehicle miles driven has fallen 4.7% on an annual basis. Why are we driving less? Is it just the price gas? Firms have shed 463,000 jobs since January, real disposable income growth slowed to 3.4%, and since the rebate checks have now run out, income growth is set to slow further. Could these factors have contributed to the reduction in vehicle miles driven as well?

The answer is not really. In a short econometric test, I find that income is not a significant factor, employment is not a significant factor, but gas prices and previous driving habits are very significant factors in determining the driving behavior of Americans. People really don’t like to change their driving habits; it takes $4/gallon for gas and several months of reduced driving to get them to do that.

I use monthly data spanning the years 1993-2008 to test the effect that income, gas prices, unemployment, and previous driving habits have on a driver’s behavior. Total vehicle miles driven describe American driving behaviors, all variables are in log, and the table below lists the results.
Income and unemployment are not presented as results because they have no significant explanatory power in describing how many miles Americans choose to drive. Interesting that consumers will continue to drive even though they are losing jobs and incomes may be falling.

Current gas prices and previous driving behavior are the biggest determinants of driving behavior (vehicle miles driven). First, for each 1% increase in gas prices, consumers reduce their driving by 0.01%. I know that it doesn’t sound like much, but it’s something isn’t it?

Perhaps the most significant result of the test is how tough it is to change driving patterns; in fact, it takes months of practice. Drivers are relatively nonresponsive to gas prices (0.01%) compared to their miles driven in previous months (0.16% on average). If a consumer reduces his driving by 1% 6 months ago, then that translates into a reduction of 0.14% miles today; 1% fewer miles 5 months ago means 0.14% less miles now; 1% fewer miles 1 month ago means 0.16% fewer miles now.

Translation: we are really drivers of habit, and those habits are hard to break! It took gas prices surging quickly and peaking to over $4/gallon to get consumers to start changing their driving patterns.

It is certainly a relief that gas prices are falling, but to what end? Consumers finally started to drive less, but all that hard work is soon to go to waste. Gas prices will continue to abate, and the American driver will start to increase the number of miles that they drive. In just a few months, we will be right back where we started 6 months ago (a 5% annual increase in miles driven).

I am surprised that Nancy Pelosi and Co. do not complain about falling gas prices. It took a lot of hard work on the part of the American driver (and the markets), and emissions from driving are finally falling; isn’t that what Democrats have been pushing for all along?

Please leave comments. Best, Nontruths


  1. We drivers really worked hard! Having listened to Pelosi for years, it is sad to say that she will not respond to this at all. She had to make the big change in her stand on off-shore drilling and that must have been VERY hard for her. Don't wait for another change. Now that school is back in session (in the west), comparison of driving habits year-to-year would be interesting. One always notices the lack of density of cars on the freeway during the summer and rejoices!

  2. The data, as reported by the Department of Transportation, is not adjusted for seasonal factors such as summer driving (less driving, for that matter). So, when you hear the reports on the media that vehicle miles driven has gone up in the Fall, beware that this data is not adjusted.

    The data that I present here has been adjusted, and should negate the effects of shifts in school schedules.

    Thanks for a great comment!