Starting in first gear

Good morning. This is the last time we talk about Monterey Car Week. Not because it wasn't a spectacle — Friday night settled that — but because a week of watching people bid the price of an office building on cars they'll never drive makes the whole hobby feel a little unhinged, and even the people who love it are tired of hearing about it. So we settle the two promises we made last week, keep it short, and move on to cars people actually turn a key in.

The rest of the issue is what you see when you stop looking at the lawn: an American brand quietly leaving the world's biggest car market, the man who designed the McLaren F1 building it again with 34 more years of knowledge, and a serious attempt at the question that came up in every tent this week — is Wall Street money the reason a Ferrari Enzo now costs what a 250 GTO cost a decade ago?

In today's newsletter, we'll get into:

The Auction Desk: the Cuda result, the $500M answer, and the numbers worth remembering

One Big Story: Chevrolet just left China after selling one car in June

Engineer's Corner: the McLaren F1 and the Gordon Murray S1, side by side

Garage Economics: the private-equity theory of hypercar prices — how much of it holds up?

Let’s roll.

THE AUCTION DESK

The hammer fell. Here's the ledger.

Source: RM Sotheby’s

We promised you the Cuda first. Mecum Lot F135, the 1971 Hemi Cuda four-speed hardtop (one of 59, $850–900K estimate), crossed the block Friday with the highest bid at $550k being short of meeting the reserve — which, next to the $3 million-plus Hemi Cuda convertible Mecum sold in Kissimmee in January, is a disappointment.

Now the $500M question. Hagerty's forecast midpoint was $470 million, the optimistic case $496 million, and the record $471 million from 2022. Through Thursday the week was running behind 2025 — $96.9 million versus $116.8 million at the same point, on fewer cars sold. Then Gooding Christie's sold the 1964 Shelby Cobra Daytona Coupe, CSX2300, the only one Carroll Shelby ever owned, for $42,905,000: nearly double the $22 million Duesenberg SSJ record for an American car, and the highest public auction result of 2026, ahead of January's $38.5 million 250 GTO. Final Monterey Car Week total is still being confirmed but it’s safe to say it likely surpassed $500M. TBD.

The rest, quotable at cars-and-coffee: at RM Sotheby's, a 227-mile 2023 Daytona SP3 made $17,825,000 against a $10–12 million estimate; a 9,834-mile Enzo made $9,410,000, which was only the eighth-highest Enzo result ever, a sentence that would have been insane to type in December; Clark Gable's Duesenberg JN made $9,080,000; a Bugatti Mistral $8,310,000; a Pagani Huayra Roadster $8,035,000; and Louis Delage's own D8-120 S Aerodynamic Coupe set a marque record at $5,780,000. Broad Arrow's black Enzo, 13,200 miles, made $10,675,000. Bonhams' Laguna Seca sale, covered Friday, finished around $12 million at 79% sold.

📈 Market signal: Look at what the eight-figure cars have in common. Of the six biggest results above, four were built after 2002. The generation that had these cars as posters now has the checkbook, and the prewar coachbuilt stuff, gorgeous as it is, is no longer where the marginal dollar goes. Hold that thought until Garage Economics.

ONE BIG STORY

Chevrolet just left the world's biggest car market

While the collector world was on the Peninsula, General Motors did something it spent two years denying it would do: it ended Chevrolet retail sales in China.

The collapse is hard to overstate. Chevrolet sold 767,001 cars in China in 2014 through the SAIC-GM joint venture, one of the top brands in the country. In 2025 it sold 5,314. In the first half of 2026, 36. In June, one Equinox. That's a 99% decline in a decade, in a market that grew the whole time; the last new Chevrolet launched there was an imported Tahoe in 2024, and dealers had already vanished from Beijing and Chongqing.

The mechanics matter more than the headline. GM is not leaving China. It just renewed the SAIC-GM venture for another 20 years, through 2047, with at least 30 new-energy vehicles planned by 2030 under Buick and Cadillac, the two brands Chinese buyers still want. Chevrolets will keep rolling out of the same Chinese factories, but for export to the Middle East, Africa, South America, Mexico and Asia-Pacific. GM's China operations are profitable again, roughly $248 million in equity income in the first half, precisely because they stopped chasing volume they were never going to win back from BYD, Geely, Chery and Changan.

