America Built a Car Wash Every Nine Hours. Then the Math Changed.

HARLINGEN, Tex. The number that ended the debate was 11 million.

That was how many gallons of water the Blue Wave car wash on the city's commercial strip had pulled from the municipal system in a single year, according to a report the Harlingen WaterWorks System handed to the city commission in November 2024. The Rio Grande Valley was in a drought. The city had 21 car washes already, more than one for every 3,500 residents (Innowave calculation), and developers were asking for more.

Mayor Norma Sepulveda read the figure aloud. On November 20, after eight months of arguing about it, the commission voted 5 to 1 to require that any new car wash sit at least two miles from an existing one and recycle most of its own water. Commissioner Rene Perez had first proposed limits the previous March and lost 4 to 1. What changed in between was not the politics. It was the arithmetic.

Harlingen is a small city, and its ordinance is a small thing. But it is one of at least fourteen such measures passed or proposed across eight states since 2023, from Birmingham to suburban Detroit to the high desert east of Los Angeles. Together they mark the moment the most prolific retail building boom of the 2020s ran into the people who have to live next to it. The boom has not ended. It has stopped making sense.

One Tunnel Every Nine Hours

In 2022, by the industry's own count, developers opened 943 express car washes in the United States. That is one new tunnel roughly every nine hours (Innowave calculation), around the clock, for a year. The figure comes from CarwashOS development tracking and Amplify Car Wash Advisors, compiled in MMCG Invest's 2026 industry review; there is no government series for this, which is part of the problem. Openings slid to about 550 in 2025 and are tracking at 450 or fewer this year. The peak is behind us. The buildings are not.

To understand how a business as humble as a car wash became a Wall Street asset class, you have to understand three inventions that arrived together around 2015. The first was the express exterior tunnel itself: no attendants inside the car, no vacuuming by staff, a conveyor that can move a vehicle through in three minutes and a hundred and fifty in an hour. The second was the unlimited monthly membership, which turned a $12 impulse purchase into a $25 or $30 recurring charge, billed to a card whether or not the car ever got dirty. The third was the sale-leaseback, in which a developer builds the tunnel for $4 million to $8 million, sells the dirt and the building to a net-lease investor, signs a lease of roughly twenty years, and pockets the difference between construction cost and what a buyer will pay for a long stream of rent.

Put those three together and a car wash stops looking like a car wash. It looks like a software company with a parking lot: recurring revenue, high margins at scale, and a way to recycle capital without waiting for the business to pay for itself. Private equity noticed. Leonard Green bought Mister Car Wash in 2014. Golden Gate took Tidal Wave in 2020, Access Holdings built Spotless Brands the same year, Warburg Pincus backed El Car Wash in 2022, KKR put a reported $850 million into Quick Quack in 2024. Raymond James's spring 2026 tally counts more than a dozen sponsor-backed platforms with over a hundred sites each.

The demand story was real, which is what made the supply story so easy to tell. The International Carwash Association's consumer surveys show the share of American drivers who pay someone else to wash their car rising from about half in 1996 to 79 percent today. The Census Bureau's County Business Patterns counted 19,807 car wash establishments in 2023, a figure that includes detailers and mobile outfits but excludes the in-bay machines bolted to gas stations. AMP's benchmarking panel covers more than 7,400 conveyor tunnels, which works out to roughly 2.2 tunnels for every 100,000 Americans (Innowave calculation).

But 79 percent is a ceiling, not a runway. When four out of five drivers already use a professional wash, a new tunnel does not create a customer. It takes one from the tunnel down the street. The industry built as if the conversion curve would keep bending upward. It had already flattened.

The Million-Dollar Tunnel

If you want a single number that captures the gap between what the boom cost and what it is worth, it is this: about $1 million a site†.

That is what Driven Brands, the Charlotte-based owner of Take 5 Oil Change and Meineke, got for its American car wash business when it agreed in February 2025 to sell roughly 380 locations to Whistle Express. The headline price was $385 million, but only $255 million of it was cash; the other $130 million was a note Driven itself agreed to carry. Many of those sites were leased rather than owned, so the figure reflects the operating business, not the land. Still, these were tunnels that had cost somewhere between $4 million and $8 million apiece to build, and the market would pay a fraction of that for them as going concerns.

