Carvana Co.: A Deep-Dive Value Investing Analysis

Hunting ten-year five-baggers among great growth stocks — pressing the upside question: "Can it get much bigger?"

• QHow high is its market ceiling? Is it enlarging an existing pie, or creating an entirely new market?6/10

  A

  Bottom line up front: Carvana is grabbing share within a huge but mature existing pie, rather than creating an entirely new market. The height of the ceiling comes from "the U.S. used-car retail pool is inherently very large + its penetration today is still very low," not from a not-yet-existing demand it conjures out of nothing. For Baillie Gifford's LTGG "five-fold in ten years" judgment this cuts both ways: the runway is long enough, but the story is one of "share migration within an old industry" rather than a "new species," and the ceiling is constrained by how big a bite it can ultimately take and how much that bite costs.

  First, how big is the pie. The report cites Cox Automotive data quoted in the company's 2025 10-K: U.S. used-car retail transactions exceeded 37 million in 2024, while Carvana's market share that year was about 1.6%; Cox forecasts U.S. used-car retail sales of about 20.4 million units in 2026. In other words, in retail alone, Carvana's share today is still in the low single digits. The latest reading shows its share has risen to "slightly above 2%", which is the right direction, but the absolute level is still extremely low. This is precisely the core of the bull narrative: going from about 2% to 8%-10% does not require the market to grow, only continued grabbing of the incumbent dealers' existing share.

  But this is exactly what shows it is doing "stock replacement" rather than "incremental creation." Demand to buy and sell used cars has long existed; what Carvana changes is the transaction method (online selection, home delivery, car vending machines, at-home pickup), not whether the transaction happens. The report's judgment here is restrained: "demand is not the problem; competition and the profit pool are... this is not a declining industry, but it is by no means a naturally high-margin one either." In other words, the physical ceiling is very high (more than twenty million retail units a year), but the ceiling on the share it can actually pocket has to be fought for head-on against CarMax (which sold about 780,000 used cars and has 255+ stores in fiscal 2026), a vast group of independent dealers, and online traffic platforms, with every additional percentage point costing customer-acquisition and operating expense.

  The company itself also points its growth toward "enlarging the existing pie" rather than "opening up a new continent": the long-term target management reiterated in Q1 2026 is to "sell 3 million cars and reach a 13.5% adjusted EBITDA margin in 2030-2035." 3 million units is several times the current pace, but it still falls within the existing pool of U.S. used-car retail, relying on rising penetration rather than category creation. The report also makes clear it does "not fully accept" this path.

  My conclusion: the ceiling is high enough to support years of strong growth (its most solid bull anchor), but it belongs to the category of "enlarging an existing pie," not "creating an entirely new market." This means the upper limit is not unlimited imaginative space; it is jointly constrained by the pace of share migration, competitors' reactions, and whether unit economics can withstand the expansion. The length of growth is there, but the "exclusivity" and "irreversibility" of that growth are not strong.

  Jun 11, 2026
• QOver the next five years, can its revenue at least double? Is growth driven mainly by volume, price, or new businesses?6/10

  A

  Bottom line up front: doubling revenue over the next five years is very promising, and the growth engine is absolutely dominated by "volume," with price a secondary tailwind and new businesses (financing/ancillary products/wholesale platform) an amplifier but not the main force. This is the most solid link in Carvana's bull logic: its unit-sales growth is genuinely strong right now. But to be honest, the doubling rests on unit expansion, and whether that unit expansion is achieved "profitably" matters more than "whether it can double."

  First, how strong is the volume momentum. In Q1 2026 Carvana sold 187,393 retail vehicles, up 40% year over year, the sixth consecutive quarter of retail unit growth of 40% or more; total revenue that quarter was $6.432 billion, up 52% year over year. In the report's table, full-year 2025 revenue was $20.322 billion, corresponding to 49% growth over 2024. Even if the growth rate gradually falls back from 40%+, as long as the five-year CAGR stays around 15%, revenue can easily double: 15% compounded over five years is about 2.01x. From the current starting point of 40%+, doubling in five years is a relatively conservative assumption rather than an aggressive one.

  Breaking down the growth: volume is the main engine. The report notes that company revenue "depends heavily on website traffic, inventory selection, brand and marketing effectiveness, customer experience, and repeat referrals," and is at bottom a high-frequency-acquisition transactional business; its vertically integrated system (the report discloses that after acquiring ADESA it gained 56 locations, of which 16 are IRC-converted, with a combined annualized inspection-and-reconditioning capacity of about 1.5 million units) is exactly the capacity laid down to support larger volume. Management's goal of "selling 3 million cars in 2030-2035" is also entirely a sales-volume metric, confirming that the doubling relies mainly on selling more cars.

