For most of the past century, buying a new car was less a purchase than a milestone. It was the thing you did when the job came through, when the family grew, when the budget finally felt like it had room to breathe.

That new-car smell was shorthand for arrival.

Detroit built an entire economy on that ritual. Automakers planned factories, staffed assembly lines, and set Wall Street expectations around one comfortable assumption. Americans would keep walking into showrooms in roughly the same numbers, year after year, the way they always had.

For decades, the assumption held. The industry treated 16.5 million to 17 million annual sales as the sign of a healthy market, a floor the country reliably bounced off after every recession and oil shock. In 2016, sales climbed to an all-time high.

Then the floor started to look like a ceiling.

New research suggests the United States may never sell 17.6 million cars in a single year again, and the cause has almost nothing to do with where interest rates or gas prices happen to sit in 2026.

Why 17.6 million became the number the auto industry trusted

For years, the math felt almost automatic. The industry assumed roughly 1% annual growth that tracked the rising population, which made 16.5 million to 17 million sales feel less like a goal than a birthright, according to CNBC.

Plan the factories around it, hire to it, promise Wall Street against it.

More Automotive:

The pool of people who can actually afford a new car has been thinning for a while. The average new-vehicle transaction price now sits near $46,000, with the typical monthly payment around $812 and finance rates parked in the mid-6% range, according to Cox Automotive.

Roughly one in five new-car loans now carries a payment of $1,000 or more, a figure once reserved for luxury vehicles. Automakers are papering over the gap with discounts, pushing average incentives to roughly $3,300 a vehicle by May 2026, up more than 20% from a year earlier, with electric models marked down by more than $10,000 apiece, according to JD Power.

Cox Automotive expects only 15.8 million new-vehicle sales in 2026, down about 2.4% from the 16.3 million the country managed in 2025, the firm said. “Our 2026 forecast reflects a slowing market, but still a good one,” said Jeremy Robb, chief economist at Cox Automotive.

The softness is not a one-year blip. The market is “steady and solid” even amid geopolitical noise and affordability strain, said Chris Hopson, manager of North American light vehicle sales forecasting at S&P Global Mobility, whose team also sees 2026 landing near 15.8 million.

The U.S. hit a record near 17.6 million vehicle sales in 2016, a peak it may never reach again.

andresr / Getty Images

What Bain found when it ran the numbers on car demand

Here is where the story turns from cyclical to structural. Analysts at consulting firm Bain & Company concluded that several signals point to a market about to shrink further, with sales potentially falling by more than 2 million units by 2040, according to Bain & Company.

The driver is demographics, not discounts. The old 1% growth rule depended on a population that kept expanding, but birth rates have fallen, and net migration is expected to slow for the next 15 years, said Gottfredson, as reported by CNBC.

The U.S. fertility rate sat near 1.6 births per woman in 2025, below the 2.1 replacement rate, according to the Centers for Disease Control and Prevention.

What strikes me in the Bain math is that much of the decline is already written. The people who will turn 16 in 2040 have already been born, so the size of the future driving-age population is essentially locked in, said Gottfredson.

Related: Bank of America just unleashed new home and auto loan perks

Behavior is changing on top of that. Half of 16-year-olds today hold a driver’s license, down from nearly 70% of 16-year-olds between 1966 and 1984, said Gottfredson. He had once pegged 2030 as the year volumes would slip below 14 million, then pushed that out because self-driving cars are arriving slower than he expected, according to CNBC.

In my read, the most overlooked signal is how fast cars leave the road for good. The rate at which vehicles are scrapped or exported, a figure Bain calls the most direct indicator of a coming decline, which Bain pegged near 6% in 2000, CNBC added.

The trajectory tells the story better than any single year does.

  • 2016: A record near 17.6 million vehicles, the high-water mark the industry still measures itself against, according to CNBC.
  • 2025: About 16.3 million sold, a genuinely strong year that still fell short of the peak, Cox Automotive indicated.
  • 2026: A projected 15.8 million, down roughly 2.4%, Cox Automotive and S&P Global Mobility confirmed.
  • 2040: More than 2 million units below recent levels, with no clear path back to the record, according to Bain & Company.

How a shrinking car buyer pool changes what you pay

This is where an abstract forecast lands in your driveway. A smaller, structurally constrained market does not deliver cheaper cars. It usually does the opposite.

Buyers are already voting with their budgets, with hybrids the only major powertrain group gaining ground through early 2026 as shoppers chased lower fuel bills without the upfront premium of a full-electric vehicle, according to Cox Automotive’s Kelley Blue Book data.

When I lined the forecasts up against the 2016 record, the affordability story and the demographic story started looking like facets of the same problem. Fewer buyers who can stretch to buy a new car push automakers to produce fewer, pricier, higher-margin models. This prices out the next wave of would-be buyers, and the cycle feeds itself.

The alternatives keep multiplying. Younger people increasingly reach for Uber (UBER) or Lyft (LYFT) instead of a car payment, said Sam Fiorani, vice president of global vehicle forecasting at AutoForecast Solutions, a firm that expects U.S. sales to hover near 16 million through 2033.

If robotaxis turn cheap and widespread within 15 years, the share of licensed adults could slip to around 85%, and vehicles per driver could fall from 1.2 to 1.1, the equivalent of 10% to 20% of households dropping a vehicle, according to Bain & Company.

For a household, that is the gap between a two-car garage and a one-car garage, or no car at all. For your kids, it may mean the new-car milestone that defined their grandparents’ adulthood simply never arrives.

What comes after “peak car” for automakers, buyers

Thriving companies will succeed because they stop chasing a volume number that is not coming back. Ford (F), General Motors (GM), and Toyota (TM) are already leaning into hybrids, trucks, and services that earn more per sale, betting the path to profit runs through margin rather than sheer unit count.

The 2016 record is not a target to which the industry is quietly climbing back. It indicates a market that has changed shape, and the smart money is building for the smaller, older, and more selective country on the other side of it.

For buyers, the real takeaway is not about timing the next incentive. It is that the rules of car ownership are being rewritten while the lot out front still looks full.

Related: Goldman exposes hidden crisis building in U.S. auto debt