The Undertaker's Margin
Somewhere in America right now, a car is dying. A hailstorm in Texas is cratering a Camry's hood. A teenager in Georgia is hydroplaning a Civic into a guardrail. A flooded retention pond in Houston is swallowing a dealership lot whole. And within seventy-two hours — sometimes faster — an insurance adjuster will declare each of these vehicles a total loss, and a tow truck will haul the carcass to a fenced lot owned by a company most people have never heard of, where it will sit on a patch of gravel until someone on the other side of the planet, hunched over a laptop in a port city in Nigeria or the UAE or Lithuania, bids on it through a proprietary online auction platform that processes over four million vehicles a year.
The company is Copart. Its market capitalization, as of mid-2025, hovers around $52 billion. It carries zero debt. It holds nearly $4.4 billion in cash. It has generated a compound annual return for shareholders exceeding 21% since its 1994 IPO — a run that places it in the same rarefied statistical neighborhood as Berkshire Hathaway, Costco, and Amazon over equivalent holding periods. Its operating margins routinely exceed 37%. It owns more than 10,000 acres of land across eleven countries. And it accomplishes all of this by performing what is, at its core, the most unglamorous task in the American automotive ecosystem: disposing of wrecked, flooded, stolen, and otherwise unwanted cars.
The paradox of Copart is the paradox of infrastructure itself. The business is invisible precisely because it works. Insurance carriers — State Farm, GEICO, Progressive, Allstate, the companies that collectively insure two hundred million American vehicles — need their totaled inventory to vanish quickly, predictably, and at the highest possible salvage price. Copart makes the wreckage disappear. And in building the logistics, the technology, the land portfolio, and the global buyer network to accomplish that disappearance at scale, it has assembled a competitive moat so wide and so strange that even the analysts who cover it sometimes struggle to articulate why no rational competitor would attempt to replicate it from scratch.
By the Numbers
The Copart Machine
$4.9BRevenue, FY2025 (est. annualized)
~$52BMarket capitalization (May 2025)
$4.4BCash on balance sheet, zero debt
4M+Vehicles sold annually
250+Locations across 11 countries
~40%North American market share
37%+Operating margin, trailing twelve months
21.7%CAGR since 1994 IPO
This is the story of how a Vietnam veteran with a high school diploma turned a single junkyard into the world's dominant salvage marketplace — and how the compounding logic of land, network effects, and
NIMBY politics created something that may be, pound for pound, one of the most structurally protected businesses in American capitalism.
Born Into the Wreckage
Willis Johnson did not stumble into the junk business. He was, in the most literal sense, raised in it. Born in 1947 on a dairy farm in Arkansas, Johnson grew up watching his father — a man who could not read — cycle through entrepreneurial ventures with the restless energy of someone who understood commerce instinctively but lacked the tools to systematize it. The family moved to California. The father acquired a small dismantling yard. Young Willis, still a teenager, learned to strip cars, sort parts, and negotiate with the kind of buyers who showed up at junkyards carrying cash and suspicion in roughly equal measure.
Then Vietnam. Johnson was drafted into the U.S. Army, served a one-year tour, was wounded badly enough to receive a Purple Heart, and returned to California with the specific combination of fatalism and urgency that combat veterans often carry into civilian enterprise. He moved his family into a trailer home. He scraped together enough money to buy a junkyard in Sacramento. The year was roughly 1972, and the American salvage industry looked almost exactly as it had for decades: fragmented, local, grimy, and profitable enough to sustain small operators but not interesting enough to attract institutional capital or managerial talent.
Johnson's first instinct was operational. Where competitors ran their yards as what they were — heaps of twisted metal behind barbed wire — Johnson organized his like a retail store. Clean aisles. Catalogued parts. Refurbished components displayed for easy inspection. He was, as one analyst would later put it, the first person in the junkyard business to understand that presentation affects price realization. He dismantled not just cars but individual parts, creating granular inventory where others saw bulk scrap. It was small-ball innovation, the kind that rarely gets written up in business school cases, but it worked. The yard grew. Johnson bought another. Then another.