  Price is a tailwind but cannot be relied on. The report is clear: the rise in per-unit revenue in 2021 and Q1 2026 "is largely tied to the vehicle-price environment and model mix... this is not the same as Coca-Cola-style brand-driven price increases." Q1 2026 revenue growth (52%) exceeded unit growth (40%), with the difference coming from higher per-unit revenue; but this swings with the used-car price cycle and is not a structural price increase Carvana can drive on its own. Treating price as a pillar of the doubling is dangerous.

  New businesses are an amplifier, not a new curve. Among the four revenue links under the report's framework, besides retail vehicle sales, the wholesale platform (more than 1,006,600 transactions in 2025, wholesale revenue of $4.052 billion), gains on selling auto loans, and extended-warranty/GAP/insurance commissions (other revenue of $1.733 billion) all amplify per-unit economic value in step with retail volume, but most of them are parasitic on retail unit growth and are not independent new markets.

  My conclusion: the probability of doubling in five years is high, and the ranking of drivers is volume >> price ≈ ancillary-finance amplification. But note the report's warning: growth itself is not the problem; the problem is the risk that "the more it grows, the more cash it needs." In Q1 2026, as volume surged, inventory again became a major use of cash, and quarterly operating cash flow fell to $107 million. So the honest answer to this question is: doubling revenue is highly likely to hold, but "doubling with quality" (volume up, unit gross profit not collapsing, cash flow not eaten up by inventory) is the real test.

  Jun 11, 2026
• QFive years from now, what will take over as the next growth engine? Does this "second curve" exist today?4/10

  A

  Bottom line up front: honestly, Carvana today does not yet have a clear, independent "second curve" capable of taking over. Its growth five years from now is most likely still an extension of the main curve: continuing to grab share in U.S. used-car retail and thickening its existing finance and wholesale businesses in step with volume. The so-called "next engine" looks more like the depth of the same curve (margin improvement, ancillary-finance monetization, maxed-out capacity utilization) rather than a new species. For Baillie Gifford's requirement to "concentrate firepower on years 3-10," this is a clear weakness: its long-term imagination depends on existing businesses continuing to deliver, rather than on a new growth pole being incubated.

  First, how long is the main curve. This is its greatest confidence: the report cites Cox data that U.S. used-car retail exceeded 37 million transactions in 2024, with Carvana's share about 1.6% (the latest has risen to "slightly above 2%"). Share migration from 2% to 8%-10% is enough to support the core growth of the next five to ten years. In other words, it hardly needs a second curve to complete the doubling; the main curve itself is long enough. This is both a strength (ample runway) and the answer to this question (the one that takes over is actually still itself).

  So are there any budding new directions? There are three worth watching, but none yet constitutes an independent curve:

  First, deeper monetization of ancillary finance and products. Under the report's framework, "other revenue" reached $1.733 billion in 2025 (gains on loan sales, extended warranties, GAP, insurance commissions). This is an important margin lever, but it is parasitic on retail volume and is not an independent new market; enlarging it is at bottom still squeezing the main curve drier.

  Second, the independent growth of the ADESA wholesale platform. The report discloses that in 2025 the wholesale platform handled more than 1,006,600 transactions with wholesale gross profit of $179 million, and that acquiring ADESA brought 56 sites, 16 IRC-converted. In theory wholesale auctions could serve third-party dealers and form B2B platform revenue; but for now it is more a supporting facility for "feeding retail + exiting low-quality inventory," and its gross-profit scale ($179 million) is negligible relative to retail, far from taking over.

  Third, category extensions such as new cars/EVs. External reports mention Carvana is strategically expanding into EV and new-car sales, a potential category broadening; but the report's body does not list it as a core growth pillar, and new-car retail is a red ocean with thinner profit and more intense competition, so whether it can become a true second curve is highly doubtful. I lean toward treating it as "an adjacent extension of the main curve," not counting it as an independent engine.

  My conclusion: its "second curve does not truly exist today"; more precisely, its growth five years out is still the depth of the same curve. This differs from growth stocks with a clear new technological pole, like "space/innovative-drug platforms." The report's overall positioning ("an excellent-operator turnaround company" rather than "a high-certainty good business") also confirms it: what deserves respect about Carvana is that it rescued its main business from the trough back to profitability, but it has not stocked up a new engine that could step in once the main curve peaks. For long-term investors, this means not paying an extra premium for "second-curve imagination"; and the current price has already priced in the flawless continuation of the main curve.

  Jun 11, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?4/10

  A

  Bottom line up front: Carvana's core advantage is scaled operating capability + a vertically integrated logistics/reconditioning network, not a network effect or brand pricing power. This moat will most likely widen mildly with scale over the next three to five years, but it is far from "wide, deep, and durable." At bottom it is "a leading operating system," which capital and time can catch up to, not an unshakable economic fortress. The report scores the moat 2.5/5, and I agree with this on-the-weak-side characterization.

