Automobile sales in the U.S. are down 35 percent for December 2008 compared to December 2007. My conclusion is that the economy is in surprisingly good shape. Why? There is hardly a single person in the U.S. who actually needs a new car. Cars have become so durable that the easiest purchase to defer is a new car. A neighbor of mine is selling a 1997 Chrysler minivan. It has 125,000 miles on it, works perfectly, and looks almost new because it lives in a garage. He doesn’t trust it for long highway trips anymore. The Blue Book value of this car is $3000 for a private party transaction, $1625 for a trade-in. It would surely be nicer to have a brand new version of this same van (about $23,000 invoice, which presumably is the maximum you’d ever have to pay), but there would be no shame or inconvenience in driving this 12-year-old minivan.
The population growth rate of the U.S. is about 0.9 percent, so we might expect about 2.5 million cars to be sold as a consequence of newly licensed drivers (there are 250 million registered cars on our roads). There are roughly 6 million car accidents every year in the U.S. If we assume that the car is destroyed in 3 million of those, we would expect another 3 million sales due to accidents. That’s an annual sales rate of 5.5 million. In fact, sales are running at a rate of closer to 10 million, which means that 4.5 million cars are being sold as luxury splurges. That doesn’t sound like a depression yet.
[January 6 Update: The New York Times has an article today on the fact that cars have gone from an average of 6 years old to an average of 9 years old.]
The sales rate for cars is stabler than you think. This shows in the age distribution of cars on the road. For 1996 data see http://www.fhwa.dot.gov/environment/conformity/emission/emismeth8.htm .
People do adjust their own new car purchase rate but overall demand is steadier. The lifespan of an automobile is controlled much more by it’s inherent design characteristics and by how well the owners care for the car. It is also significantly affected by how much the car is used. If you decrease use by 10% you also increase life (in years) by about 10%. Other than these adjustments, the major demand driver for demand is the manufactured lifespan of the cars.
There is some short term flexibility, by accepting a lower than usual level of reliability and from driving fewer miles. I’m sure that both are happening right now. The VMT statistics show an 8% drop in miles driven and I suspect that people are making the fix decision more often than the scrap decision.
You see all of this in the 1996 age statistics. Almost every car survives their first 5 years. The next 5 see a modest drop, and years 10-15 are when the heavy scrap rate takes place. I suspect that the current distribution is shifted somewhat toward longer life in the current data. There has been pressure on auto makers to build cars with better lifespans.
Shouldn’t there be a replenishment rate as well? The existing fleet needs to be refreshed eventually. If we assume a full refresh cycle to be about 25 years (pretty generous I think) for 250 million vehicles, wouldn’t that add another 10 million per year to the mix? So I think the numbers work better like this:
2.5 million for population growth
3 million destroyed due accidents etc.
10 million refreshed
That gives a total of 15.5 million. If we’re down to 10 million now, that’s about a 35% drop as would be expected. This doesn’t seem like a splurge time to me.
jzs: My point was that cars on the road today are in such good shape and so durable that replenishment can be deferred for a couple of years. Replenishment is a luxury, not a necessity.
I would argue that a 25 year old car is not in good shape and may not be worth keeping alive (it may cost more for ongoing maintenance than to buy and finance a replacement, not to mention fuel economy and safety), which is why I picked a 25 year refresh.
But even if we pick a 50 year refresh cycle (and I think that’s pushing it, since I don’t know how economical or safe it would be driving around in a 1958 anything) that adds 5 million cars to your original estimate, bringing the total to 10.5 million cars.
We are effectively at a 56 year refresh rate at 10 million cars per year (250 mil divided by the 4.5 mil excess in your numbers). The original point I was trying to make is that I think the economy is struggling substantially, and is not in as good of a shape as you opine. If we assume these numbers are correct, we went form a 25 year fleet refresh to a 50+ year cycle; more than double in one year.
Safety advances — side air curtains, ABS, traction and electronic stability control — are reasons to get a new car, even if your present car is still reliable.