The pivotal insight — the one that would ultimately generate tens of billions of dollars in equity value — arrived not through analysis but through adjacency. Johnson, already operating dismantling yards, noticed a related but distinct business: salvage auctions. Insurance companies, after declaring vehicles total losses, needed a mechanism to dispose of them. The existing system was clunky. Carriers would consign wrecks to local auction houses, which held periodic live auctions where a thin crowd of local buyers — dismantlers, rebuilders, small dealers — would bid on inventory they could physically inspect. The process was slow, the buyer pools were shallow, and price realization was mediocre. Johnson saw the gap.
In 1982, he founded Copart. The name — an amalgam whose precise etymology Johnson has been characteristically vague about — would come to define an entirely new model for salvage disposition. But in those early years, it was just another small auction operation in Vallejo, California, distinguished mainly by its founder's relentless energy and his willingness to drive personally to insurance offices, building relationships one adjuster at a time.
I know what I don't know. I also think it's a good idea to learn as much as you can.
The Incentive Inversion
What made Copart different — and what would eventually make it dominant — was not a single technological breakthrough but a structural innovation in how the middleman got paid.
The traditional salvage auction model worked on flat fees. An auction house would charge the insurance carrier a fixed amount per vehicle — typically $50 to $175 — to store, market, and sell the wreck. The auction house's incentive was volume throughput, not price maximization. Whether a totaled Honda Accord sold for $800 or $2,400, the auctioneer's take was the same. This misalignment was so deeply embedded in the industry that most participants didn't even recognize it as a problem.
Johnson recognized it as an opportunity. Copart introduced what it called the "percentage-of-sale" model — a commission structure in which Copart's fee was tied directly to the final auction price. If Copart could get a higher price for the wreck, Copart earned more. The insurance carrier earned more. The incentives were aligned. This was, in principle, identical to the commission structures that real estate brokers and fine art auction houses had used for centuries. But in the salvage industry, it was revolutionary.
The second-order effects were profound. Under the flat-fee model, auction houses had no reason to invest in marketing, buyer development, or technology that might improve price realization. Under the percentage model, every dollar of incremental buyer demand flowed partially to Copart's bottom line. This created a powerful feedback loop: Copart invested in attracting more buyers and improving auction technology, which drove higher sale prices, which attracted more insurance consignors, which gave Copart more inventory to offer buyers, which attracted still more buyers. The flywheel was born — not as a metaphor, but as an economic mechanism with real, measurable cash flow consequences.
The flat-fee competitors, meanwhile, were trapped. Switching to percentage-based pricing would have required them to simultaneously invest in buyer development capabilities they didn't have, while accepting the short-term revenue uncertainty of a variable-fee model. Most chose to do neither.
The IPO and the Consolidation Machine
Copart went public on NASDAQ in March 1994. The timing was good. The American salvage auction industry was still radically fragmented — hundreds of small operators, many family-owned, most operating in a single city or region. Johnson saw consolidation as the next lever. Public-market capital gave him the currency to buy.
🔑
Copart's Consolidation Era
Key milestones in the rollup strategy
1982Willis Johnson founds Copart in Vallejo, California.
1994IPO on NASDAQ. Begins acquisition-driven national expansion.
1998Launches first internet-based bidding platform — VB1.
2003Transitions entirely to virtual bidding with VB2.
2004Reaches $400 million in annual revenue; 107 North American locations.
2007Enters the UK market through acquisition.
2010Jay Adair succeeds Willis Johnson as CEO.
2012Launches VB3, third-generation virtual auction platform.
The rollup was methodical. Johnson and his team would identify local salvage auction operators — often in secondary markets, often run by aging owners without succession plans — and acquire them at reasonable multiples. Then they would integrate the acquired yard into Copart's national platform, plug the local inventory into the broader buyer network, and watch as price realization climbed. The economics were straightforward: a wrecked car that might attract five bidders in a local live auction could attract fifty or five hundred bidders on a national (and eventually global) online platform. More bidders meant higher prices. Higher prices justified higher consignment volumes. And the whole cycle accelerated.
By 2004, Copart had over 100 locations across North America and $400 million in revenue. But the real transformation was just beginning.