  First, where the real advantage lies. The report's judgment closest to a moat points to economies of scale + operating density: after acquiring ADESA it gained 56 locations, and by the end of 2025, 16 had completed IRC conversion, with existing IRCs and auction sites together providing annualized inspection-and-reconditioning capacity of about 1.5 million units; in 2025 the wholesale platform handled more than 1,006,600 transactions. This system brings "a larger inventory pool, closer geographic coverage, lower transport friction, faster turnover." This is the real reason it runs faster than rivals, and it strengthens as volume grows, which is the source of the "widening": Q1 2026 still maintained 40% retail unit growth, and site IRC conversion and rising capacity utilization will further dilute unit costs.

  It also has brand and experience differentiation: nationwide mindshare, online ordering, home delivery, car vending machines, proprietary logistics software, and same-day delivery in some markets. These genuinely save time versus a traditional dealer and will thicken slightly as user word of mouth accumulates.

  But the moat's boundaries are very clear, which means it will not "become very deep":

  First, weak network effect, low switching costs, limited pricing power. The report says bluntly that "consumers buying or selling a car are not locked in by accumulated historical data, and dealers can source and dispose of inventory across multiple channels"; the competitive dimensions are "transparency, convenience, price, selection, and vehicle condition," and rivals "do not need to fully replicate the car vending machine to compete." Buying a car is a low-frequency, high-value decision, and users are not locked in; this is a structural weakness that will not change within three to five years.

  Second, brand ≠ pricing power, and there are reputational cracks. The report notes that Carvana has historically drawn regulatory scrutiny in several states over title/registration delays, vehicle condition, and complaints, with Connecticut reaching a $1.5 million settlement in 2025. The strong rivals are not weak: CarMax has 255+ stores and in fiscal 2026 sold about 780,000 used cars and 540,000 wholesale vehicles, with a longer operating history and a more mature store network.

  Third, the moat has not been cycle-tested. The report stresses that Carvana's high margins of the past two years "likely include operating improvements, but may also be mixed with the inventory cycle, the purchasing environment, fixed-cost dilution, and a favorable loan-sale market window"; in 2021-2022 operating cash flow was -$2.594 billion and -$1.324 billion, and net income -$287 million and -$2.894 billion. It has proven it "can climb out of the trough," but has not proven it "can earn money steadily whether or not the trough returns."

  My conclusion: over the next three to five years the moat's direction is mild widening (scale/density effects strengthen with volume), but its width has a structural ceiling (weak network effect, low switching costs, no pricing power, strong rivals all around). The report's phrasing is most precise: "strong operations has not yet settled into a strong moat." For long-term investors, this means its lead is an "operating lead" rather than an "institutional monopoly," and the moat is not enough on its own to support the current high valuation premium; this is one of the foundations of the report's conclusion of "hold with caution, no margin of safety at present."

  Jun 11, 2026
• QIf its core business were disrupted, does it have the DNA to reinvent itself? How does it treat mistakes and bad news?5/10

  A

  Bottom line up front: Carvana has one very hard piece of evidence of "self-reinvention": it climbed out alive from a near-collapse, high-leverage/high-loss crisis in 2022-2023, a real crisis-reinvention DNA that deserves credit. But its way of handling mistakes and bad news still carries clear flaws: the roots of the crisis (aggressive leveraging, a huge acquisition, over-expansion) were created by management itself, and its most dangerous "disruption" scenario is actually not technological disruption but the funding chain and cycle reversal, which is precisely where it has stumbled in its history. So the answer is: reinvention capability has been proven, but self-discipline has not.

  First, the reinvention evidence it has actually shown. The report records it fully: facing a life-or-death moment, in 2023 the company swapped $5.5 billion in principal of unsecured debt for $4.2 billion in principal of secured debt plus $341 million in cash, producing an $878 million debt-restructuring gain in accounting terms; in 2024-2025 it again used ATM offerings to raise equity at high prices, applying the proceeds to reduce debt and strengthen liquidity. The result was a strong recovery from the trough: in the report's table, 2021-2022 operating cash flow was -$2.594 billion and -$1.324 billion, turning positive to $803 million in 2023 and reaching $1.036 billion in 2025 with operating profit of $1.881 billion. The report judges that this move of "repairing the balance sheet with overvalued stock and buying back high-cost debt with cash" is rational for existing shareholders. This is a genuine "reinvention from the cliff's edge," proving that management can make cool-headed financial self-rescue at the life-or-death moment, which is stronger than many growth companies that only know how to burn cash.

