Business

Business

Copart runs the auction, not the car lot. In fiscal 2025 it turned $4.65 billion of revenue into $1.70 billion of operating income — a 36.5% margin — mostly by charging fees to both sides of an online salvage auction it never takes inventory in. The result is a lightly-taxed, cash-generative toll on a flow of totaled vehicles, funded by insurers, sitting on $4.8 billion of cash and T-bills with almost no debt. This chapter traces that engine and fixes the question the rest of the report works through.

The toll on totaled cars

The core mechanic is narrow and durable. An insurer decides a damaged car is a total loss, hands it to Copart, and Copart sells it through its online VB3 auction to a global pool of registered buyers — dismantlers, rebuilders, dealers, exporters, and the public. In the United States and most markets Copart acts as an agent: it never owns the car, and it books revenue as the fees it charges for remarketing the vehicle, not the vehicle's selling price [1]. Insurance companies supplied 81% of the vehicles it processed in fiscal 2025 [2].

Copart charges both sides. Sellers pay listing, selling, transportation, storage, and title-processing fees; buyers pay purchasing and bidding fees. Every one of these is recognized as net revenue at the time of auction, not the gross price of the vehicle [3]. The flagship seller arrangement, the Percentage Incentive Program, pays Copart a predetermined percentage of each vehicle's sale price — so its own revenue rises and falls with the hammer price, aligning it with the seller to merchandise the car for the highest bid [4]. A parallel Consignment Program charges a fixed fee instead [5].

The exception matters for reading the income statement. Under a Purchase Program used primarily in the U.K., Copart buys the salvage vehicle outright at a formula price — a percentage of its estimated pre-accident value — and resells it for its own account, taking title and inventory risk [6]. Those cars flow through revenue at their full selling price. The arena mechanics — total-loss economics, insurer supplier power, and the buyer liquidity that makes the auction work — are laid out in Industry; what follows is how they show up in Copart's numbers.

A fee business wearing a retailer's revenue line

Copart reports two revenue lines, and they behave nothing alike. Service revenue — the agency fees — was $3.97 billion in fiscal 2025, 85% of the total. Vehicle sales — the purchased cars, mostly U.K. — was $678 million, 15% [7].

Source: FY2025 10-K, Consolidated Statements of Income (FY2023–FY2025) [8]; FY2021–FY2022 from the FY2022 10-K income statement [9].

The chart carries the point: every dollar of growth over five years came from the fee line. Purchased-vehicle sales have been flat near $650–680 million since fiscal 2022, and management notes the U.K. principal book is shrinking as sellers there migrate onto the consignment model [10]. That is the business voting with its own contracts: the agency model is where the economics live.

They live there because purchased vehicles are a low-margin pass-through. In fiscal 2025 the $678 million of vehicle sales carried $603 million of vehicle cost, leaving roughly $75 million of gross profit — an 11% margin on that line [11]. The fee line, by contrast, has almost no matched cost of goods; its expenses are the yards and the people that run the auctions. So the 15% of revenue that looks like retailing contributes a sliver of profit, and the 85% that is fees does the work.

Operating leverage, and where it stalled

The fee model produces margins few industrials reach. Operating margin was 36.5% in fiscal 2025 [12]. But the more useful fact is the direction: margin has compressed every year since a fiscal 2021 peak of 42.2%.

Source: derived from reported operating income and revenue, FY2025 10-K [13] and FY2022 10-K [14].

The compression is not a demand problem — revenue grew 73% across the window. It is cost. Facility operations rose from 39% of revenue in fiscal 2023 to 42% in fiscal 2025, and general and administrative from 7% to 9%, as Copart added yards, staffed for catastrophe response, and invested in technology [15]. The fiscal 2021 peak also flattered the base — pandemic-era auction prices were unusually high against a lean cost structure. Operating leverage is real over the long arc, but it has run in reverse recently, and "asset-light" is the wrong label for a business whose costs are 281 physical yards of labor and land [16]. The counter-fact to that read is that revenue keeps compounding through the cost build, so the margin give-back has cost shareholders growth, not dollars.

The two reported segments show where the quality sits: derived from the segment note, U.S. operations earned a 38.4% operating margin in fiscal 2025 versus 27.3% internationally [17], so the mature domestic network subsidizes a still-scaling overseas footprint.

Sources: revenue and operating margin, FY2025 10-K income statement [18]; cash and held-to-maturity investments, FY2025 10-K balance sheet [19]; free cash flow derived as operating cash flow less capital expenditures [20].

The balance sheet: land, cash, and no lender

Copart's balance sheet is unusual for how little of it is financed by anyone but retained earnings. At July 31, 2025 total assets were $10.1 billion against total liabilities of just $883 million — and there is essentially no debt in that figure [21]. Two asset classes dominate. The first is the operating base: $3.60 billion of property and equipment, the acreage under 281 auction yards worldwide [22]. Land is the moat the fee model is built on — permitted, zoned salvage yards near population centers are slow and costly to replicate — and Copart owns much of it outright rather than renting.

The second is a war chest: $2.78 billion of cash and $2.01 billion of held-to-maturity securities, $4.79 billion together, largely U.S. Treasury bills [23]. That pile now earns real money: net interest income was $179 million in fiscal 2025, up from $66 million two years earlier, and it lifts pre-tax income by roughly a tenth [24]. The same cash is also the report's first open question about capital: it sat idle for years while the company paid no dividend and bought back no stock. How that changed is taken up in History.

Cash conversion is the through-line of the model. The auction takes no working capital to grow — Copart collects from buyers and pays sellers, so receivables and a tiny inventory are the whole of it — and operating cash flow reached $1.80 billion in fiscal 2025 [25]. After $570 million of capital spending — most of it land and yard build-out, not maintenance — free cash flow was $1.23 billion, close to 80% of net income [26].

Source: operating cash flow, FY2025 10-K liquidity discussion [27]; free cash flow derived as operating cash flow less capital expenditures; FY2021–FY2022 as reported.

The reinvestment need is modest and largely discretionary — capex is going into new yards that expand capacity, so a slower-growth Copart could convert an even higher share of earnings to cash. That optionality is real, and it is why the balance sheet, not the income statement, is where the capital-allocation debate will land in later chapters.

What the engine is levered to

The model carries two concentrations. Supply is concentrated: insurers are 81% of the units, so the addressable flow rises and falls with claims volumes and with how often a damaged car is written off rather than repaired — the total-loss frequency [28]. Revenue is also levered to price, because a large share of service fees is tied to the vehicle's selling price; Copart itself flags that used-car values, commodity prices, and the mix of cars sold move that price, and that it cannot quantify their effect in advance [29].

The two long-run tailwinds are structural but not guaranteed. Total-loss frequency has trended up for three decades as repair costs rose against used-car values, and the average car on U.S. roads aged from 11.1 years in 2012 to 12.8 in 2025 — an older, more totaled fleet feeds the auction [30]. The buyer side is what converts that supply into price; the reach of Copart's global, cross-border buyer pool is laid out in Industry. Recent U.S. insurance-unit softness, and whether it reflects lost share or a shrinking claim pool, is a live debate carried in Competition.

The central question this report follows is whether Copart's fee-based auction toll can keep compounding units and hold operating margins near 36% as its two engines — a rising total-loss frequency and record auction prices — mature, and whether a balance sheet holding $4.8 billion of near-idle cash is a strength or a question of its own.