History

History

The primary record for Copart runs deep from fiscal 2020 forward — five annual reports, the earnings calls from fiscal 2022 onward, and the 8-K stream — over a corporate spine that reaches back to a single 1982 salvage yard. Copart was founded in 1982 with one auction facility [1], incorporated in California and reincorporated in Delaware in January 2012 [2], and has traded on NASDAQ under CPRT since March 17, 1994, paying no cash dividend in the three decades since [3].

Within the covered window the record has a few clean breaks: a used-vehicle super-cycle that lifted selling prices and margins into fiscal 2021–2022; a margin-compression stretch as land, technology, and storm costs outran volume from fiscal 2024; a widening gap between rising total-loss frequency and flat-to-declining insurance unit volumes; a three-CEO reshuffle in four years that ended with the founder-era chief executive returning in mid-2026; and a capital-allocation pivot in fiscal 2026 — the first material buybacks since 2019 — against a share price sitting roughly 54% below its trailing three-year high. This tab records those beats, the promises management made against them, and how the explanations changed. It does not judge them.

The arc in dated beats

Sources: FY2025 proxy, "Our History Drives Our Future" [3]; FY2025 10-K, listing and international history [3][4]; FY2024 10-K, acquisitions and stock splits [5][6]; officer 8-Ks (2024, 2026) [7][8].

The business model is the through-line. Copart runs a 100% online auction — its VB3 virtual-bidding platform, patented in 2008, is what let it eliminate the expense and capital of live auctions [9]. The international footprint was built by acquisition, in a sequence the 10-K states plainly: "We first expanded our operations outside the U.S. in fiscal 2003 with an acquisition in Canada," followed by the U.K. in fiscal 2007–2008, the UAE, Brazil, Germany, and Spain in fiscal 2013, and later entries through fiscal 2022 [10].

Revenue and margins, FY2020–FY2025

Source: FY2025 10-K, Consolidated Statements of Income (FY2023–FY2025) [11]; FY2023 10-K (FY2020–FY2022) [12].

Revenue more than doubled over six years, from $2.21 billion in fiscal 2020 to $4.65 billion in fiscal 2025 [13][14]. Operating margin tells a second story: it peaked at 42.2% in fiscal 2021, at the height of the used-vehicle boom, then drifted lower every year to 36.5% in fiscal 2025 — the arithmetic behind management's repeated promise of "operating leverage" and the record of what actually happened to margins over the same span.

Source: derived from reported operating income and revenue, FY2023 and FY2025 10-Ks [15][16].

Leadership: three chief executives in four years

Control has stayed close to the founder. Willis J. Johnson founded Copart, served as CEO from 1982 to February 2010, and has chaired the board since January 2004 [17]. His son-in-law, A. Jayson (Jay) Adair, rose from manager of operations in 1989 to president in 1996 and CEO in February 2010; he became co-CEO in April 2022 and Executive Chairman in April 2024. Jeff Liaw joined as CFO in September 2016, became president, then co-CEO in 2022, and sole CEO in April 2024 [18]. The March 2024 succession made Liaw "the Company's sole principal executive officer and principal operating officer" effective April 1, 2024 [19]. Leah C. Stearns has been CFO since December 2022 [20].

That planned succession reversed roughly two years later. On June 29, 2026 the board appointed Adair — the Executive Chairman — as CEO again, effective July 31, 2026, with Liaw stepping down as CEO and resigning from the board; the filing states his "decision to resign was not the result of any disagreement with the Company" [21]. The press release called Liaw the company's "third-ever CEO" and named him a Special Advisor to Adair [22]. His separation terms include a $450,000 lump sum plus $200,000 during the transition, waived equity holding periods, and eliminated option price hurdles [23]. On July 8, 2026 the company named Jane Pocock, its UK chief executive, as President effective August 1, 2026 [24].

Capital allocation

For the covered years, capital allocation reads as reinvest-and-accumulate. Copart returned effectively nothing to shareholders: it has paid no dividend since 1994, and despite an open repurchase authorization it bought back no stock in fiscal 2021 through 2025 [3][16]. The two 2-for-1 splits of November 2022 and August 2023 were cosmetic, not returns of capital [25].

Sources: capex per MD and A, land and facility driven [26]; free cash flow derived from reported operating cash flow less capital expenditures, FY2020–FY2025; zero repurchases FY2021–FY2025 [16].

Instead the cash went two places. First, into land and facilities: capital expenditure ran roughly $340–590 million a year, "primarily related to acquiring land, opening and improving facilities, capitalized software development costs… and lease buyouts" [27]. Second, into the balance sheet: by July 31, 2025 Copart held $2.78 billion of cash and restricted cash plus $2.01 billion of held-to-maturity securities — roughly $4.8 billion — against an undrawn revolver [28][29]. On what to do with it, the FY2025 10-K says only that the company is "considering a variety of alternative potential uses for our remaining cash balances and our cash flows from operations" [30].

The two acquisitions in the window were small and largely non-cash — which is why the cash-flow acquisition line reads near zero. Hills Motors, a U.K. parts recycler, was bought for $106.6 million cash in fiscal 2022; Purple Wave, an online heavy-equipment auctioneer, was an 80% stake acquired in October 2023 for $112.1 million in Copart stock [31]. The earlier National Powersport Auctions deal (fiscal 2017) is named but its economics are not disclosed in the covered filings [32].

