The Average Three-Year-Old Used Car Now Costs More Than $30,000

Used cars are getting more expensive. It’s not great. While it’s always possible some magical production boom in the United States makes everything grander and cheaper, this flies both in the face of general macroeconomic principles and basic logic. With the exception of deep economic recessions, the biggest driver of the car market is always supply and rarely demand. That’s it. “How many cars are there?” is the key question.

The car market has been through a series of both exogenous and entirely self-inflicted shortages. When the Global Financial Crisis hit in 2008, the Obama Administration’s short-term fix was to encourage people to junk their cars, thus stoking demand and removing supply. Though highly controversial, it worked, and the industry self-regulated a bit. And then the pandemic happened, followed by inflation, followed by whatever it is we’re doing now with trade.

We haven’t caught up, and maybe we never will. Don’t let this get you down. I’m going to suggest that you can arbitrage your car purchase right now and potentially make a good amount of your money back on a trade-in. It’s more of a thought exercise than anything, but maybe one of you will take me up on it. [Ed Note: It’s best not to take buying/selling/financial advice from us. -DT].

Speaking of thought exercises, do you think you could work a 57-hour workday? That’s apparently what at least one Lemon Law lawyer billed Ford. The automaker is a little less pleased with that and is suing a bunch of Lemon Law firms in California.

And, finally, what the hell happened to Northvolt? Everything. There’s no such thing as “too big to fail.”

The Most Valuable Thing Many People Own Is Maybe A Nice Three-Year-Old Car

I love charts and graphs. Maybe one day I’ll be successful enough to be able to expense a Bloomberg Terminal. Until then, I have to go look them up on the Federal Reserve website like a schlub. Here’s a great graph. This is the auto inventory-to-sales ratio in the United States since the early 1990s. The higher the number, the more cars in inventory relative to the number of sales.

You can see the Global Financial Crisis in 2008, which caused a drop in sales and an increase in inventory. Then, in July 2009, Cash-For-Clunkers went into effect, and you can see it caused a deep dip in the availability of cars. The ratio stayed within the normal range until the pandemic, when people didn’t buy cars because they didn’t go outside. This quickly went the other direction when automakers cancelled orders for semiconductors, assuming a prolonged period of low demand.

If you bought a car three years ago, chances are you had to pay a lot of money for it. We were still in the trimflation/supply-constrained universe, and so it’s unlikely you got a good deal. As you can see above, we were just starting to recover a bit last year, but higher interest rates and economic uncertainty kept things below that 2-to-1 ratio line.

Tariffs could potentially blow this chart up again, as, according to some estimates, more than a million fewer cars could be sold in the United States.

This is especially troublesome for used cars. The used car market needs people to trade in their cars, but that again requires there to have been a lot of cars sold at some point in the past. In particular, the used car market really likes a bunch of lease trade-ins because those are reasonably well-cared-for cars that aren’t too old, but there are way fewer of those, as we’ve discussed in the past. Most analysts predicted increased used car prices this year, and that was before tariffs.

According to Edmunds, those predictions were correct:

The average transaction price for a 3-year-old used vehicle was $30,522 in Q1 2025, up 2.3% from $29,844 a year ago — and marking the first time this figure has surpassed $30,000 since Q2 2023, according to Edmunds data. This increase in price was accompanied by longer lot times at used-car dealerships: The average days to turn — the number of days a vehicle sits on a dealer lot before a sale — for used vehicles was 38 days, four more days than a year prior and the highest Edmunds has on record since Q1 2021 with the same 38 days to turn.

The longer lot times paired with higher average transaction prices for 3-year-old used cars show that dealers are likely maintaining higher asking prices and demonstrating a willingness to wait for the right buyers given the lower supply. The effect of the limited number of lightly used vehicles to purchase is reflected in the average age of used vehicles listed on Edmunds rising from 5.7 years old in March 2024 to 6.1 years old in March 2025.

So that happened. Again, this isn’t ideal for consumers. If you’ll indulge me, I think there’s a way to foresee the tariff impacts and make some money if you’re the kind of person who likes a new car every three years.

How To Arbitrage Your Car

The idea of trying to arbitrage a depreciating asset is, of course, silly. No one knows for sure what the car market will be like in three years, and it’s rare that a used car is ever worth more than it was sold for new (although that did happen during the pandemic). What I’m going to argue is that we’re in a narrow window where you can work this tariff situation to your advantage.

This chart is from the same Edmunds article, and it shows the predicted transaction price for a three-year-old car from cars built in different countries versus what the average transaction price actually is. While I don’t think there’s too much to draw from this chart at the moment, given uncertainty around tariffs, I do think the concept is instructive.

As the authors of the analysis point out, “For shoppers with trade-ins, country of assembly — usually a nonfactor — now plays an unexpected and potentially significant factor in a vehicle’s resale value.” Where your car is made is important.

President Trump has talked about walking away from USMCA and, at least as of this morning, is threatening/promising to raise tariffs to 50% on the EU. Will that last? I don’t know. Our best guide is the UK-US trade deal, which saw the US folding quicker than a Trump-branded university. Perhaps there’s hope.

