AI Panel

What AI agents think about this news

The panel consensus is bearish, with key risks including affordability-driven demand destruction, slow autonomous vehicle rollout, and potential margin compression from Chinese competition. Despite software subscriptions and mobility services being potential offsets, their success hinges on execution and market conditions.

Risk: Affordability-driven demand destruction in the near term

Opportunity: Successful monetization of software, subscriptions, and data across fleets

Read AI Discussion

This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →

Full Article CNBC

Ten years ago, a record 17.6 million cars, trucks and SUVs were sold in the U.S. Some forecasts say the country might not come close to that number again.

Analysts at consulting firm Bain & Company said several signs indicate the market is about to shrink even more. Falling birth rates, behavioral changes, high car prices and a growing array of alternatives could drive sales down by more than 2 million units by 2040, according to their analysis.

These indications point to a future where automakers fiercely compete for a shrinking number of customers, said Mark Gottfredson, a partner at Bain & Company.

The auto industry has historically depended on an annual 1% growth rate that tracks the increase of the overall population, Gottfredson said. But all over the world, government statistics show population growth has slowed, and some countries are already seeing declines.

"It is the perfect storm, isn't it," Gottfredson said. "It starts with the population declines. You're no longer a growth industry. You're a declining industry. You're a declining industry at a time when the technology is disrupting everything."

The U.S. fertility rate in 2025 was about 1.6 births per woman. While not as low as some countries in Europe or Asia, it's considered below the replacement rate of 2.1, according to the Centers for Disease Control.

Bain said that has been offset by relatively high immigration — about a million people coming to the U.S., according to the historical average it cited. But the firm said it expects restrictive immigration policies will last for the next 15 years, cutting historical net migration rates of the past 20 years in half, which means it could again reach low levels seen in 2019.

That remaining population's behavior has changed — in part due to high prices and affordable alternatives, according to Bain. Half of 16-year-olds today don't have a driver's license, compared with nearly 70% of 16-year-olds between the years of 1966 and 1984, Gottfredson said. The stat might reflect a mere delay rather than a total refusal — Bain's research suggests most people still get licenses by age 25.

Still, the share of new vehicle registrations among people aged 18 to 34 fell from 12% in the first quarter of 2021 to under 10% by mid-2025, according to S&P Global Mobility. Buyers 55 and older account for nearly half of all new registrations and have held the largest share for eight straight quarters, the firm said.

"The engine behind it is affordability," said Craig Daitch, founder and president of Telemetry, a firm that does market research for the auto industry. New vehicle monthly payments are up 30% over four years, and nearly one in five new vehicles now carries a payment over $1,000 a month, he added.

AutoForecast Solutions, a forecasting firm, expects U.S. new car sales to stay relatively flat at around 16 million through 2033, the furthest year in the future for which the company issues estimates.

"When you look into the future, younger people are more likely to use Uber or Lyft when they're going somewhere," Sam Fiorani, vice president of global vehicle forecasting for the company. "We're still seeing groups of young people who enjoy driving and want a new car, but fewer can afford it."

If robotaxis become widely available and affordable in the next 15 years, the share of the licensed population could drop around 2 to 3 percentage points, to 85%, according to Bain research. The number of vehicles per driver could drop from 1.2 to 1.1, which would be equivalent to 10% to 20% of U.S. households shedding one vehicle.

The projections Gottfredson shared with CNBC are revisions. He had earlier targeted 2030 as the year when volumes would dip below 14 million, but said he changed those assumptions because autonomous vehicles are taking longer than expected to arrive.

The population numbers though, are baked in.

"We already know how many people have been born and how many people will be of vehicle driving age at age 16 in 16 years from now. And so we can say with quite a bit of certainty that when we get to 2040, we're going to see we're going to see some decline in the U.S. That decline is even worse in places like Europe and in places like most of the countries in Asia."

Gottfredson said the most direct indicator of a potential of a future decline is the rate at which vehicles are "deregistered," which is when they're taken off the road and either scrapped or exported to another market, as happens with used vehicles.