The Internet as Structural Weapon
In 1998, Copart launched its first internet-based bidding system — VB1, short for Virtual Bidding. The name was modest. The implications were not.
Before VB1, even a national salvage auction network was limited by geography. Buyers had to attend live auctions in person, which meant that the effective buyer pool for any given vehicle was constrained to whoever could drive to the yard that day. VB1 began to dissolve that constraint. Registered buyers could now place bids remotely, asynchronously, from anywhere with an internet connection. It was crude by later standards — the interface was basic, the real-time interactivity limited — but it expanded the addressable buyer universe from local to national overnight.
The 2005 Forbes profile of Copart captured the moment with precision: by July 2004, Johnson had all 107 of his lots linked to second-generation IT. Registered buyers anywhere in the world could bid online, in real time. "This means no more schlepping to live auctions," Forbes noted. "Hunched over their home monitors, the trader can now view each vehicle on the calendar, scan repair invoices, call up past sales prices and trade with a tap of their finger."
VB2 followed, then VB3 — each iteration adding richer vehicle data, better imaging, more sophisticated bidding mechanics. The third-generation platform, VB3, became the industry standard: a proprietary auction system connecting vehicle consignors to approximately one million registered members in over 185 countries. The internationalization was the killer feature. A totaled Lexus in Dallas was no longer competing for the attention of a handful of Texas dismantlers. It was competing for the attention of exporters in Nigeria, rebuilders in the UAE, parts dealers in Eastern Europe. The global buyer base transformed salvage economics. Vehicles that might have sold for scrap value in a local auction now commanded prices that reflected their residual utility in markets where labor was cheap, parts were scarce, and environmental regulations were permissive.
Hunched over their home monitors, the trader can now view each vehicle on the calendar, scan repair invoices, call up past sales prices and trade with a tap of their finger — all while having more information available than ever before.
The irony is rich. Copart — a company whose core product is literal wreckage — became one of the earliest and most effective adopters of internet-enabled marketplace dynamics. While the tech industry was busy building auction platforms for collectibles (eBay), advertising (Google), and consumer goods (Amazon Marketplace), a junkyard operator in Fairfield, California was quietly building the same economic architecture for the least glamorous inventory category imaginable. And because the inventory was unglamorous, nobody from Silicon Valley bothered to compete.
The NIMBY Moat
Land. It always comes back to land.
The salvage auction business requires a physical footprint that is, by its nature, unpleasant. You need large, accessible lots — typically ten to fifty acres — located near population centers (because that's where cars get wrecked), with good highway access (because tow trucks need to get in and out quickly), and minimal residential adjacency (because nobody wants to live next to a field of crushed metal). You need zoning permits. You need environmental approvals. You need community acquiescence.
Getting all of these things simultaneously has become, over the past two decades, nearly impossible for a new entrant. This is the NIMBY moat — and it may be the single most important source of Copart's competitive advantage.
The mechanism works like this: municipalities have become increasingly hostile to industrial land uses, particularly those involving damaged vehicles, fluid leaks, and heavy truck traffic. Zoning boards reject new salvage yard applications. Environmental agencies impose escalating remediation requirements. Community groups organize against proposed facilities. The result is that the existing stock of permitted salvage auction land is essentially frozen. New supply cannot enter the market.
Copart understood this decades before its competitors. Beginning in the late 1990s and accelerating after 2010, the company began systematically purchasing the land beneath its auction yards rather than leasing it. This was expensive. It required massive capital outlays that depressed near-term returns on invested capital. But it was strategically brilliant.
Owned land serves three purposes for Copart. First, it eliminates lease renewal risk — a landlord cannot decide to repurpose a yard for higher-value development. Second, it appreciates over time, particularly as adjacent areas urbanize and the regulatory barriers to new salvage land intensify. Third, and most importantly, it creates a structural barrier to entry that no amount of technology, capital, or managerial talent can overcome. You cannot build a competing salvage auction network without land. And the land, effectively, is no longer available.