  On "the DNA to reinvent itself when the core business is disrupted" (the chained question's implicit premise): we must honestly distinguish two kinds of disruption. Technological disruption (say, some better way to buy cars replacing the online self-operated model): Carvana's proprietary logistics software, vertical integration, and aggressive pursuit of operating efficiency do show an ability to iterate processes quickly, and relative to traditional dealers it is itself the disruptor, which gives it a degree of adaptive resilience. But model/financial disruption (tightening of loan-sale and securitization capacity, reversal of the interest-rate and vehicle-price cycle) is its real Achilles' heel: the report repeatedly stresses that it is highly sensitive to "consumer affordability, interest rates, vehicle supply, loan-sale capacity, and capital-market sentiment," and the 2022 crisis was precisely the result of such shocks compounded by its own recklessness. Whether it can reinvent itself again in the next such shock has, for now, only the single sample of "it was rescued last time," which does not constitute a stable pattern.

  Now how it treats mistakes and bad news, where the flaws are more evident. The report notes that in 2022 the company spent about $2.2 billion to acquire ADESA and recognized an $847 million goodwill impairment that same year, "hard to call evidence of excellent capital allocation"; the crisis itself was self-inflicted "when the company was expanding most aggressively and the capital environment was worst." The governance structure also weakens external constraints on error-correction: the Garcia family firmly controls it through dual-class shares, with the family holding more than 85% of the voting power and Class B carrying ten votes per share, plus ongoing related-party transactions with DriveTime; this means outside shareholders can hardly impose checks when management errs. Facing bad regulatory news (title/registration/vehicle-condition problems in several states, Connecticut's $1.5 million settlement in 2025) the company chose to settle, a pragmatic handling, but these problems themselves reflect recurring execution cracks in its large-scale standardized fulfillment.

  My conclusion: reinvention DNA: present, and verified by one real crisis (a plus); but the culture of self-discipline and error-correction: doubtful. What it does best is "stay cool and self-rescue after causing trouble"; what reassures least is "repeatedly and actively creating risk for the sake of scale." The report's "triggers for re-evaluation" list has one item that hits the mark: "management again adds leverage to make big acquisitions for scale rather than per-share intrinsic value." In other words, what truly tests its reinvention DNA is not external disruption but whether it can restrain itself from repeating the 2022 mistake, and on this there is still not enough long-term evidence.

  Jun 11, 2026
• QDoes management (especially the founder) have long-term vision, with interests deeply tied to the company? Are they willing to sacrifice current profit for five to ten years out?6/10

  A

  Bottom line up front: the founding Garcia family does have long-term vision, with interests deeply tied to the company, and is clearly willing to sacrifice current profit for years down the road; from this pure angle of "long-termism + founder present," Carvana is actually a relative strength in the Baillie Gifford framework. But it must come with an equally important qualifier: this binding is a "control-style binding" achieved through dual-class shares and related-party transactions, and for ordinary outside shareholders, the founder's interests are often aligned in direction with yours, but the distribution mechanism is not friendly. So the honest answer here is: deep binding, long vision (credit); but the governance structure means the benefits of this binding do not necessarily flow fairly to minority shareholders (a deduction).

  First, the evidence of binding and long-term vision (a strength). The founder father and son's control of the company is textbook-level deep binding: the report discloses that as of March 10, 2026, Ernest Garcia II held about 50% of the voting power and Ernest Garcia III about 33%, with the family holding more than 85% of the vote combined and Class B carrying ten votes per share when conditions are met. Ernest Garcia III is the founder and CEO, long in his seat, with his personal wealth highly tied to the company: exactly the type of "founder betting his own fate on the company" that Baillie Gifford favors.

  Is it willing to sacrifice current profit for the long term? The answer is a clear yes, with behavioral evidence. The most powerful is that management publicly anchors a ten-year target: it reiterated in Q1 2026 the goal to "sell 3 million cars and reach a 13.5% adjusted EBITDA margin in 2030-2035", and it keeps investing cash into long-cycle infrastructure such as reconditioning, logistics, title/registration, and ADESA integration. The report also records its long-term orientation during the crisis: it would rather issue dilutive equity at the trough and use the proceeds to cut debt and strengthen liquidity in order to preserve long-term survival, a real choice of "bearing current cost to live to year 10." In Q1 2026, even as volume surged and quarterly operating cash flow fell to $107 million, the company still prioritized investing in capacity rather than dressing up short-term profit, a long-term direction.

  But the governance structure brings three concerns that must cost points (a core reservation the report repeatedly stresses and I agree with):

  First, the absolutization of control weakens accountability to minority shareholders. Dual-class shares make it nearly impossible for outside shareholders to impose checks when management errs, a lesson already paid for in the history of 2022's aggressive leveraging + the roughly $2.2 billion ADESA acquisition + the same year's $847 million goodwill impairment.

  Second, related-party transactions dilute the "binding". The company has related-party arrangements with DriveTime covering leases, services, and the wholesale platform; DriveTime is controlled by the Garcia family, and the related transactions cannot be assumed to be at arm's length. With the founder standing on both sides of the deal, the binding of interests is no longer purely "aligned with minority shareholders."

