You’re reading Dispatch Markets, a weekly economics newsletter that features contributions from Scott Lincicome, Kyla Scanlon, Karl Smith, Marian Tupy, and Adam Ozimek. If you’d like access to more Dispatch reporting and analysis, consider becoming a member today.
In September of last year, the price of a new vehicle hit a record high, breaking the $50,000 mark for the first time. My gut reaction is that such a figure seems excessive for a typical household. Yet ordinary buyers do purchase cars at that level all the time. Is that a problem? How did we arrive at this point?
There are many forces pushing up the cost of cars. Government policy has certainly played a role. But a substantial share of the climb to the $50,000 vehicle is driven by what buyers actually want.
Does that mean consumers have no grounds to grumble? It turns out that they both are getting what they want and have legitimate reasons to complain. It’s a bit paradoxical. The key takeaway is that more affordable options are still possible.
What’s up with prices?
Let’s begin with the data. There are two principal ways to measure what’s going on. Depending on which gauge you use, new cars either look incredibly expensive or unusually affordable by historical standards.
For now, I am keeping inflation out of the picture so we can examine nominal sticker prices that have so many people buzzing. We’ll get to the role of overall inflation in a later section.
Average price: This figure is simply the yearly average transaction price of a vehicle sold in a given year. I assemble it from two data sources. Beginning in 1990, I rely on numbers from the National Automobile Dealers Association (NADA), a trusted and frequently cited source. For earlier years, I project the series backward using BEA’s (Bureau of Economic Analysis) growth rates for average vehicle prices.
Average prices clearly show a sharp rise over time, approaching the $50,000 level. If we look back three decades, new-vehicle prices were a little over $20,000, implying a roughly 122 percent increase over that span. The surge after the pandemic has been especially strong, with prices climbing by about 25 percent since 2019.
Quality-adjusted prices: The CPI (consumer price index) produces monthly estimates of new-vehicle prices that are adjusted for quality. This approach measures how much prices have increased over time while holding the mix of models constant and attempting to account for changes in quality (more on that shortly).
The CPI for new vehicles appears far less dramatic than the unadjusted average prices. This measure stayed remarkably flat for more than two decades, signaling little to no growth in quality-adjusted prices. Since the pandemic, however, vehicle prices have risen, and they’re up about 24.2 percent over the last 30 years.
The gulf between these two measures is striking. Did the price of vehicles rise by a quarter or more than double? Remember that these are nominal figures, not adjusted for all inflation. If we convert them into constant dollars, the erosion in affordability is much less stark, yet still meaningful. Real average prices have risen about 20 percent over the last 30 years, while the real CPI for autos has fallen by about 32 percent. In other words, the quality-adjusted price of vehicles has, in real terms, actually edged downward.
So are cars more expensive than ever or more affordable than ever? What’s driving the discrepancy between the two measures?
The slow death of the car
A key factor pushing up average prices is that buyers are shifting away from the cheaper end of the market. This appears in the long, steady decline in the share of new-vehicle sales that are passenger cars. Buyers have gradually redirected their purchases toward what BEA labels “light trucks,” a broad category encompassing trucks, vans, SUVs, and crossovers. Three decades ago, cars made up half of all new-vehicle sales; today, they account for only about 16 percent. This is a slow, persistent trend that stretches back at least to the 1970s.
This move away from cars has significant implications for average prices because cars are typically cheaper than light trucks. On average, buyers pay about 50 percent more for a new light-truck model than for a new car. Crossovers bear some resemblance to cars, but BEA categorizes them as light trucks, and the price gap is sizable. While sticker prices have surged toward the $50,000 mark, BEA’s consumer-price estimates for new cars hover just under $32,000.
One way to visualize the impact of fewer cars on average prices is to imagine a household today buying the same mix of vehicles as in 1996, but facing today’s price levels. If that were the case, the overall average price would be about 16.8 percent lower. In other words, a world where cars remained a larger share of sales would feel notably cheaper.
