The Moat

The Moat

Copart's fee toll earns a 36.5% operating margin because something protects it. This chapter tests what. The advantage is real and shows up in prices, margins, and returns: Copart's operating margin is more than double that of RB Global, the only scaled salvage rival, and management's own disclosures show its auction consistently clears higher prices. The moat rests on auction liquidity, permitted land, and integration into insurers' claims workflows. The open question is whether recently soft assignment volume reflects a claims-frequency cycle or the start of share loss.

The advantage shows up in prices and margins

The advantage appears in the numbers. In FY2025 the business converted $4.65 billion of service revenue and vehicle sales into $1.70 billion of operating income — a 36.5% margin [1]. Its only scaled competitor in insurance salvage is RB Global, parent of Insurance Auto Auctions (Competition holds the full named-rival record) [2]. RB Global earned $713.4 million of operating income on $4.59 billion of revenue in calendar 2025 — a 15.5% margin, less than half Copart's [3].

Copart FY (ended July 31); RB Global fiscal year (ended December 31). RB Global blends IAA salvage with Ritchie Bros. commercial-asset auctioneering, so this is a company-level, not a salvage-only, comparison. Sources: Copart FY2025 10-K, Consolidated Statements of Income [4]; RB Global FY2025 10-K, Statements of Operations [5].

The margin gap is a mix, not a like-for-like: RB Global carries a large commercial-asset auction business alongside IAA. But the direction is unambiguous, and it holds where it matters most — in the price the auction delivers to the seller. Management states that Copart's average selling prices grew faster than the Manheim Used Vehicle Value Index and "more than fivefold that of service providers similar to ours" [6]. That is the moat expressed as output: the platform that clears the highest price wins the seller, because the seller is paid on the price.

Returns tell the same story once the balance sheet is read correctly. Copart's reported return on capital employed was about 18% in FY2025, down from roughly 28% in FY2022. That decline is a balance-sheet artifact, not erosion of the moat. Once the $4.79 billion of cash and Treasury bills — idle on a business that runs on negative working capital [7] — is set aside, the operating business earns roughly 30% on the capital actually tied up in yards and receivables: $1.70 billion of operating income, taxed at 18%, over about $4.6 billion of net operating assets. The reported figure is dragged down by the denominator, not by any weakening of the moat. What to do with that cash is a capital-allocation question the report takes up separately (Business flagged it); it is not evidence of decay here.

Auction liquidity: the network that sets the price

The engine underneath the price is liquidity — the depth of the buyer pool bidding on each lot. Copart maintains a database of roughly one million registered members [8], of which about 300,000 are paying, active members drawn from virtually every non-sanctioned country [9].

International buyers account for ~40% of vehicles sold at Copart's U.S. auctions and roughly half of proceeds; in FY2026 they paid ~38% more per vehicle than comparable U.S. buyers. Sources: Q4 FY2025 call [10]; Q1 FY2026 call [11].

Three features make this pool hard to copy. It is global: international members buy about 40% of the vehicles sold at U.S. auctions and, because they take the more valuable cars, close to half of the proceeds, paying roughly 38% more per vehicle than domestic buyers — the demand side of that mix is taught in Industry [12] [13]. It is fragmented: the top ten individual buyers together take only a low single-digit percentage of all vehicles sold, so no handful of whales can be poached to hollow out demand [14]. And it compounds: management reports that unique bidders per auction have risen to all-time highs since 2022, and that a strong supermajority of insurance units now sell on a pure-sale basis rather than with reserve prices — a sign consignors trust the platform to find the price on its own [15].

The head start behind this is structural. Copart has been an exclusively online auction since 2003 — nearly two decades before its competitors moved online, and they did so only when COVID-19 forced the change [16]. A rival can build a website; it cannot instantly assemble the two decades of global buyer registration, licensing, and repeat behavior that give a single lot its depth of bidding.

Land, permits, and the physical barrier

Liquidity needs somewhere to put the cars. Copart carries $3.6 billion of net property and equipment — the yards where salvage is stored, inspected, and staged for auction [17]. The barrier is not the money; it is the permission. The company itself notes that new yards require land that is appropriately zoned, and that agreements to buy facilities can be blocked by "zoning restrictions or difficulties obtaining and maintaining use permits" [18]. It goes further in its risk disclosures: "public opposition in some communities to different aspects of our business operations has impacted our ability to obtain required business use permits" [19].

That is a moat and a constraint in one sentence. A well-funded entrant cannot buy its way to a national salvage-yard network quickly, because acres near major highways that a town will zone for wrecked cars are scarce — which is precisely why the same friction slows Copart's own expansion. The barrier that keeps competitors out also caps how fast the incumbent can add capacity, and shows up as periodic yard-fill during catastrophe events. The land moat is real; it is not free of cost to its owner.

Switching costs: booked net, wired into claims

On paper Copart looks exposed on the supply side. It obtains its vehicles from insurers under customizable seller agreements with no long-term lock-in [20]. The customer-concentration and contract-terms record sits in Competition. Management concedes that “the absence of long-term contractual commitments between us and our sellers… there can be no assurance that our competitors will not gain market share at our expense" [21]. Retention here is earned every quarter, not contracted.

What replaces the contract is workflow. Copart has been forward-integrating into functions the insurer used to run in-house: decision-support tools that help carriers call a total loss faster, advance-charge mitigation, and titling and loan-payoff services attached to each vehicle [22]. The more of a carrier's claims process runs through Copart's systems, the more disruptive it is to move volume to IAA — not because a contract forbids it, but because the plumbing would have to be re-laid. This is a genuine switching cost, but a softer one than a signed term; it holds as long as the price and service stay ahead. The clearest evidence it is working is the price itself: sellers keep the highest-clearing auction because that is where their proceeds are largest.

The live test: soft volume, and whether it is share

The moat's durability faces a real-time test on volume rather than margin. In the second quarter of FY2026, Copart's U.S. units fell 9.5% year over year and U.S. insurance units fell 10.7%; average selling prices rose 6% but could not fully offset the unit decline, and net income fell 9.5% [23].

"Adjusted" removes prior-year catastrophe units and the reclassification of low-value units to the direct-buy channel. Source: Q2 FY2026 earnings call, CFO remarks [24].

The headline overstates the weakness. Stripping out a hard prior-year catastrophe comparison and units Copart deliberately moved to a lower-cost "direct buy" channel, the underlying U.S. decline was about 4.5%, and insurance units about 4.8% — which management ties to lower claims frequency across the industry [25]. Assignments — the vehicles carriers actually hand Copart, the cleanest read on seller share — fell only low single digits [26]. On the evidence available, this reads as a claims-frequency cycle, not defection.

The honest limitation is that Copart's own disclosures cannot fully separate an industry claim-count decline from a loss of share to IAA. RB Global's total gross transaction value was roughly flat year over year, consistent with a soft industry rather than a competitor taking Copart's cars — but its salvage volumes are not cleanly isolated from its commercial-asset business in the data here. A like-for-like salvage-share reconciliation against IAA is the piece this chapter cannot close, and it is where a later chapter should go next.

The measured read: the moat is wide. It is defended by liquidity that compounds, land that cannot be quickly zoned, and workflow integration that raises the cost of leaving — and it is visible in a margin more than double the only scaled rival's and in prices that outrun the used-car index. The risk to that read is a durable decline in salvage assignments that turns out to be share rather than cycle, not a price war. What would change the read is simple and checkable: Copart's assignment volume falling for several consecutive quarters while IAA's salvage GTV rises. Until that shows up, the toll looks defensible.