  Third, the TRA and ongoing dilution tilt economic interests toward the controlling party. The report discloses that at year-end 2025 the TRA liability was as high as $2.228 billion, of which $1.721 billion belonged to related parties; in 2023 it also raised the authorized share count under the 2017 incentive plan from 16.1207 million shares to 36.1207 million shares. This means that although the founder is tied to the company's fate, through the TRA and Up-C structure a large slice of future tax benefits and economic interests goes preferentially to LLC-unit holders/related parties rather than to ordinary Class A shareholders.

  My conclusion: long-term vision: strong; founder's interests tied to the company: extremely deep; willing to sacrifice current profit for the long term: yes, evidenced by the ten-year target and crisis-period behavior (these three are among Carvana's few clear pluses in the Baillie Gifford framework). But "tied to the company" does not equal "tied to you the minority shareholder": dual-class shares, related-party transactions, and the TRA determine that the benefits are not necessarily fairly distributed. In the report's words: "even if the operating team is capable, that does not guarantee governance is friendly." For long-term investors, this is a combination of "the founder is very right, but the rules for splitting the pie are unfavorable to you."

  Jun 11, 2026
• QIf it disappeared tomorrow, how much would customers miss it? Is its growth sustainable and not reliant on harming society or regulation?4/10

  A

  Bottom line up front: if Carvana disappeared tomorrow, customers would face some inconvenience, but would not miss it much: what it offers is "a more hassle-free car-buying/selling experience," not an irreplaceable necessity, and users could quickly turn to CarMax, dealers, or other online platforms. On the "social/regulatory sustainability" dimension, its growth method is basically not harmful to society, but it does carry recurring regulatory friction (title/registration, vehicle condition, consumer complaints), not zero hidden risk. On balance: indispensability is on the weak side, and social sustainability is neutral-to-manageable. To Baillie Gifford's probing question "can customers not do without it," Carvana cannot give a strong answer.

  First, "indispensability" (the chained question's first layer). The report's judgment is direct: buying and selling cars is a low-switching-cost, weakly-locked-in transaction. It writes that "consumers buying or selling a car are not locked in by accumulated historical data," that customers are "not locked in by long-term contracts, and revenue is repeatable but not recurring," and that at bottom it is "a high-frequency-acquisition, low-switching-cost transactional business." This means what Carvana gives users is an experience increment (online selection, home delivery, car vending machines, trade-in, at-home pickup, same-day delivery in some markets), not "you cannot get it done without it." Once it disappeared, substitutes are ready and strong: CarMax has 255+ stores and sold about 780,000 used cars in fiscal 2026, plus a huge number of independent dealers and other online/auction platforms. The report sums up the competitive dimensions as "transparency, convenience, price, selection, and vehicle condition," all of which rivals can provide, none of which Carvana holds exclusively. So the degree to which "customers would miss it" is low to medium. It makes transactions more comfortable, but it is not the sole channel for a life necessity.

  The fair credit to give is: on the convenience experience it does have real differentiated stickiness. For users already used to buying cars online, going back to the traditional process of visiting stores and haggling is an experience downgrade; repeat referrals and word of mouth also show some users approve. But this is "likes to use" rather than "cannot do without," far below in magnitude the indispensability of search, operating systems, or payment networks where "society would seize up if they disappeared."

  Now "social and regulatory sustainability" (the chained question's second layer). This one needs an honest two-sided assessment:

  On the positive side, its core business does not harm society: it makes used-car transactions more transparent and convenient, standardizes vehicle inspection, reconditioning, and logistics, and overall improves market efficiency, with no structural problem of "growing by harming users or society." Demand truly exists over the long term (the report cites Cox data that U.S. used-car retail exceeded 37 million transactions in 2024), and what it does is upgrade the experience of a stock market, not manufacture false demand.

  But the negative is real regulatory friction: the report records that Carvana has previously faced regulatory pressure in several states over title and registration processing, temporary tags, and vehicle-condition descriptions, with Connecticut reaching a $1.5 million settlement in 2025. The report specifically warns: because the model relies heavily on "nationwide, standardized, large-scale fulfillment," a slip in execution standards would "amplify reputational damage faster than for a traditional small dealer." In addition, in 2026 there were short-seller challenges to its financial presentation and related-party relationships. None of these is "the growth method itself harming society"; rather it is "compliance-execution cracks under large-scale rapid expansion," a manageable risk but one that needs continued investment (the company is indeed increasing title/registration investment) to keep down.

  My conclusion: indispensability: on the weak side (if it disappeared customers would be inconvenienced but have ample substitutes, hardly missing it much); social/regulatory sustainability: neutral-to-manageable (the business itself benefits society, but it comes with recurring compliance friction that needs ongoing governance, not zero hidden risk). Combining the two layers exactly confirms the report's on-the-weak-side judgment of only 2.5/5 for the moat: a business whose customers "like to use it but can easily replace it" and which needs to keep handling regulatory friction lacks the deep "cannot do without it" stickiness Baillie Gifford values most.