Driven had seen it coming. The company wrote off $125 million of car wash trade names in 2022, then took an $851 million goodwill impairment on the unit in the third quarter of 2023, wiping out essentially the whole premium it had paid to assemble the chain. Its chief executive at the time, Jonathan Fitzpatrick, blamed "continued weak consumer demand and increasing competition in the U.S. Carwash sector" and froze new investment. Two years later the company was out entirely, having also sold its international wash arm for about €406 million.

The second signal came from Plano, Texas. Zips Car Wash, backed by the private equity firm Atlantic Street Capital, had grown to about 260 locations in 23 states by borrowing. When it filed for Chapter 11 on February 5, 2025, it owed $653.9 million in funded debt, about $2.5 million for every tunnel†, and had $1 million in cash. Its annual interest bill had climbed from $59 million to $93 million as rates rose. It had 625,000 members, which was not the problem. In a court declaration, its restructuring officer, Kevin Nystrom, described an industry where "upwards of 900 new car wash sites have opened annually," and the company moved to reject 41 leases on locations it could not make work. Zips emerged in April owned by its lenders, among them HPS and PennantPark, and now operates "more than 185" sites, roughly 75 fewer than it entered with. Atlantic Street's equity was wiped out.

The third signal was the quietest and the most telling, because it came from the best operator in the business. Mister Car Wash went public in June 2021 at about $15 a share and touched $23.53 that July. In February 2026, Leonard Green, which already controlled about two-thirds of the stock, took it private at $7.00 a share, an enterprise value of about $3.1 billion and roughly nine times 2025 adjusted earnings. Leonard Green's opening bid had been $6.25. The deal closed on May 19. Mister had done almost everything right: 549 locations, 2.47 million members, revenue still growing 6 percent. The public market valued the whole thing at about $5.6 million a site†, which is to say roughly replacement cost. For the category leader, after a decade of growth, the premium for being a chain had gone to zero.

Three transactions, three price points: $1 million a site for a distressed strategic seller, $2.5 million of debt a site for a chain that could not refinance, $5.6 million a site for the best in class. The cost to build a new one has not moved. That spread, between what a tunnel costs and what a buyer will pay once it is standing, is the bubble. It is not a bubble in whether people want clean cars. It is a bubble in development economics.

Price, Not People

The industry's defenders have a fair point, and it is worth stating plainly: the money is still coming in. Mister's first-quarter 2026 revenue rose 6 percent to $277.9 million. Members accounted for 76 percent of its wash sales, up from 73 percent a year earlier. Eric Wulf, the chief executive of the International Carwash Association, told attendees at the trade group's spring show that tunnel revenue had grown 5 to 7 percent over the past year and that 2021 and 2022 were the anomaly, not the new normal. The ICA's own first-quarter Pulse survey found demand stable.

Look at where the growth comes from, though, and a different picture emerges. The same survey found that operators ranked market saturation as their top concern for 2026, ahead of labor and ahead of interest rates. Mister's same-store sales grew 1.2 percent in the second quarter of 2025, 1.6 percent in the fourth, and 3.9 percent in the first quarter of this year. Its revenue per member rose about 4 percent to $29.23, almost entirely because it raised the price of its base plan. Growth was coming from charging existing customers more, not from more customers.

The membership software company Rinsed, whose panel covers more than 3,000 sites, puts figures on this. In the second quarter of 2026, membership revenue at its clients rose 10.4 percent while revenue from walk-in retail customers fell 3.9 percent. Monthly churn hit 7.9 percent, the highest the company has recorded, with voluntary cancellations up 9 percent from a year earlier. Nearly three in ten new sign-ups were people who had been members before and quit. Rinsed's analysts estimate that as much as half of the industry's membership revenue growth now comes from price increases rather than new members. Steve Gaudreau of Brink Results, a consultancy, puts typical churn at "between 7% and 10% monthly."

Seven or eight percent a month means a tunnel's entire member base turns over in roughly a year. The membership is not an annuity. It is a treadmill, and the treadmill speeds up every time a competitor opens across the intersection with a $10 introductory offer. Lenders who underwrote these loans on recurring revenue would do well to treat churn the way an apartment lender treats vacancy: not as a footnote but as the variable that decides whether the debt gets paid.

The most candid assessment came from the man running the biggest chain. On Mister's first-quarter 2025 earnings call, John Lai, the chief executive, said "rationality is setting in as our competitors are reevaluating their growth trajectories." A quarter later he referred to "the irrational behavior we've seen over the last five years." That is the leader of the industry describing the industry's 2020-to-2024 build-out as irrational, on the record, to analysts. Nine months after that, his company was private and no longer had to explain itself to anyone.