Copart now controls over 10,000 acres across its global network. The company's capital expenditure in recent years has ranged from $335 million to $640 million annually, with the majority directed toward land acquisition and facility development. As of mid-2025, the balance sheet carries $4.4 billion in cash, zero debt, and $1.3 billion available on a revolving credit facility — $5.6 billion in total liquidity, much of it earmarked for the patient, opportunistic accumulation of irreplaceable real estate.
Our conservative balance sheet, our net cash balance, equips us such that we can be patient even through a crisis. Our insurance clients have a long memory, and they know those things. They know that when land is available for us to acquire, so that we can preserve it for the industry's use for the next 50 years, that we'll do so gladly and proudly, despite it of course coming with big ticket prices as well.
Think about what this means for a hypothetical competitor. To build a national salvage auction network from scratch, you would need to identify, purchase, zone, permit, and develop hundreds of large industrial lots in major metropolitan areas across the United States — a process that would take years, cost billions, and face community opposition at virtually every site. By the time you finished, Copart would have spent those same years deepening its buyer network, refining its technology platform, and strengthening its insurance carrier relationships. The moat doesn't just exist. It widens with time.
The Duopoly That Isn't
For most of its modern history, the salvage auction industry has been a duopoly: Copart and IAA (Insurance Auto Auctions). Together, they control an estimated 80% or more of the North American salvage auction market, with Copart holding roughly 40% and IAA the remainder. The structure looks superficially like Visa and Mastercard, or Coca-Cola and Pepsi — two dominant players splitting a market. But the comparison is misleading, because the competitive dynamics are deeply asymmetric.
IAA's trajectory diverged from Copart's at several critical junctures. Where Copart was founder-led for decades, IAA cycled through corporate parents and private equity sponsors. Where Copart accumulated cash and bought land, IAA accumulated debt and leased facilities. Where Copart invested in proprietary technology early and aggressively, IAA was slower to digitize its auction platform. The result, by the 2020s, was a structural gap in financial strength, technology capability, and — critically — balance sheet resilience.
In 2023, IAA was acquired by Ritchie Bros. Auctioneers (now RB Global), a Canadian heavy-equipment auction company, in a deal that merged IAA into a larger but heavily leveraged enterprise. As of early 2025, the combined entity carried approximately $3.84 billion in net debt against roughly $1.22 billion in EBITDA — over three turns of leverage. Compare that to Copart's $4.4 billion in net cash. The disparity is not merely financial; it is strategic. In a business where insurance carriers need to trust their auction partner to perform through catastrophe events — hurricanes, hailstorms, pandemics — balance sheet strength is a competitive weapon. Copart's CEO Jeff Liaw has said as much explicitly.
The UK market offers a revealing parallel. In 2023, the UK
Competition and Markets Authority (CMA) investigated the salvage vehicle market and found significant concentration, with Copart and its main UK competitor (e2e) dominating. The CMA's final report, published in mid-2023, examined whether the duopoly structure was harming competition. The investigation itself was an acknowledgment of just how concentrated — and how structurally entrenched — the salvage auction market had become.
The Johnson-Adair Dynasty
Willis Johnson ran Copart with the instincts of a scrapper and the discipline of a capital allocator for nearly three decades. But in 2010, he handed the CEO title to Jay Adair — his son-in-law.
Adair had been with the company since the early 1990s, rising through operations with a deep understanding of both the physical logistics of salvage yards and the technology investments that were transforming them. Where Johnson was folksy, intuitive, and drawn to the deal, Adair was systematic, technology-oriented, and focused on building the platform that would scale Copart from a national business to a global one. The succession was seamless in a way that founder-to-next-generation transitions rarely are, in large part because Adair had been effectively co-running the company for years before the formal handoff.
Under Adair's leadership, Copart accelerated its international expansion — moving into Brazil, Germany, Finland, Spain, the UAE, Oman, and Bahrain — and invested heavily in VB3, the auction platform that connected the global buyer network. Revenue roughly tripled during his tenure. Margins expanded. The land-buying program intensified. And the stock compounded at rates that made Copart a quiet favorite among quality-oriented fund managers who appreciated its combination of growth, profitability, and capital discipline.