  Jun 11, 2026
• QWhat are the unit economics of this business (gross margin, incremental returns)? Do they improve or worsen as it scales? Where does the money it earns go?4/10

  A

  Bottom line up front: Carvana's unit economics have indeed improved markedly with scale over the past two years, its most persuasive bull evidence; but it is inherently a thin-margin retail/turnover business (not the high-margin leverage of software SaaS), and the improvement comes mainly from scale diluting fixed costs and operating density rather than from structurally high margins. The money earned is currently used first to repair the balance sheet (cut debt) + invest in capacity (reconditioning/logistics/ADESA) rather than dividends or buybacks. The honest conclusion: the unit-economics improvement is real and the scale effect is real, but its "good" is "going from very bad to decent," and the sustainability of margins has not yet been verified across a complete cycle.

  First, the relationship between gross profit and scale (a strength). In the report's table, gross profit recovered from $1.929 billion in 2021 and $1.246 billion in 2022 (the crisis trough) to $2.876 billion in 2024 and $4.192 billion in 2025; Q1 2026 gross profit was $1.271 billion. The operating margin moved from a huge loss to 2025 GAAP operating profit of $1.881 billion and an operating margin of 9.3%. What better reflects "scale-driven profitability" is EBITDA leverage: full-year 2025 adjusted EBITDA reached $2.237 billion with an 11.0% margin, and Q1 2026 adjusted EBITDA was $672 million with a 10.4% margin. Maintaining a double-digit EBITDA margin while unit volume grew +40% for six consecutive quarters shows that incremental units are indeed contributing positive operating leverage; this is the very evidence of "unit economics improving as scale grows," and is where Carvana is genuinely strong.

  But three qualifiers must be honestly flagged, to avoid misreading it as a high-quality compounding machine:

  First, the absolute gross margin is retail-grade, not software-grade. 2025 total revenue was $20.322 billion and gross profit $4.192 billion, a gross margin of about 20.6%; this is a physical retail business that requires real money spent on inspection, reconditioning, transport, and inventory, with a ceiling far below the 70%-margin of software. The report repeatedly stresses that it is "not a pure software platform; at its core it is still a highly operational auto retail and turnover system with a financial component." Its unit-economics improvement is "efficiency leadership among peers," not "a naturally high-margin industry."

  Second, the current high margin may contain cyclical tailwinds. The report warns: the high margins of the past two years "likely include operating improvements, but may also be mixed with the inventory cycle, the purchasing environment, fixed-cost dilution, and a favorable loan-sale market window"; and it makes clear that "if unit gross profit and SG&A leverage reverse at the same time, the profit swing would be large." Q1 2026's EBITDA margin (10.4%) is already slightly below full-year 2025 (11.0%), and management itself admits there are margin headwinds. So the "bigger is better" trend holds, but it is not monotonic or risk-free.

  Third, ancillary finance amplifies per-unit economics but also increases reliance on external funding. Under the report's framework, "other revenue" (gains on loan sales, extended warranties, GAP, insurance) reached $1.733 billion in 2025, an important amplifier of per-unit value; but this ties unit economics to "loan-sale/securitization capacity," and the report lists "tightening loan-sale capacity" as one of the signals that would overturn the investment logic.

  Where does the money earned go? The report gives a clear picture: debt reduction + capacity investment first, rather than returning capital to shareholders. After the crisis: in 2023 it used debt swaps + cash to redeem high-cost debt, and in 2024-2025 it raised equity to cut debt and repurchased/redeemed a fair amount of 2028 secured notes; meanwhile it kept investing cash into reconditioning, logistics, title/registration, and ADESA integration (management laying capacity for the 2030-2035 target of 3 million units/13.5% EBITDA margin). What to watch is the report's point that "stock-based compensation is a real shareholder cost": 2025 stock compensation was about $96 million and ongoing dilution is a real price; plus the $2.228 billion TRA liability (of which $1.721 billion belongs to related parties) will swallow, over years to come, part of the tax benefit that should belong to public shareholders. In other words, where the money goes is rational for long-term survival (cut debt, invest in capacity), but a portion leaks to the controlling party through dilution and the TRA.

  My conclusion: unit economics: substantial improvement in recent years, with a real scale effect (a clear plus); but the magnitude is retail-grade thin margin, and the high margin contains components not cycle-tested (a qualifier); the money earned: rationally directed to debt reduction and capacity, but a portion is siphoned off by dilution and the TRA. The report's conservative Owner Earnings estimate (a range of about $550-700 million, midpoint about $640 million, far below the $1.895 billion GAAP net income) is exactly the reminder: even as unit economics improve, the cash truly available for free distribution to shareholders is much less than the income statement suggests.