Read it alongside two headlines from the same week: Ford is moving some Lincoln production out of China and back to the U.S., and Tesla is cutting output at Shanghai. This is the American industry conceding, brand by brand, that it can't sell mainstream cars to Chinese consumers on price, cadence or software, and repositioning China from "growth market" to "export base."

The takeaway: Two weeks ago we called the border the news; this week the news is the exit. A decade ago every strategy deck in Detroit had a slide about China as the second home market. It's now a factory that happens to be in China. If you've wondered why the domestic lineup keeps narrowing to trucks, SUVs and halo cars, this is part of the answer: the volume that used to subsidize variety isn't coming from Shanghai anymore.

ENGINEER’S CORNER

Gordon Murray builds the McLaren F1 again, on purpose this time

Source: Top Gear

Two things happened at The Quail on Friday that belong here rather than in the auction tent. Gordon Murray Special Vehicles unveiled the S1, a road-going homage to the McLaren F1, and — for a company that said it would build one-offs and handfuls — announced a run of 64 cars, all sold. To see why the car matters and why that number is a wink, go back to the original.

The F1, 1992. Murray's brief was almost monastic: three seats with the driver in the middle, a naturally aspirated V12, a manual gearbox, and a target weight of 1,000 kg. No turbos, no power steering, no ABS, no traction control, no brake servo. He asked BMW M for at least 550 hp from a compact V12; Paul Rosche's team delivered the 6.1-liter S70/2 with 627 hp and 480 lb-ft, more power than requested and 16 kg heavier than requested, an overrun Murray still grumbles about. The tub was the first carbon-fiber monocoque in a production road car. Because the engine ran hot and sat close to it, the bay was lined with 24-karat gold foil, roughly 16 grams, because gold reflects heat better than anything cheaper. Two electric fans in the diffuser pulled the boundary layer off the underbody to sharpen the ground effect and cool the brakes. It weighed 1,138 kg, missed its target, and was still the lightest thing in its class by a mile. In 1998, with the rev limiter raised, it ran 240.1 mph at Ehra-Lessien. McLaren built 106 F1s in all; 64 were the standard road car.

The S1, 2026. The S1 sits on a version of the chassis from last year's S1 LM (five built, the first reportedly selling for over $20 million). Where the LM wore the wing, splitter and skirts, the S1 strips them off and replaces them with active aero integrated into the rear diffuser — a direct descendant of the F1's fan-and-diffuser thinking, now with actuators. Every carbon panel except the roof and engine cover is new; the round lights in the front intakes nod to Murray's 1992 F1 Monaco show car. The engine is the 4.2-liter Cosworth V12: 681 hp, 367 lb-ft, a 12,100-rpm redline, but retuned so peak torque arrives from 2,500 rpm, because this one is meant to be driven from Los Angeles to Monterey without complaint. Six-speed manual with a longer sixth for touring. Weight target: under 997 kg. Polished Inconel exhaust and, yes, 24-karat gold heat shielding. Ride height up 10 mm over the LM, more compliant dampers, switchable power steering, and dihedral doors that finally have proper drop-down glass.

What changed in 34 years. Almost nothing in the philosophy, almost everything in the execution. Same three seats, same central driver, same three pedals, same twelve cylinders, same allergy to weight. The differences are where the technology finally caught up with the idea: fixed aero becomes active; a BMW V12 that revved to 7,500 becomes a Cosworth that revs to 12,100 and still pulls at 2,500; 1,138 kg with a 266 kg engine becomes a target under 997 kg with more horsepower. The F1's gold foil was an engineering solution. The S1's is an engineering solution and a love letter.

The takeaway: Murray missed 1,000 kg in 1993 by 138 kg and spent three decades bothered by it. The S1 is what happens when the same engineer, with modern carbon, a modern engine and no F1 team's accountants over his shoulder, gets to hit the number. And 64 is not a coincidence — it's the F1 road-car count. He built the whole run again.

GARAGE ECONOMICS

Is private equity buying the Ferraris?

Source: Getty Images

Here's the theory from every tent this week: hypercar prices went parabolic because private equity — "Wall Street," "the funds" — started treating limited-run Ferraris like apartment buildings. It explains a bizarre year and supplies a villain. So let's take it seriously and see how much survives.