The Landlords Have Not Gotten the Memo

Here is the strangest number in the whole story. In April 2026, the net-lease brokerage B+E counted about 190 car wash properties listed for sale nationally, at an average capitalization rate of 6.21 percent, an average lease term of nineteen years and an average price of $5.11 million. In the same season, The Boulder Group's survey of all single-tenant net-lease property put the average cap rate at 6.82 percent, rising to 6.92 percent by the third quarter, a level the firm called the highest in more than a decade.

Read that again. Buyers were paying more, per dollar of rent, for a car wash than for the average net-lease building in America. The average net-lease building is a Dollar General or a Walgreens with an investment-grade corporate guarantee and four walls that can become a vape shop or a dentist's office when the tenant leaves. A car wash is a 4,000-square-foot concrete tunnel full of conveyor track and blowers, leased to a private company with no credit rating, that cannot be converted to anything else without a demolition permit. Whistle Express properties were trading at a 5.99 percent cap; Quick Quack's at 5.56 percent.

Part of the explanation is tax policy. The 2025 federal budget law restored full bonus depreciation, and a car wash, classified as a 15-year property, lets an owner write off most of the purchase price in year one. That drew a wave of individual 1031 exchange buyers and family offices who were pricing the tax shield, not the tenant. Essential Properties Realty Trust, one of the largest institutional owners of car wash real estate, sold 19 of them in the fourth quarter of 2025 for $48.1 million; its chief executive told analysts the firm was selling into a bid driven by accelerated depreciation, and its car wash exposure fell to 13.7 percent of rent. Getty Realty, whose portfolio is 8 percent car washes, said this spring that it now underwrites the category to a higher coverage ratio than it did five years ago, and that its bucket of tenants with rent coverage below 1.0x grew because of newly built tunnels still ramping up.

The arithmetic of the sale-leaseback, which looked like free money on the way up, works in reverse on the way down. Sell a $5.11 million property at a 6.21 percent cap and you have committed to about $317,000 a year in rent before any escalator†. Cover that 1.5 times, which is what a cautious landlord wants, and the tunnel needs roughly $475,000 of annual operating profit†. A mature express wash in a good location can clear that. An average one cannot: industry-wide operating margins run about 14 percent, against the 35 to 50 percent four-wall margins the pro formas assume. The gap between those two numbers is where the rent goes unpaid. Even Mister Car Wash, which reported no new sale-leasebacks in the first quarter of 2026, carried 493 car wash leases and saw its rent expense rise 7 percent to $31.8 million in that quarter alone. Zips rejected 41 leases in bankruptcy. Somebody owned those buildings.

The large public landlords have mostly been careful. Agree Realty's investor materials say it has deliberately avoided or sold out of car washes, alongside movie theaters. Realty Income does not break the category out, and no car wash operator appears among its twenty largest tenants. The exception is NNN REIT, where Mister Car Wash is the second-largest tenant at 120 properties and 3.7 percent of rent, with Tidal Wave and Super Star Car Wash adding another 2.7 percent. NNN reports third-quarter results on November 4. John Feeney, a senior vice president at Boulder, said after the Zips filing that "investors will scrutinize any future investments into car wash real estate." The listing data suggests many of them have not started.

What the Banks Already Know

The clearest early-warning system for an overbuilt category is not the trade press or the stock market. It is the loan desk. And the loan desk pulled back two years before anyone wrote a headline.

The federal 7(a) and 504 loan data shows 5,807 approvals totaling $6.83 billion to car washes since fiscal 2010, about 4,860 of which were actually funded. The 7(a) program, the workhorse for independent operators, approved 391 car wash loans in fiscal 2021. By fiscal 2025 that was down to 243, a 38 percent drop (Innowave calculation). The first half of fiscal 2026 annualizes to about 162 (Innowave calculation), which would be the lowest count in at least fifteen years. The median 7(a) car wash loan shrank from $1.23 million in 2021 to $428,000 in 2025, as the big ground-up construction loans migrated to the fixed-rate 504 program or stopped getting made at all.

The charge-off numbers look benign: about 5.25 percent of resolved 7(a) car wash loans, against 7.45 percent across all industries. But that comparison is less reassuring than it sounds, for a reason anyone who lived through 2008 will recognize. Loans default in years three through six, not years one and two. The 2021-to-2024 vintages, written at peak construction costs into the most crowded corridors, have barely begun to season. The low loss rate reflects the loans that were made before the boom, not the ones made during it.