  Jun 11, 2026
• QFor it to rise fivefold in ten years, what conditions must hold simultaneously? Are these conditions realistic? What expectations does today's share price imply?3/10

  A

  Bottom line up front: for Carvana to rise fivefold in ten years from the current price (about $68/share post-split, market cap about $74 billion), a string of near-perfect conditions must hold simultaneously; and today's share price has already priced in most of "sustained high growth of the main curve + high margins maintained + a funding chain that stays smooth throughout." The report's valuation conclusion is very clear: the current price is about 113%-210% above base-case intrinsic value and still about 24%-80% above the bull range, with "no margin of safety." So the honest answer here is: fivefold in ten years is not impossible, but the conditions it requires have a low combined probability, and today's price already implies the expectation that "most of these good things will happen"; you are not betting on improvement at a low level but paying at a high level for "flawless continuation."

  For fivefold in ten years, the following conditions must hold simultaneously (the chained question's first layer):

  First, sales keep expanding rapidly to approach management's vision. Management's target is to sell 3 million units in 2030-2035 at a 13.5% adjusted EBITDA margin. Current annualized retail volume is on the order of 700,000-800,000 units (Q1 2026 alone was 187,393 units); reaching 3 million units is a several-fold jump, requiring share to be grabbed from "slightly above 2%" to near double digits without being countered by rivals.

  Second, unit gross profit is not eaten by competition, and SG&A/vehicle keeps falling. The report lists this as a key assumption and warns that "if unit gross profit and SG&A leverage reverse at the same time, the profit swing would be large." Fivefold in ten years requires margins not just to hold at larger scale but to move toward a 13.5% EBITDA margin; yet Q1 2026's EBITDA margin (10.4%) is already slightly below full-year 2025 (11.0%), with margin headwinds present.

  Third, loan-sale/securitization and other financing capacity does not break for ten years. The report repeatedly stresses its high sensitivity to "loan-sale capacity, interest rates, and capital-market sentiment," and lists "funding tightening" as a signal that would overturn the logic. Over ten years the U.S. must not see a single cyclical shock strong enough to cut off its funding chain, which is a very strong assumption.

  Fourth, governance and regulation no longer become a major drag. Dual-class shares, DriveTime related-party transactions, and the $2.228 billion TRA liability ($1.721 billion belonging to related parties) must not further erode common shareholders' economic interests, and regulatory friction must not recur on a large scale.

  Fifth, the valuation multiple does not contract sharply. This is the most overlooked one: even if all four above are achieved, if the multiple the market assigns reverts to normal from its current high, a five-fold share price cannot be realized; and the current multiple is already at a high level (see below).

  Are these conditions realistic? My judgment: each on its own is not outlandish, but the combined probability of all five holding continuously for ten years is low. The report's characterization is incisive: this is a "big market, poor industry, where the strong executor may win" track, with a moat of only 2.5/5, no pricing power, and no complete-cycle test. Multiply the five "must all be right" together and the margin of safety vanishes.

  What expectations does today's price imply (the chained question's second layer)? This is the most crucial part of the whole question, and the easiest to be misled by a "cheap P/E." The report punctures this trap: the "about 7.9x P/E" shown on market terminals is a false impression caused by a mismatch between the post-split price and pre-split EPS/share-count conventions; restating $68.28 to the pre-split equivalent of about $341.40 and comparing it with $8.63 EPS, the true historical P/E is close to 39.6x. I independently verified this: stockanalysis.com shows CVNA with a market cap of about $74.2 billion and a P/E of about 39, consistent with the report's framework, not 7.9x. A stricter framing is the report's owner-earnings view: using the conservative Owner Earnings midpoint of about $640 million as the denominator, the multiple on a $74 billion market cap is much higher. In other words, the expectation today's price implies is "high growth + high margins + smooth financing continuing flawlessly for years": the ten-year annualized return the report calculates from the current price is conservative scenario -10%~-5%, base -2%~+2%, and only +5% to +8% in the bull case, which already shows the price has exhausted the upside.

  My conclusion: fivefold in ten years requires five demanding conditions to hold simultaneously, with a low combined probability; and today's price of about $68/a $74 billion market cap has already priced in most of these good things (true P/E about 39x, by no means the misleading 7.9x). The report's fair buy range is $15-22 (post-split), and the current price is far above it. The essence of Baillie Gifford-style fivefold-in-ten-years is "buying a great growth stock cheap when the market does not get it"; but Carvana right now is a case where "the market already gets it and has fully priced it," so what you buy is high-priced flawless expectations, not an undervalued growth option. This is the core of the report's "Watch, no margin of safety" conclusion.