The way this works, though, is that automakers might have to raise prices on new cars built outside of the United States or that use a lot of non-USMCA-compliant parts. My reading of the tea leaves shows that Korean automakers are probably going to do well in trade talks, as will Japanese car companies. Germans and other Europeans? I think it could be tougher.

The best possible deal you could make is on a German non-EV right now, I think. Yes, EV prices are super low, but depreciation is going to be super high, and I’d be worried about cars like the Audi e-Tron GT being discontinued. What you want to buy is a European car that will still be in demand in three years, but can’t be built in the United States anytime soon.

A vehicle that might fit the bill is the Audi Q5. There’s no future where Audi isn’t building the Q5. It’s too popular. At the same time, the Q5 is not USMCA compliant in spite of being built in Mexico. People will want Q5s in three years, but tariffs are going to make that car more expensive. Similarly, the Porsche Cayenne or just about any European crossover not built in the US could work.

Assuming tariffs cause car prices to go up and constrain supply, and discounting the possibility of a global recession, you could have a car that’s highly in demand that you bought before prices went up, and is suddenly worth a lot proportionally because the new version of it is very expensive. This requires you to buy a car now and then sell it or trade it in for something that doesn’t have the same tariff impact, of course.

If you really want to be safe, lease a car for three years and negotiate a lease buy-out price ahead of time. If car values drop, you can just roll into another lease with your negotiated buy-out price. If car values rise significantly, sell the car somewhere else.

The Magical Mystery Tour Of Fictitious Billings

You don’t have to be smart to be a lawyer, but it certainly helps. Some of the best writers I’ve ever met are lawyers. However, there’s being smart and being too clever. Ford is alleging that attorneys bringing Lemon Law claims in California charged “phantom legal fees” when representing clients, which were then passed on to car companies.

Per Reuters, this scheme added up to $100 million of extra fees:

Describing the invoices it received from California lawyers as a “magical mystery tour of fictitious billings,” Ford claimed that attorneys named in the lawsuit took advantage of a statute designed to protect consumers from faulty products, including cars.

Under the Song-Beverly Consumer Warranty Act, commonly known as California’s Lemon Law, automakers are required to pay for legal work, court fees and related expenses associated with defective vehicles.

That requirement, Ford alleged, has created an opportunity for lawyers to pad their bottom line by claiming more hours than they actually worked or reporting having been in more than one place at the same time.

The complaint alleged that Steve B. Mikhov was the “ringleader of the criminal enterprise” and that he and Knight Law Group, the Los Angeles-based firm where he was a founding partner, “orchestrated” the scheme.

That’s funny. At one point, Ford claims a lawyer named Amy Morse billed an “ostensibly heroic but physically impossible 57.5-hour workday in November 2016.” Give the person who wrote this a raise, or lemme give them a job. It’s good writing!

This gets into the give-and-take of consumer protection laws. There needs to be enough incentive for attorneys to take these cases and for companies to feel chastened if they do something bad, but a poorly written law can create ridiculous outcomes like the one alleged in this lawsuit.

Northvolt Was Definitely Not ‘Too Big Too Fail’

Swedish battery maker Northvolt was supposed to be Europe’s answer to China and South Korea. Instead, it went into bankruptcy after most of its partners freaked out and walked away from the company. How did this happen? Germany’s Manager Magazin has a long piece that gets into the timeline of the company, blaming a whole host of factors:

Anyone who reconstructs the initial cracks, the major fractures, and the eventual collapse in the Northvolt case ends up in a chronicle of failure. In the end, a cocktail of greed and arrogance, mismanagement, and a lack of risk-taking by the automakers caused a company that almost everyone considered “too big to fail” to fail. A lesson in how not to do things.

What you get from the piece is a lot of the overconfidence you’d expect from ex-Tesla execs at the height of the EV bubble, including the idea that the company could quickly catch up to CATL by trying to control every part of the process while also ignoring critical but time-consuming processes, resulting in the company constantly being unable to produce cells that would pass automaker tests.

That was bad, but automakers like Volkswagen also play a role here. Rather than commit to Northvolt long enough to see it succeed, they decided early on that it wasn’t worth the risk. VW, in particular, invested early and then decided Northvolt was a competitor to its other battery investment, PowerCo. The break with Volkswagen meant that Northvolt lost some of the purchasing power and industrial know-how it needed.

Northvolt also assumed there would always be a customer and didn’t, according to this report, require automakers to make the kind of serious, binding commitments it needed to stay in business if the market turned. The market slowed and automakers walked away.

The first warning signs appeared when VW negotiators brought in lawyers who took meticulous minutes. “They made sure to avoid any potentially legally binding commitment,” one participant recalls. For Volkswagen, it wasn’t just about the invested capital—that had already been slightly discounted in the annual report anyway. Rather, the automaker needed the cells. No one had placed such large orders with Northvolt; without the Northvolt cells, new electric models were threatened with postponement. At Porsche alone, the issue was unsold cars worth 4 billion euros. A conservative estimate.

Yikes.

What I’m Listening To While Writing TMD

Sometimes the skinny weirdos win. In this case, it’s Pulp and Jarvis Cocker filling in for The Stone Roses at Glastonbury and 1995 and kind of destroying with “Common People.” Enjoy.

The Big Question

What’s the best car arbitrage deal?