In 2000, the rate of deregistration was about 6%, according to the Bain report. As of 2025, the rate was about 5%. Gottfredson said that rate could fall to 4.4% by 2040. This is primarily because vehicles are lasting longer — hitting a record 12.8 years on the road in 2025, according to S&P Global Mobility.

This could reverse. The longevity of electric vehicle batteries is still uncertain. It is also unclear how long automakers will be willing or able to update the software that is increasingly vital to new cars.

However, auto forecasters say that with vehicle prices as high as they are, the industry will have to find a way to keep cars in service.

"Today's vehicles can't have a limitation of five to 10 years," Fiorani said. "It's not practical for a person who's spending $50,000 or $100,000 that it's going to be junk in less than a decade."

Should these trends hold, the auto industry in the U.S. is liable to become ever more competitive. Consumers have their choice of about 450 nameplates in the country already.

"The competition in the U.S. is going to be ferocious," Gottfredson said. "There's too many automakers and too many brands competing for the consumers. The market is going to have to consolidate."

AI Talk Show

Four leading AI models discuss this article

Opening Takes
C
ChatGPT by OpenAI
▲ Bullish

"Profitability for automakers in a slower-growth market hinges on software-enabled monetization per vehicle, not on the pace of new-vehicle volume alone."

While Bain flags a smaller U.S. auto market by 2040, the contrarian view is that a volume hit need not equal a profit hit. Demand could shift from ownership to mobility services, and OEMs that monetize software, subscriptions (OTA updates, battery-as-a-service), and data across fleets can sustain margins even with flatter new-vehicle sales. The real hinge is the profitability of per-vehicle monetization and the pace of autonomous/robotaxi rollout, which could unlock new revenue streams. The article glosses over potential policy tailwinds, used-vehicle dynamics, and the resilience of replacement demand. Execution in software/services, not just unit volumes, will drive outcomes.

Devil's Advocate

Against: even with software monetization, a sustained decline in new-vehicle volumes could compress OEM cash flow; robotaxi progress and regulatory hurdles might keep ownership central longer than expected.

GM
G
Gemini by Google
▬ Neutral

"The shift from volume-based unit sales to high-margin recurring software revenue will likely decouple industry profitability from the total number of vehicles sold by 2040."

The Bain thesis conflates unit volume with industry value. While demographic headwinds and shifting ownership models threaten the '17 million unit' benchmark, they ignore the shift toward high-margin software-defined vehicles and subscription services. Automakers like GM and F are pivoting from pure hardware sales to recurring revenue models. If the industry consolidates as predicted, pricing power may actually improve, offsetting volume declines. The real risk isn't just fewer cars sold; it's the capital intensity of the EV transition colliding with a lower-volume environment, which could compress EBITDA margins for legacy OEMs that fail to scale their software ecosystems quickly enough to replace lost hardware volume.

Devil's Advocate

If autonomous vehicle adoption fails to materialize, the industry is left with high-cost, software-heavy inventory that consumers cannot afford, leading to a catastrophic margin collapse rather than a pivot to services.

Automotive Sector
C
Claude by Anthropic
▼ Bearish

"The market isn't shrinking uniformly—it's bifurcating into affordable-used and premium-new, squeezing mid-market manufacturers hardest."

The article conflates structural headwinds (demographics, affordability) with cyclical ones (high prices, rate environment). Yes, U.S. auto sales peaked at 17.6M in 2015—but we're still at ~16M today despite a decade of supposed decline. The real risk isn't a 2M-unit drop by 2040; it's margin compression NOW. Bain's deregistration thesis is backwards: if vehicles last 12.8 years and prices stay elevated, used-car supply tightens, supporting new-car pricing power. The robotaxi wildcard is real but speculative. What's concrete: younger cohorts can't afford $50K vehicles at 7%+ rates. That's a demand destruction problem for volume players (F, GM) masquerading as a market-size problem.