Although the decline of cars clearly elevates average prices, it has no direct effect on the CPI for vehicles. The CPI tracks price changes for the same model over time. If the price of a Toyota Corolla sedan increases, that shows up as a higher CPI for autos. If buyers switch from Corollas to more expensive Tundra trucks, that does not show up in the index as a price increase for a given model. It does, however, influence how much autos weigh in the overall CPI over time.
Quality matters
Another reason for higher vehicle prices is simply that automobiles have grown nicer. Or at least they include more features that buyers value. The CPI attempts to remove the effects of rising quality. The BLS collaborates with automakers to quantify how much extra content they add each year, and statisticians subtract the retail value of that content from the price change to arrive at a “quality-adjusted” price.
For example, in its 2026 model-year light trucks report, the BLS found manufacturing costs rose by $187.93 relative to the prior year due to quality changes. The retail value of those changes was $201.93, broken down as follows:
– $12.98 for safety equipment improvements.
– $33.03 for infotainment system improvements.
– $155.92 for other changes, such as propulsion improvements and shifts in standard versus optional equipment.
Over the past 30 years, the BLS has recorded total nominal quality improvements of $6,048 for light trucks. That amount represents about 28 percent of the $21,883 increase in nominal retail prices noted in the same BLS reports. For cars, the share is about 29 percent. In other words, roughly a quarter to a third of the rise in nominal prices over three decades reflects measured improvements in quality.
If those figures feel substantial, consider that the average weight of vehicles in a given model rose by about 21 percent from 1996 to 2018. The graph (not shown here) illustrates this with examples like the Toyota Corolla and the Ford Explorer, both up by roughly 20 percent in weight. Similarly, average horsepower within models rose by about 54 percent, and fuel efficiency improved by around 6 percent.
A further sign of better quality over time is that vehicles tend to stay on the road longer. In 2025, the average age of vehicles on the road was about 12.8 years, up from 8.5 years in 1996.
Of course, not every change that raises costs is a consumer benefit. Some safety standards are mandated and included as quality improvements. The costs associated with meeting safety rules can be substantial. McKinsey estimated that between 2001 and 2010, government requirements added roughly $400 in new components to a typical midsize car purely for safety. If buyers value these changes less than they cost, the CPI for autos would be biased downward.
Conversely, improvements that do not add cost will not be removed from the price as a quality adjustment. For example, better engine performance due to innovation would not be accounted for as a quality change unless it came with a higher price tag.
Net, the quality adjustments used by the BLS likely understate the true value of improvements. If that is the case, a genuinely quality-adjusted CPI for autos would indicate even larger real declines in price after accounting for quality.
Is there anything to complain about?
The analysis thus far might seem to invert affordability: a 20 percent real rise in average prices over 30 years, paired with a 32 percent drop in real quality-adjusted prices, doesn’t scream a disaster. Still, there are two legitimate grievances: progress has slowed, and the most affordable vehicles are gradually vanishing.
First, note from the second chart that the inflation-adjusted CPI for vehicles fell more quickly up to around 2010. From 2010 to 2019 the pace of decline slowed, then prices surged during the pandemic and have since been rising more slowly again. In short, improvements have occurred more slowly than they used to — a valid reason to complain, even if the headline $50,000 average exaggerates the issue.
Second, the most affordable models are indeed disappearing, a factor not captured by the CPI for vehicles. The CPI tracks changes in price for specific models over time, so when a model is discontinued, like the Fiesta in 2023, the index does not reflect a higher price for that particular model because its price didn’t increase. It’s not a measurement flaw when a model disappears, but when the market’s cheapest options vanish in large numbers, affordability declines in a real sense that the CPI doesn’t capture.
And the cheapest options are indeed dwindling. Clifford Winston, writing in the New York Times, notes that in 2012 there were about a dozen U.S. offerings priced below $25,000 in today’s dollars. By his count, there are only four left now.
To be clear, affordability isn’t extinct. It can still be achieved by opting for base models and avoiding pricey add-ons. Toyota maintains a dedicated webpage showing Corolla configurations with base prices under $25,000. But once you start adding options, a Corolla can top $30,000. Another example is the 2025 Mazda CX-50, which ranges from $28,950 to $43,300 depending on trim and options.