In 2024, Jeff Liaw — a former investment banker and long-time Copart executive who had been serving as co-CEO alongside Adair — became sole CEO. Liaw represents a new archetype for the company: financially sophisticated, fluent in the language of institutional investors, and oriented toward the next phase of growth, which will require both the continued expansion of the global marketplace and the intelligent deployment of Copart's enormous cash pile.
The Johnson family's continued involvement — Willis remains a significant shareholder and board presence — creates the alignment structure that long-term investors prize. Copart has never been run for quarterly earnings optimization. It has been run for multi-decade value creation, with the patience that comes from having the founder's wealth tied to the same stock that outside shareholders own.
The Secular Tailwinds No One Talks About
The bull case for Copart is often expressed in terms of its moat, its management quality, and its financial metrics. But the truly remarkable feature of the business is that the addressable market keeps growing — not because Copart is expanding into new categories, but because the underlying economics of automobile insurance keep pushing more vehicles into the salvage channel.
Three forces drive this secular expansion:
Total loss frequency is rising. Modern vehicles are increasingly expensive to repair. Advanced driver-assistance systems (ADAS), cameras, sensors, lidar units, aluminum body panels, and complex electronics mean that even moderate collisions can generate repair estimates that exceed the vehicle's market value. When the cost to repair exceeds the cost to replace, the insurer declares a total loss and consigns the vehicle to a salvage auction. In the early 2000s, total loss rates in the U.S. were roughly 10–12% of all insurance claims. By the mid-2020s, that figure has risen to approximately 20% or higher. Every percentage point increase in the total loss rate represents hundreds of thousands of additional vehicles flowing into the Copart ecosystem annually.
Vehicle complexity is accelerating. Electric vehicles compound this trend dramatically. An EV with a damaged battery pack is often uneconomical to repair — the battery alone can represent 30–40% of the vehicle's value. As EV penetration grows, salvage volumes should grow with it. Copart has already begun investing in EV-specific handling capabilities, including specialized storage protocols for lithium-ion battery risks.
Used car prices remain elevated. Higher residual values for used vehicles translate directly into higher salvage prices, which benefits Copart through its percentage-of-sale commission structure. The post-pandemic surge in used car prices provided a tailwind that has only partially normalized, and structural supply constraints in the new vehicle market continue to support elevated used values.
The counterargument — that autonomous vehicles will eventually reduce accident rates and shrink the salvage pool — deserves serious consideration but operates on a timeline measured in decades, not years. Level 4 and Level 5 autonomy remain commercially unproven for mass-market deployment. Even optimistic projections suggest that the current vehicle fleet will take fifteen to twenty years to turn over. And even fully autonomous vehicles will be subject to weather events, infrastructure failures, and the simple physics of objects colliding at speed. The salvage pool is not going away. It is, by almost every near-term measure, expanding.
Catastrophe as Business Model
There is something uncomfortable about a business that benefits from destruction. Copart does not cause accidents, hurricanes, or hailstorms. But it processes their aftermath, and its financial results are correlated — sometimes dramatically — with the frequency and severity of catastrophic events.
Hurricane season is Copart's surge-pricing moment. When a major storm floods tens of thousands of vehicles in a metropolitan area — as happened with Hurricane Harvey in Houston in 2017, or Hurricane Ian in Florida in 2022 — Copart's volume spikes. The company has developed specialized catastrophe response capabilities: rapid deployment of tow trucks, temporary storage expansion, expedited title processing, and dedicated insurance carrier coordination. The ability to handle these surges — to absorb 50,000 or 100,000 additional vehicles in a matter of weeks — requires the kind of operational depth, geographic coverage, and balance sheet strength that only Copart possesses at scale.
The catastrophe business is lumpy and unpredictable, which is precisely why it favors the incumbent with the deepest resources. A smaller competitor cannot maintain the excess capacity needed to handle a major hurricane; maintaining that capacity during normal periods is too expensive. Copart can, because its land portfolio and financial reserves function as a permanent call option on catastrophic volume. The hurricanes always come. The question is when, not whether.