What a default looks like in practice is visible on the business-for-sale sites. In Big Spring, Texas, a population of about 26,000 on Interstate 20, an express wash built in 2024 for more than $3.5 million was listed this year at $1.65 million, land included. That is not a distressed sale. It is simply what a tunnel in a small West Texas town is worth once the developer's optimism has been subtracted from the price.

The feasibility math has gotten harder to pretend about. Published lender models for a $5 million to $6 million express project put the break-even before debt service at roughly 150 to 160 cars a day and the point at which the loan covers itself at 250 to 300. A third tunnel in a two-tunnel trade area does not get to 250. Beginning with applications received on October 1, 2026, the revised federal standard operating procedure for 7(a) and 504 loans requires acquisition loans to show 1.25 times debt coverage on the business's actual historical cash flow, not the buyer's projections. That single sentence closes the door on the leveraged purchase of an underperforming wash by someone who believes they can run it better, which was for several years the main thing holding up prices in the resale market.

The pattern the data describes is a familiar one. Credit did not cause the bubble; the sale-leaseback and the membership model did that. But credit withdrew first, quietly, and the withdrawal is what will determine how long the adjustment takes. Tunnels do not close the way restaurants do. They sit, under-trafficked, servicing debt from the member base, until a lease matures or a lender loses patience.

The Moat Nobody Meant to Build

Which brings the story back to places like Harlingen, and to a twist that the industry has been slow to notice: the city councils trying to stop the boom are, without intending to, handing the winners of it a permanent advantage.

A moratorium hurts a developer with a site under contract. It does nothing to the tunnel already running across the street, and it does something quite good for the one that would otherwise have faced a new competitor. Every two-mile spacing rule is a zoning-granted exclusive territory for whoever got there first. The industry's trade groups have lobbied against these ordinances. Its largest incumbents, if they were honest, would send flowers.

What began as temporary pauses has been hardening into permanent code. Hesperia, California, in the high desert along Interstate 15, passed a 45-day moratorium in February 2023 after approving five new washes in four years, then adopted permanent standards that June requiring express tunnels to sit at least 2,000 feet from one another. Brook Park, Ohio, a Cleveland suburb of about 18,000 with more than six car washes, enacted a one-year moratorium in May 2023 and a year later replaced it with a cap of one car wash per 10,000 residents, which works out to fewer than two (Innowave calculation) for a city that already has six. Nobody is closing the existing ones. They now operate behind a wall.

The Rio Grande Valley did the same thing in the language of water rather than land use. Brownsville set a five-mile spacing rule in the spring of 2024. Edinburg followed on September 17, 2024, folding a two-mile rule and a recycling mandate into the drought-contingency chapter of its city code; within three months it had granted a variance to Quick Quack for a site 1.3 miles from the nearest competitor. Harlingen adopted its two-mile rule in November. Mission, the last holdout, passed a one-mile rule in January 2026 and then a retroactive 180-day moratorium in March, and was promptly sued by a mayoral candidate who argued the city had skipped the notice-and-hearing steps that Texas House Bill 2559 now requires before a municipality can freeze development. That case is a preview of the next phase: state legislatures, prodded by developers, limiting what cities can do.

The list is not exhaustive and it does not include the cities, like Whittier, California, and Ceres, California, where councils have directed staff to study a pause without yet imposing one, or Mesa, Arizona, where a fifteen-year-old rule limiting auto uses near arterial intersections has drawn a $2 million takings claim. Warren, Michigan, which froze new washes and gas stations in May, counted 25 of the former and 69 of the latter in 35 square miles. The council member in Whittier who asked for a study said there were "way too many car washes being built in the city." It is hard to find anyone outside the industry who disagrees.

For anyone underwriting a car wash today, the zoning map has become as important as the traffic count. A tunnel inside a spacing radius is a protected asset. A tunnel outside one, in a city that has not yet acted, is a tunnel waiting for its competitor. The difference between those two is now worth more than the difference between a good operator and a bad one.

What Self-Storage Had That Car Washes Don't

America overbuilds a retail format about once a decade, and the comparisons are instructive mostly for how they differ.

The optimists point to self-storage, which went from 3.87 square feet per American in 2000 to 6.78 in 2007, saw occupancy slip, and then absorbed the glut so completely that the country now has 7.28 square feet per person and nobody calls it a bubble. The difference is demand. The share of households renting a storage unit rose from about 9 percent in 2005 to 12.6 percent in 2024, a 40 percent increase in the customer base that soaked up the excess supply. Car washes have no equivalent runway. Professional-wash usage is already at 79 percent of drivers. There is no second act in which a third of non-users convert.

