Car Sales Hit Their Natural Limit, Again
U.S. car and light-truck sales set an all-time record in 2016, with almost 17.6 million vehicles sold. It's not looking like that record is going to be broken this year. Report Bloomberg's Jamie Butters and David Welch:
March was supposed to be the month U.S. auto sales rebounded from decreases in January and February. Instead, ample discounts were unable to spur demand for models like GM’s Chevrolet Malibu and Ford Fusion, which are being surpassed by crossovers as the new American family vehicle of choice. Deliveries of those models each plunged by more than 35 percent in March.
Overall, March sales came in at a seasonally adjusted annual rate of 16.53 million vehicles, according to WardsAuto, which if it holds up for the full year could make this the worst year for auto sales since 2013.
This is a negative and potentially significant business-cycle indicator, with implications for the stock market, manufacturing employment and the risk of recession. But it also seems to be part of a long-run trend that was interrupted for several years by the Great Recession and subsequent recovery and is now reasserting itself. We Americans appear to need fewer and fewer cars -- per person, that is.
On a per-capita basis, light-vehicle sales have been on a long, if unsteady, decline in the U.S. since the 1970s. 1 That's mainly just because there are lots of used cars out there. The average age of cars and light trucks on the road in the U.S. is now 11.6 years, up from a little over five years in 1969.
Back in the golden age of American auto manufacturing, cars weren't made to last long. In the 1930s, General Motors president Alfred P. Sloan Jr. even pioneered the practice of planned obsolescence -- changing styles annually so that cars would start to look out of date after a couple of years. In the 1970s, tough economic times and increased foreign competition led to a heightened focus on quality and durability that, while not entirely putting an end to planned obsolescence, led to big changes in consumer and manufacturer behavior. (I would also bet that Consumer Reports magazine played a big role in this, although the evidence on that is mixed.) Cars got better, and styles changed more slowly.
Over the past decade, though, there's been another thing weighing on auto sales. Demand for the transportation services cars provide, as measured by vehicle miles traveled per person, seems as if it may have peaked.
Car ownership -- measured as vehicles owned per person and per household -- also peaked in 2006, according to Michael Sivak of the University of Michigan's Transportation Research Institute. And Sivak and his colleague Brandon Schoettle have documented a steady decline since the 1980s in the percentage of younger Americans (ages 16 to 44) with driver's licenses (although the trend has been going in the opposite direction for older Americans, especially those 70 and over).
There was a time, four or five years ago, when such statistics could credibly be cited as evidence of a sea change in American lifestyles. Cities were on the rebound, suburbs were on the decline, and those danged millennials just weren't buying cars or houses or anything. Since then, population growth has been faster in the suburbs than in cities, public-transit ridership has declined, and millennials have started buying stuff. Some of what was going on was just the aftereffects of a really terrible recession.
Still, the post-recession rebound in miles traveled does appear (in the above chart) to be fading. After decades and decades of upward movement, except during recessions, the trendline is now looking awfully flat. Why is that? I would guess that it's a combination of a real if modest shift in demand toward walkable, transit-friendly neighborhoods (which would be bigger if those neighborhoods weren't so danged expensive); the continuing rise of telecommuting; and the still-high number of Americans who have dropped out of the labor force (the prime-age labor-force participation rate, while it has risen for the past couple of years, is still, at 81.7 percent, well below the 84.4 percent of early 2000). 2
As someone who is always looking for signs of peak auto or even peak human environmental impact, I may be inclined to overstate the importance of what could turn out to be just a bad couple of months for car sales. But I do think it would be prudent to expect that new light-vehicle sales in the U.S. will over time continue to trail population growth. And with U.S. population growth projected to average only 0.6 percent a year through 2060, that doesn't leave a lot of room for sales growth.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
The Bureau of Economic Analysis's monthly light-vehicle sales data only goes back to 1976, but other, less-complete data sources (such as the BEA's numbers for just cars) indicate that the all-time peak in per-capita sales was in the early 1970s.
Yes, the rise of car-sharing and ride-hailing services is probably having some kind of impact, too, although it's a little hard to say at this point what exactly it is. It may be putting upward pressure on miles traveled (and depressing transit usage) in some cities. And while you'd think these services would depress new car sales eventually, so far Uber and Lyft's aggressive efforts to get more drivers on the road may have actually led to more car purchases.
To contact the author of this story:
Justin Fox at email@example.com
To contact the editor responsible for this story:
Brooke Sample at firstname.lastname@example.org