Devil's Advocate

If robotaxis scale faster than Bain's revised timeline and capture even 5-10% of miles by 2035, fleet utilization collapses vehicle-per-capita demand far beyond the 10-20% household shedding the article mentions. Meanwhile, if immigration policy actually tightens as assumed, population growth could underperform even Bain's already-conservative forecast.

F, GM (legacy volume OEMs)
G
Grok by xAI
▼ Bearish

"Demographic certainty plus slower replacement cycles will force margin-eroding price competition and consolidation among traditional automakers through 2040."

The Bain analysis correctly flags baked-in demographic headwinds: U.S. fertility at 1.6 and halved net migration will shrink the future driving-age cohort, cutting annual sales by over 2 million units by 2040. Longer vehicle life (12.8 years) and lower deregistration (potentially 4.4%) already signal slower replacement cycles. Yet the piece underplays how robotaxi delays and sustained high prices could force automakers to extend profitable ICE/EV platforms rather than lose volume outright. Slower AV adoption also preserves traditional ownership longer than modeled.

Devil's Advocate

Restrictive immigration assumptions could reverse under new administrations, and the 18-34 registration drop may simply reflect delayed purchases that rebound once wages rise or used-EV prices fall.

auto sector
The Debate
C
ChatGPT ▼ Bearish
Responding to Gemini
Disagrees with: Gemini

"Software monetization alone cannot reliably offset lost hardware margins due to scale, churn, and regulatory/cost headwinds."

Gemini argues high-margin software subscriptions will offset volume declines, but that assumption hinges on durable per-vehicle monetization and near-perfect execution. If autonomous rollout stalls, consumer take-up remains weak, and data/cybersecurity costs and channel fragmentation erode margins. OEMs may burn cash on software ecosystems without scale, leaving EBITDA pressure unaddressed. A pure pivot to software is not a given shield for 2040 volume headwinds.

G
Gemini ▼ Bearish
Responding to Claude
Disagrees with: Gemini ChatGPT

"The real risk isn't just lower volume, but a margin-crushing price war against global competitors as domestic demand plateaus."

Claude is right that affordability is the immediate crisis, but everyone is ignoring the 'China factor' as an existential threat to US OEM margins. If domestic demand for $50k+ vehicles hits a demographic wall, legacy players like GM and F cannot simply pivot to software; they will be forced to compete on price against hyper-efficient, vertically integrated Chinese EVs. This isn't just a volume decline—it is a structural margin reset that software subscriptions won't fix.

C
Claude ▼ Bearish
Responding to Gemini
Disagrees with: Gemini

"Affordability collapse precedes Chinese margin pressure; both are structural, neither is offset by software monetization at scale."

Gemini's China threat is real, but it's a *margin* problem, not a *volume* problem for the U.S. market itself. Domestic demand destruction from affordability (Claude's point) happens first and independently of Chinese competition. A $50K vehicle unaffordable at 7% rates stays unaffordable whether the competitor is Ford or BYD. The sequencing matters: affordability kills U.S. unit sales by 2027-2030; Chinese price pressure compounds margin erosion after. Software subscriptions don't solve either.

G
Grok ▬ Neutral
Responding to Gemini
Disagrees with: Gemini

"Tariff walls could delay Chinese price pressure long enough to change the margin timeline Claude described."

Gemini flags Chinese EVs as the margin killer, yet this underweights how sustained 25-100% Section 301 tariffs and potential new import quotas could keep BYD-scale pricing out of the U.S. market until at least 2030. That buffer lets GM and F defend ASPs longer even as affordability erodes units, but it also masks the need to fix domestic replacement cycles before tariffs inevitably ease.

Panel Verdict

Consensus Reached

The panel consensus is bearish, with key risks including affordability-driven demand destruction, slow autonomous vehicle rollout, and potential margin compression from Chinese competition. Despite software subscriptions and mobility services being potential offsets, their success hinges on execution and market conditions.

Opportunity

Successful monetization of software, subscriptions, and data across fleets

Risk

Affordability-driven demand destruction in the near term

This is not financial advice. Always do your own research.