Copart, Inc. CPRT

Key data

The business

Copart runs online auctions for wrecked and salvaged cars. When an insurer declares your crashed vehicle a total loss, it hands the car to Copart, which stores it in one of hundreds of yards, photographs it, and sells it to the highest bidder among roughly a million registered buyers across 185 countries, many of them dismantlers and rebuilders overseas. Copart takes a fee from the seller and the buyer on each car, and on some vehicles it buys the car outright and resells it for its own account, a lower-margin principal model that carries inventory risk. The moat is two things at once: a two-sided network where sellers go for the buyers and buyers go for the supply, and the physical land itself, licensed yards near major cities that are nearly impossible to permit and assemble today. Copart and IAA split most of the North American salvage market. No single customer was more than 10% of revenue over the last nine months, but a handful of large insurers together drive most of the volume, and they have walked from specific markets before.

The numbers

The last five quarters show a business that stopped growing and then defended its profit by other means.

The soft quarter was January 2026: operating income fell to $388.7M, the weakest in over a year. April snapped back to a record $464.3M, yet net income of $402.4M still landed below the $406.6M of a year earlier. Diluted EPS rose anyway, $0.43 against $0.42, because the share count keeps shrinking. That is the whole tension in one line: profit dollars are flat, per-share profit inches up on buybacks. Revenue says the same, up just 2.1% in April and dead flat across nine months, $3.5B against $3.5B. The earlier worry, that growth under 3% with no volume pickup would keep the multiple under pressure, held: nine months in, there is still no reacceleration.

Step back and the compounding is real but decelerating. From fiscal 2021 to fiscal 2025, EPS grew from $0.97 to $1.59, about 13% a year, and operating cash flow compounded faster still, near 16%. Then fiscal 2026 flattened: nine-month operating income of $1.3B is level with the $1.3B a year earlier, and nine-month net income is flat to the dollar. The machine that reliably added 10% to 13% a year has, for three quarters, added nothing. Underneath that flat line, the mix drifted the wrong way.

Service revenue, the high-margin auction and fee business that is the whole point of owning Copart, slipped 0.4%, while lower-margin principal-model vehicle sales rose. The one non-consensus belief here is narrow: that flat assignment volume is cyclical, tied to used-car values and total-loss frequency, not structural. The print that settles it is the next revenue growth number.

Management

In June 2026 Jeff Liaw stepped down as chief executive, becoming a senior advisor through July 2027, and Jay Adair returned to the CEO seat. Adair's return puts capital allocation back in the hands of the executive who oversaw the company's expansion. Insiders have been sellers, not buyers: ten sales worth about $20.8M over the year and no purchases, led by director Thomas Tryforos at roughly $14.8M across September and November 2025 and director Daniel Englander at $2.2M in July 2026; plan status is not disclosed in the filings reviewed, so read them as ordinary director diversification. The balance sheet is a fortress, net cash, effectively no long-term debt, a current ratio above seven and working capital of roughly $4.5B, which is exactly what lets a flat year pass without stress.

How it fails or surprises you

The flat line is the trend, not a pause. Nine-month revenue and operating income are both flat to last year, the fact this read explains least. If total-loss frequency keeps falling and used-car values stay soft, assignment volume does not recover, and 21 times earnings compresses toward the 18 times floor of its twelve-year range. First tell: a fourth straight sub-3% revenue quarter when fiscal Q4 reports around September.

A founder with working capital. Adair returns to a company sitting on roughly $4.5B of working capital, no dividend, and only modest buybacks. The risk is a large, out-of-character acquisition that spends the cash at a worse return than the yards earn. Tell: an 8-K announcing a deal or a step-change in buyback pace.

Volume and margin reaccelerate together (right tail). If used-car values firm and international, chiefly Germany and the UK, picks up, assignment growth can return to high single digits while incremental margins on the fixed yard base run above 40%. At the low end of its historical multiple, the market pays nothing for this. Tell: a quarter with revenue growth back above high-single-digits and the operating margin widening, visible first in the service line.

Closing thoughts

The read turns on whether fiscal fourth-quarter revenue, reporting around September, shows growth returning or another flat print. Another flat quarter confirms that a 13% compounder has become a no-growth utility and validates the slow re-rating toward 18 times. A return to mid-single-digit or better growth converts today's 21 times, near the cheapest this stock has traded in twelve years, into an unusual entry on a genuine duopoly. An ambiguous print, 2% to 3% again, leaves the standoff intact and the multiple drifting lower, which is what has happened for three quarters. The left tail, a structural decline in cars written off, is real but slow and survivable: net cash and no debt mean a flat decade dents the return, it does not impair the business.

The bet is still that when your car gets totaled, the insurer keeps handing it to Copart to auction, and that cars keep getting wrecked and written off every year. What breaks it is volume, and one pair of numbers tells you first: assignment-driven revenue growth against the operating margin. Flat revenue with a flat margin says the compounder has stalled; growth returning with the margin widening says the market mispriced a durable franchise at the low end of its range.