What needs explaining. In January, Mecum sold the Bachman collection at Kissimmee: 19 model records fell in one sale, including all five Ferrari halo cars (288 GTO, F40, F50, Enzo, LaFerrari), topped by a Giallo Modena Enzo at $17,875,000. The previous Enzo record, the Pope's car in 2015, was $6.05 million and had stood eleven years; it was then beaten four times in a month. Paris followed with a €8.1 million Enzo plus world records for a 288 GTO and an FXX K Evo; Amelia in March saw a Carrera GT at $6.7 million, 205% above the record on New Year's Day. Friday's SP3 landed ten million above the previous non-charity record. Trade people assumed January was one bidder and wouldn't travel. It traveled.

The literal version doesn't hold. Real private equity buys operating companies with borrowed money and needs cash flow and an exit. An Enzo produces no income, costs 10–12% in buyer's premium going in and a commission going out, and can only be valued by selling it. Collector-car funds have existed for a decade and most of the early ones quietly failed for exactly these reasons. Blackstone with a paddle? We can find no evidence of it.

But the spirit of the theory holds up better than the letter. Something is buying these cars with the discipline of a portfolio rather than the passion of a collector, and it looks like this. Italian asset manager Azimut launched an "evergreen" collector-car fund in 2023 that only buys cars worth over €1 million, and U.S. accredited-investor funds are publishing 2026 investment theses whose target lists — modern halo Ferraris, the McLaren F1, air-cooled 911s, Gullwings, pre-1970 racing Ferraris — read like the auction catalogs they just bought from. Sotheby's Financial Services now openly advertises lending against top-tier Enzos so buyers can "preserve liquidity"; the 1989–91 Ferrari bubble ran on borrowed money too. And the buying is concentrated in a way passion isn't: of 59 post-1980 cars sold above $2 million in the first quarter, Hagerty found 84% went to U.S. buyers, while 83% of values in its UK price guide sat flat and its overall Market Rating hovers near a 15-year low even as 2025 set a record for auction dollars. That is not "the car market is hot." That is a small number of buyers with very large budgets moving in the same direction on the same sixty cars — with the seller helping: Ferrari built 600 Daytona SP3s, sold most to prior Monza buyers at around $2.3 million, and two have now traded publicly for $17.8 million and $26 million.

What the theory gets wrong. It names the wrong villain and skips the simpler explanation. Hagerty overlaid the year's top auction price on Fed data for top-1% incomes: the two tracked closely until 2011, and since then the record has beaten that income measure nine times. The relevant buyer isn't the top 1% anymore, it's the top 0.01%, for whom $18 million is not a stretch. Add Friday's generational rotation — millennial and Gen Z money pricing the posters it grew up with — and you get most of this year's chart without a single fund manager. Wealth concentration plus nostalgia is a boring answer. Boring answers usually win.

The takeaway: "Private equity is buying the Ferraris" is directionally right and literally wrong. It's not buyout shops; it's the financialization of the top of the market — dedicated funds, family offices and lenders, layered on a very small, very rich buyer pool that loves the same forty models. Financial money leaves the way it arrives: all at once, in the same direction. Three-digit moves in a single quarter have happened once before in this hobby, in 1990, and the correction that followed took prices to a quarter of peak. Nobody serious is predicting that. Everybody serious is watching for it. And if you own anything other than the sixty cars this money wants, this year's headlines have very little to do with what your car is worth — which is the part nobody in the tents wanted to hear.

🏁 THE COOL-DOWN LAP

The best thing anyone said about the market this month came from a collector who wasn't selling. Hagerty's John Mayhead, writing in Square Mile, tells of a British F50 owner whose car is valued at around £4 million and who was offered £10 million for it by another collector. He turned it down. His entire explanation, as relayed: he just likes driving it.

Which is, when you strip away every chart above, the whole point.

Share the Downshift. Forward them this email. Every subscriber from your link gets you closer to the Daily Downshift starter-pack.

That's a wrap. See you Wednesday for my person BaT experience last month. Same inbox, lower gear.

Was this email forwarded to you? Subscribe free at Daily Downshift. Drop a gear. Read the good stuff.

Consider this the feedback loop: reply with what worked, what didn't, and what you want more of. Reader replies are the single biggest input into what gets written next.

Marsel

What did you think of today's email?

Your feedback helps me create better emails for you!

Login or Subscribe to participate

Reply

Avatar

or to participate