The pessimists point to Quiznos, which peaked near 4,700 units in 2007 and had fewer than 400 by 2017, or to frozen yogurt, which lost about 15 percent of its chain locations between 2018 and 2020. Those comparisons are imperfect in the other direction. A failed sandwich franchise strands a franchisee and a strip-mall landlord who can re-tenant the space in six months. A failed car wash strands a $5 million special-purpose building that cannot be anything but a car wash, on a pad that was bought at car wash prices, under a lease that may have fifteen years left on it. The capital at risk per unit is an order of magnitude larger and far less liquid.

So the shakeout will not look like a wave of closures. It will look like a slow re-pricing. Expect the following, in roughly this order. First, the independents who built in 2022 and 2023 at the top of the cost curve and are now reaching the end of their interest-only periods will sell, at Big Spring prices, to the regional chains, who will run the tunnels on their existing overhead and not care that the buildings cost twice what they paid. Second, the 1031 buyers who bought car wash net leases at sub-6 percent caps for the depreciation will find, at the first lease renewal or the first tenant failure, that the dark value of a tunnel is the land minus the demolition. Third, the private equity sponsors who cannot find an exit at the multiples they paid will do what Wildcat Capital is reportedly doing with Club Car Wash and Express Wash Concepts: roll the assets into a continuation vehicle, which is the polite term for selling a company to yourself at a price you set.

It is also worth watching what the consolidators are actually doing, as opposed to saying. Whistle Express bought Driven's 380 sites in early 2025 and announced a chain of 530 locations. Within eight months it had sold 53 of them to Tsunami Express, 13 in Utah to Quick Quack and 5 in Lubbock to Mister Car Wash, and now describes itself as having "more than 450." The biggest buyer of the cycle has been a net seller ever since. That is not a company that thinks every tunnel is worth owning.

The Quiet Part

The most consequential change of 2026 is one that almost no one outside the industry registered. With Mister Car Wash private and Driven Brands gone, there is no longer a single publicly traded American car wash operator. For five years, Mister's quarterly filings were the one place where anyone, a banker in Tulsa, a city planner in Hesperia, a 1031 buyer in Scottsdale, could see audited numbers on membership, churn, rent and same-store sales. That window closed on May 19. What remains are vendor panels, broker surveys and press releases, all of them produced by people with something to sell.

That opacity is not an accident of timing. It is what the end of a cycle looks like. Price discovery stops being public at exactly the moment the prices start to fall, because the people holding the assets would rather not discover them. The continuation vehicles, the seller notes, the lender-owned restructurings: each is a way of avoiding a mark.

For those who still need one, a few dates matter. NNN REIT reports on November 4 and is the only large landlord that names its car wash tenants. Rinsed's third-quarter report, due this month, will say whether churn kept climbing past 7.9 percent. The federal 7(a) and 504 loan data for the full fiscal year will show whether fiscal 2026 was in fact the lowest year for car wash lending since the program started tracking it. And somewhere in Texas, another city commission is reading a water report.

In Harlingen, the two-mile rule has held. Blue Wave is still open. So are the twenty others. The developers who were waiting on permits went elsewhere, to towns that have not yet done the math. The tunnels already standing have, in effect, been granted something the market never gave them: a reason to believe the one down the street will never be built.

That is not a bad ending for them. It is a worse one for everyone who paid 2022 prices for the belief that it would.

† Innowave calculation from the deal terms and filings named in this article.

Figures in this article are drawn from SEC filings by Mister Car Wash and Driven Brands; the Zips Car Wash Chapter 11 docket (N.D. Tex., Case 25-80069); the federal 7(a) and 504 loan data (FOIA files) as compiled by WashIndex and MMCG Invest; the International Carwash Association's CAR WASH Pulse survey; Rinsed's quarterly industry reports; B+E and The Boulder Group net-lease research; Raymond James's Car Wash Insight (Spring 2026); the Census Bureau's 2023 County Business Patterns; and municipal records and local reporting in each jurisdiction named. Opening counts are industry estimates; no government agency tracks car wash construction. Ordinance numbers for Harlingen and Edinburg could not be confirmed from primary text and are cited by date and code section. Per-site valuation figures, marked Innowave calculation or † in the text, are calculated from disclosed deal terms.