JManc
May 16, 2009, 6:36 AM
http://www.infrastructurist.com/2009/05/07/stats-whiz-takes-crack-at-explaining-why-americans-are-driving-less/
Nate Silver, the precocious statistician who knew the final results of the 2008 election before it even happened, has taken on the tricky and high-stakes question of why Americans are driving less.
http://www.infrastructurist.com/wp-content/uploads/american-driving-stats-300x256.jpg
“High stakes” because the decades-long trend in increasing car usage has been a defining force in post-war America–shaping everything from our daily routines to our community structures to our natural landscape. But that trend line suddenly broke downward in 2007 and are now, month after month, U.S. motorists are logging fewer miles. So what brought about the reversal though and will it last? Was it higher gas prices? Or a tanking economy leading to fewer commuters? Or was something else at work — a slow-dawning realization that 50-mile commutes are madness and that life is bit better if we take a more balanced approach to transportation?
Writing for Esquire, Silver built a regression model to tease apart the various causes — and found that gas prices and unemployment alone don’t quite explain the current data:
The model predicts that given a somewhat higher unemployment rate but much lower gas prices, the lower gas prices should have won out: Americans should have driven slightly more in January 2009 than they had a year earlier. But instead, as we’ve described, they drove somewhat less. In fact, they drove about 8 percent less than the model predicted.
So–as we’ve speculated about on this site recently–maybe there is some kind of cultural shift at work here.
But the jury isn’t in yet:
There is strong statistical evidence, in fact, that Americans respond rather slowly to changes in fuel prices. The cost of gas twelve months ago, for example, has historically been a much better predictor of driving behavior than the cost of gas today.
The real test will come as the summer unfolds and Americans have had time to get “used to” lower gas prices.
Gas prices peaked in July of last year. So if Silver’s 12-month lag effect holds in this case, August and September will be telling months. We’ll be watching.
Nate Silver, the precocious statistician who knew the final results of the 2008 election before it even happened, has taken on the tricky and high-stakes question of why Americans are driving less.
http://www.infrastructurist.com/wp-content/uploads/american-driving-stats-300x256.jpg
“High stakes” because the decades-long trend in increasing car usage has been a defining force in post-war America–shaping everything from our daily routines to our community structures to our natural landscape. But that trend line suddenly broke downward in 2007 and are now, month after month, U.S. motorists are logging fewer miles. So what brought about the reversal though and will it last? Was it higher gas prices? Or a tanking economy leading to fewer commuters? Or was something else at work — a slow-dawning realization that 50-mile commutes are madness and that life is bit better if we take a more balanced approach to transportation?
Writing for Esquire, Silver built a regression model to tease apart the various causes — and found that gas prices and unemployment alone don’t quite explain the current data:
The model predicts that given a somewhat higher unemployment rate but much lower gas prices, the lower gas prices should have won out: Americans should have driven slightly more in January 2009 than they had a year earlier. But instead, as we’ve described, they drove somewhat less. In fact, they drove about 8 percent less than the model predicted.
So–as we’ve speculated about on this site recently–maybe there is some kind of cultural shift at work here.
But the jury isn’t in yet:
There is strong statistical evidence, in fact, that Americans respond rather slowly to changes in fuel prices. The cost of gas twelve months ago, for example, has historically been a much better predictor of driving behavior than the cost of gas today.
The real test will come as the summer unfolds and Americans have had time to get “used to” lower gas prices.
Gas prices peaked in July of last year. So if Silver’s 12-month lag effect holds in this case, August and September will be telling months. We’ll be watching.