The buyback posture, said and then done

The clearest said-versus-did event in the record is the buyback. Copart framed repurchases as rare and opportunistic — "to the extent we had excess cash, we have leveraged the share repurchase program, but have done so very opportunistically" (Feb 2024) [33]. Asked directly in November 2025, management deflected: there "for sure will come a day," but the how, when, and where was "a conversation for another day" [34]. One quarter later it had begun buying: over 13 million shares for more than $500 million by February 2026 [35], scaling to more than 43.4 million shares for over $1.6 billion fiscal-year-to-date by the May 2026 call — the first material repurchases since roughly $365 million in fiscal 2019 [36].

Sources: no repurchases FY2021–FY2025 [16]; FY2026 repurchases per Q2 and Q3 FY2026 calls [37][38].

The debt side stayed conservative throughout: the December 2021 facility carried a $1.25 billion revolver that was never drawn, and in January 2026 Copart replaced it with a new $1.25 billion unsecured revolver maturing in 2031 [39][40].

Management said versus what followed

Copart declines to give numeric guidance — "as a rule, we don't provide forward-looking guidance" (May 2024) [41]. What it puts on the record instead is a stable set of qualitative commitments — reinvestment ahead of returns, and "operating leverage" over the long run. The table below pairs the commitment with the measurable outcome that followed.

Sources: no-guidance policy [42]; reinvestment and operating-leverage commitments [43][44]; "partially recede" and Purple Wave GTV target [45]; margin compression [46]; buyback reversal [47][48].

The reinvestment commitment was kept and quantified: in Q1 FY2024 the CFO put "our first priority is to deploy capital to grow our core business" alongside "over the long run, we continue to expect operating leverage as we grow," with roughly 80% of that quarter's $162 million of capex going to physical capacity [49]. The operating-leverage half is the one to watch against the margin line above — the same promise had been voiced by the prior CFO in fiscal 2022 [50], yet gross margin fell about 340 basis points to 42.4% in the fourth quarter of fiscal 2024 [51], with U.S. yard depreciation and amortization named as the driver [52].

Explanation drift

The most instructive part of the record is how management's account of the same recurring problems moved over time. Three threads stand out.

Total-loss frequency versus volume

Copart's volume thesis rests on total-loss frequency — the share of damaged vehicles insurers write off rather than repair. The framing shifted as the numbers stopped cooperating.

FY2022 (Nov): "Total loss frequency is negatively correlated with used car prices." — Jeff Liaw [53]

FY2022 (May): total-loss frequency "has declined over the course of the past 12 months," but "we believe the market will ultimately revert to the historical norm of steadily rising total loss frequency." [54]

FY2024 (Nov): frequency "troughed at just north of 17%" in mid-2022 and "is now 19.3%," with U.S. insurance volumes up 9.7% but "still… suppressed." [55]

FY2025 (Feb): frequency "hit 23.8%, an all-time high, though a portion of that is attributable to those storm events." [56]

FY2026 (Nov): with frequency flattening, "reading a whole lot into 80 basis points versus 130 or versus plus 30… is more noise than it is signal" — the same quarter U.S. insurance units fell 9.5%. [57][58]

FY2026 (May): "we are very much not passive beneficiaries of an increase in total loss frequency. We have helped to drive it upwards." [59]

The account travelled from a macro correlation, to a multi-decade trend that would "revert," to a hedge that all-time-high frequency was storm-inflated, to near-term flattening being "noise," to Copart itself as an active cause of rising frequency. Across the same span reported U.S. insurance volumes went from +9.7% to roughly -4% [60][61].

Catastrophe events, from "incremental" to "not profitable"

Storms recur in the record with a changing frame — first as a small cost of serving insurers, later as an outright drag. Hurricane Ida in Q1 FY2022 was "an operating loss from the event… of a few million dollars," roughly 100–150 basis points of gross margin [62]. Hurricane Ian added "approximately $25 million" of cost in Q1 FY2023 [63]; Helene and Milton cost $29 million in Q1 FY2025 [64]. By September 2025 the framing had turned: "on a truly fully loaded basis over a long horizon… catastrophic events are surely not… profitable for Copart" [65]. Those same prior-year storms then became the comparison that suppressed every fiscal 2026 print — the prior year "included revenue from over 49,000 CAT-related vehicles" [66].

Selling prices, benchmarked to a moving yardstick

As used-vehicle values fell from their fiscal 2022 peak — worldwide selling prices had risen 13% year over year that May [67] — Copart began benchmarking its own prices against the Manheim index to show resilience. In Q3 FY2024, U.S. insurance selling prices were down "less than 2%" against a "nearly 14%" Manheim decline [68]; by September 2025 management said its prices grew "more than fivefold that of service providers similar to ours" [69]. As the comparison narrowed in fiscal 2026 — U.S. insurance ASPs up 4.1% — the label held: "a seasonally adjusted all-time record high" [70].

KPI and segment definitions

Two definitional changes matter for anyone comparing periods.

The first is a reporting change: beginning with the Q1 FY2026 results (November 20, 2025), Copart added formal segment reporting — "The Company has provided segment reporting information as a new table within this press release" [71]. The new table splits results into United States and International, showing U.S. operating income of $375.0 million and International of $55.7 million for the quarter [72]. The change was disclosed in the release and 10-Q rather than explained on the call.

The second is a measurement habit rather than a formal restatement: how management defines the growth number it emphasizes. It introduced an ex-Hurricane-Ian "normalized" volume frame in Q2 FY2024, then a quarter later grew "somewhat hesitant to provide, quote, normalized growth trends that exclude the effect of storms" [73]. By fiscal 2026 the emphasis had shifted again — to units "excluding catastrophic activity" and to a "direct buy" referral model that removes low-value cars from reported purchased-unit counts, making reported U.S. purchase units fall 23.6% but "just 8% on a normalized basis" [74]. The headline growth figure and its basis moved together; each period's basis is worth reading before comparing it to the last.