This dynamic reinforces the insurance carrier relationships that are Copart's commercial backbone. Carriers need a salvage partner who will not buckle under catastrophe load. They have long memories, as Liaw has noted. During the COVID-19 pandemic, when driving activity collapsed and salvage volumes plummeted, Copart did not lay off staff, did not suspend capital expenditure, did not breach commitments to its carrier partners. That institutional reliability — purchased at the cost of short-term margin compression — is a form of competitive moat that does not appear on any balance sheet.
The Marketplace Geometry
Copart's value proposition is, at its core, a marketplace arbitrage. The company sits between two groups — insurance carriers who need to dispose of vehicles and global buyers who want to acquire them — and earns its margin by making the matching process more efficient than either side could achieve on its own.
The geometry of this marketplace is asymmetric in instructive ways. On the supply side, Copart has a relatively small number of very large customers. The top U.S. insurance carriers — State Farm, GEICO, Progressive, Allstate, Liberty Mutual, USAA — collectively generate the vast majority of total loss volume. These relationships are contractual, long-term, and deeply integrated into carrier workflows. Switching costs are high because the insurance adjuster's process — declaring a total loss, assigning the vehicle, coordinating pickup, processing title transfer — is woven into Copart's systems.
On the demand side, the buyer base is vast, fragmented, and international. Approximately one million registered members across more than 185 countries participate in Copart auctions. These buyers range from large-scale dismantlers and rebuilders to individual exporters and hobbyists. No single buyer represents a meaningful share of total demand. This asymmetry gives Copart bargaining leverage on the demand side — buyers need access to Copart's inventory far more than Copart needs any individual buyer — while the supply-side concentration creates sticky, high-value relationships that competitors cannot easily disrupt.
The marketplace also exhibits classic network effects, though of a specific and somewhat unusual kind. More buyers create higher prices, which attract more consignors. More consignors provide more diverse inventory, which attracts more buyers. But the network effects are reinforced by — and in some ways secondary to — the physical infrastructure. Unlike a pure software marketplace (Airbnb, Uber), Copart's network effects are anchored in atoms: the land, the tow trucks, the storage capacity, the title-processing infrastructure. This makes the flywheel harder to start but also harder to disrupt.
$4.4 Billion in Cash and Nowhere Obvious to Put It
The most interesting strategic question facing Copart in 2025 is not about its competitive position, which appears unassailable, or its growth trajectory, which remains favorable. It is about capital allocation.
Copart generates enormous free cash flow — trailing twelve-month operating cash flow recently hit an all-time high — and has accumulated $4.4 billion in cash on a balance sheet with zero debt. The company has never paid a dividend. It has been modest in its share repurchase activity. And while land acquisition remains a significant capital expenditure, it does not absorb the full torrent of cash the business produces.
The math is relentless. With limited options to deploy capital at rates of return comparable to the core business, Copart's cash pile is likely to grow by another $1.3 billion or more over the next twelve months. Within a few years, the company could be sitting on $7 billion or $8 billion in cash — an extraordinary figure for a business with a $52 billion market cap.
Management has framed the cash as a multi-purpose strategic asset: insurance against catastrophe, fuel for land acquisition, proof of financial stability for carrier partners, and optionality for transformative investments or acquisitions. All of this is true. But the tension between disciplined capital allocation and mounting cash accumulation is real. At some point, the opportunity cost of holding billions in low-yielding assets becomes a drag on per-share value creation.
Willis Johnson, the founder, was a deal-maker by temperament. He would have found uses for the capital. Jeff Liaw, the current CEO, is a more deliberate allocator. The question is whether deliberation becomes inertia — or whether Copart finds, in its next phase, the kind of transformative deployment opportunity that justifies the wait.
On May 22, 2025, Copart reported its fiscal third-quarter results. Revenue: $1.2 billion, up 7.5% year-over-year. Net income: $406.6 million. Nine-month revenue: $3.5 billion, up 11.2%. Nine-month net income: $1.2 billion. The numbers were good by any reasonable standard. The stock fell 10% the next day. Investors, apparently, wanted more.
The cash pile kept growing. The land kept appreciating. The wrecked cars kept arriving. Somewhere a tow truck was already on the road.