Nevertheless, affordable choices are clearly shrinking over time, and that gives consumers a legitimate cause for complaint.
Policy matters
A second credible grievance is that government policy is pushing prices higher. The most significant factor here is likely trade policy.
Tariffs on trade have risen markedly during the latter part of the Trump administration. There is a global 25 percent tariff on imported vehicles and parts, with somewhat lower rates for a handful of countries with trade deals. In addition, there are 50 percent tariffs on imported steel and aluminum. Automakers estimate that they have paid roughly $35.4 billion in tariffs since 2025.
But the trend did not begin in 2025. A longstanding 25 percent tariff on light trucks has been in place, and the United States–Mexico–Canada Agreement, effective in 2020, included various protectionist provisions. One notable rule is the Labor Value Content (LVC) requirement, which stipulates that 40–45 percent of a vehicle’s value must be produced by workers earning at least $16 an hour, along with a mandate that 70 percent of steel and aluminum come from North America. These rules raise costs directly and add administrative burdens to prove compliance, complicating an already intricate North American auto-supply chain that crosses borders multiple times before final assembly.
Another avenue through which policy pushes up prices is a dense system of state-level dealer franchise laws. These laws prevent manufacturers from selling cars directly to consumers and regulate market entry and exit, among other things. The effect is to “almost guarantee dealership profitability and survival.” Some estimates put the added cost at around $4,000 to $5,000 per new car.
Automakers falling behind
Policy shifts could potentially help with one part of the problem, but the fix is complex: automakers seem to be lagging the productivity frontier by not embracing newer manufacturing methods. On the frontier, you have Tesla, and perhaps more prominently, a wave of Chinese automakers; behind the frontier is basically everyone else, and the gap between leaders and stragglers has widened substantially.
There isn’t a single reason for the lag. The Center for Automotive Research (CAR) recently summarized a series of industry expert roundtables convened to discuss the challenges of delivering affordable electric vehicles. The experts agreed that next-generation manufacturing techniques could substantially reduce car costs. This includes modular design, unboxed manufacturing (a more modular approach), and gigacasting. One estimate suggests unboxed manufacturing could cut production costs by 25 to 30 percent.
The difficulty in adopting these new manufacturing techniques is likely holding many companies back from reaching the frontier. But it is one of many obstacles, including vertical integration, standardization of products, and overall product engineering improvements.
Automakers are trying to learn from Tesla and Chinese producers. Yet, by most accounts, no one believes they have reached that level yet. Could policy help? Outlining a concrete plan would go beyond the scope of this piece, but it is a serious issue worth thoughtful consideration.
Signs of stress
Meanwhile, it’s hard to fault consumers for sensing they aren’t getting a great bargain. There are clear signs of strain: repossessions are increasing, a larger share of car payments exceed $1,000 per month, and more buyers are financing for longer than five years.
Even if part of the problem stems from buyers choosing bigger, higher-quality vehicles, they are doing so in a market with fewer affordable options. Across the quality spectrum, it is likely that policy and automakers’ slow progress toward the technological frontier are pushing costs higher than they ought to be.
I’m not sure what to do about consumer preferences for spending more and more, but there is room to improve vehicle affordability even without fully reversing that trend.
Markets FTW
USA Today has updated its ranking of the 10 best buffets in the United States. And to the surprise of no buffet enthusiasts, Lancaster, Pennsylvania — my hometown — places two of the top three spots.
If you have never visited No. 1, Shady Maple, you’re in for a real treat. It features 200 feet of Pennsylvania Dutch cuisine and accommodates about 2,000 guests. This isn’t a paid plug for Shady Maple. Shady Maple serves as a showcase for the power of capitalism.
Chart of the Week
The outlook remains highly uncertain out there.
Worth Your Time
– Stereotypes about fields of study are guiding people toward the wrong degree
– Debt and deficits: Time to worry?
– Automakers are grappling with whipsaw regulatory costs by releasing fewer new models
– The “cow number paradox” in cow productivity and milk’s price
– Immigrants and U.S. billion-dollar firms
Disclaimer: The opinions expressed above do not necessarily reflect those of the presenting sponsor.