  Jun 11, 2026
• QWhy has the market not yet realized all this? Does it not understand, not respect, or not see far enough? What would become the "narrative turning point"?3/10

  A

  Bottom line up front: Baillie Gifford's standard framing of this question is "why has the market not yet realized this stock's greatness"; but for Carvana, the honest answer must be flipped: the market has actually quite fully realized its merits, assigning it a high valuation of about $74 billion market cap and a true P/E of about 39x. What is truly "not yet fully digested by the market" is instead its fragile side: the "cheap illusion" created by the convention mismatch, the profit distorted by tax/TRA, and the deep reliance on the funding chain and the cycle. So this is not undervaluation caused by "not understanding/not respecting/not seeing far," but closer to overvaluation where "some investors are led by misleading metrics and a strong volume narrative and underprice the risk." The answer to this question for Carvana is exactly a mirror of the report's "no margin of safety" conclusion.

  Let me take apart Baillie Gifford's three kinds of "the market has not realized" one by one:

  "Not understanding": partly true, but in the opposite direction. There is indeed one thing the market widely gets wrong, but the direction of the error is "thinking it is cheap" rather than "overlooking its greatness." The report notes that the "about 7.9x P/E" shown on terminals is a product of the mismatch between the post-split price and pre-split EPS/share-count conventions; the true historical P/E is close to 39.6x. I independently verified this: stockanalysis.com shows CVNA with a market cap of about $74.2 billion and a P/E of about 39, confirming the report's framework. Moreover, under the Up-C structure Carvana Co. holds only about 64.9% of the economic interest in Carvana Group, and 2025 net income is severely distorted by a $2.785 billion tax benefit and a $2.2 billion TRA expense, all of which is "very high accounting noise, easy for a layperson to misread" (the report's words). So "not understanding" is real, but it makes some people overestimate how cheap it is rather than underestimate its value.

  "Not respecting": basically untrue. The market is far from slighting it. CVNA carries a Zacks Strong Buy rating and is sought after by analysts on the back of the split and its expansion into EVs/new cars; its valuation multiple is at a high level, and six consecutive quarters of 40% volume growth are enthusiastically priced by the market. This is not a neglected, disdained stock but a hot growth stock.

  "Not seeing far": present in both directions, but not a reason for undervaluation. Bulls are indeed "seeing far," pricing in ahead of time the vision of 3 million units and a 13.5% EBITDA margin in 2030-2035; this is precisely the over-optimism of "seeing too far and treating the vision as certainty" rather than the undervaluation of "failing to see long-term value." Bears, meanwhile, fix on the structural weaknesses the report repeatedly stresses: a weak moat (2.5/5), no pricing power, no complete-cycle test, high sensitivity to the funding chain, and governance tilted toward the controlling shareholder. The report clearly records that "in 2026 there were indeed short-seller challenges to its financial presentation and related-party relationships," which shows the disagreement is large, but the focus of the disagreement is risk, not "no one has discovered it is valuable."

  On "what would become the narrative turning point" (the chained question's implicit premise): this is the most actionable part of the question. I think the turning point lies not in "the market suddenly discovering its greatness one day" (it is already fully priced) but in the concrete events where the current flawless narrative is falsified or validated; the report's "signals that would trigger re-evaluation" gives a clear list:

  Downward narrative turning points (more likely, more dangerous): first, "volume up but cash flow not up" appearing for several consecutive quarters: Q1 2026 volume was +40% but quarterly operating cash flow already fell to $107 million, and if inventory keeps eating cash, the market will re-rate it as a "high-volatility used-car retailer" rather than a "high-momentum growth stock"; second, unit gross profit falls and SG&A/vehicle no longer improves, proving the margin is not a structural advantage (Q1 EBITDA margin already shows margin headwinds); third, loan-sale/securitization capacity tightens, exposing its reliance on external funding; fourth, regulatory events flare up again or short-seller challenges are borne out; fifth, the convention mismatch is widely corrected by the market: as more people realize the true P/E is about 39x rather than 7.9x, the support for "cheap" will loosen.

  Upward narrative turning points (possible, but already partly priced): volume keeps growing at 40% and the EBITDA margin moves steadily toward 13.5%, cash flow amplifies with scale rather than being eaten by inventory, and the governance or TRA burden eases substantially; these would strengthen the bulls, but because the current price already implies flawless continuation, the upside is suppressed by valuation (the report calculates that at the current price the ten-year bull-scenario annualized return is only +5% to +8%).

  My conclusion: the market is not "failing to realize Carvana's greatness"; rather it "has fully or even excessively paid for its merits and underpriced its risk." What is truly not fully digested is the fragile side, not the value. The narrative turning point most likely comes from "the flawless expectation being broken by one of cash flow/margin/financing/regulation" rather than "value being rediscovered." This is fully consistent with the report's overall judgment: a company worth respecting that genuinely makes money, paired with a price that prices in flawless execution far into the future; for long-term value investors, the correct posture is to track patiently and refuse to chase a hot narrative on price, rather than betting on an "undervaluation repair" that does not exist.

  Jun 11, 2026