CarMax, Inc. (NYSE: KMX) — A Good Operator Renting a Bad Neighborhood on a Subprime Loan

Date: June 7, 2026 Price reference: ~$47.15 (June 5, 2026) · Market cap ~$5.43B · ~141.8M shares · No dividend Fiscal year: Ends end-February (FY2026 = year ended Feb 28, 2026) Sector / industry: Consumer Discretionary — Automotive Retail (used-vehicle retail + captive auto finance)

This article carries NO buy/sell recommendation and NO price target in its analytical body (Sections 1–15). Valuation is discussed only as embedded expectations and scenarios. The single, deliberate exception is the “Author’s Take” block immediately below, which is clearly fenced as a subjective view.

⚡ Author’s Take

This block is the author’s own independent, subjective opinion. It is general information, not investment advice. The analytical body that follows (Sections 1–15) takes no position and contains no price target outside this block.

Verdict: AVOID at ~$47 — a quality operator in a structurally bad industry, with its earnings engine quietly migrating into subprime credit at a 32-year delinquency high. Not a high-conviction short (cheap on book, off-lease supply recovery and an 11.6% short interest can fuel a cyclical bounce). Accumulate-on-weakness only toward book value — roughly the $30–36 zone (~0.75–0.9x tangible book; ~11–13x trough-normalized EPS). Fair ~$45–52; rich above ~$60.

CarMax is the best-run retailer in a commoditizing, hyper-fragmented, cyclical business — and that is precisely the problem. After thirty years it holds only ~3.6% of its addressable used-car segment and that share is now falling, while Carvana grew units 43% last year and is building 2.5x its current capacity. The market is being asked to pay ~$47 for normalization — that the $141M Edmunds impairment and restructuring don’t recur, that used-unit volume inflects positive off the 2025–27 off-lease supply trough, that SG&A re-leverages, and that CarMax Auto Finance (CAF) holds ~$560M of income. The trouble is what sits underneath: CAF generates roughly 70–90% of normalized pre-tax profit, and management is pushing it down-spectrum (Tier-2 originations targeted to roughly double) just as net charge-offs have climbed from 1.93% to 2.34% and recovery rates have fallen from 53% to 45%. You are buying a low-margin retailer whose profit is really a leveraged subprime-tilting loan book, late in the consumer-credit cycle. The framing here is value-trap-risk, not deep-value: in the bear case the multiple and the earnings de-rate together. Capital allocation reinforces the caution — they bought back stock at $125 in FY22, then paused at ~$40 in early FY26 on a leverage technicality; they impaired their only acquisition; and the board’s answer was to fire a long-tenured insider CEO and import a hotel executive with no auto or credit background.

What keeps me off the short side: the off-lease supply trough genuinely reverses through 2027 (a real volume tailwind), the stock is cheap on book (1.14x, 7th percentile of its own history) and sales, the recourse balance sheet is sound (~$2.2B corporate debt; the scary $18B is non-recourse against the loan book), and a heavy short base plus a new-CEO strategy reveal (June 2026) are squeeze catalysts. Conviction: medium. Flips bullish if used-unit comps inflect durably positive while CAF credit metrics stabilize and share stops bleeding to Carvana. Flips bearish if CAF net charge-offs accelerate through allowance coverage as the down-spectrum vintages season, forcing provision spikes in the segment that carries the whole company.

One-liner: “A good operator renting a bad neighborhood — financed by a subprime loan, and the rent is rising.”

1. Executive Summary

CarMax is the largest used-vehicle retailer in the United States — 780,684 retail used units, 538,203 wholesale units, and 256 stores in FY2026 — paired with a consolidated captive lender, CarMax Auto Finance (CAF), which carries a ~$16.3B auto-loan book. It is a genuinely well-managed company with a distinctive, customer-friendly model (no-haggle pricing, instant online appraisal, omni-channel buying) and a remarkably stable gross-profit-per-unit engine. It is also, on the evidence, a structurally disadvantaged business: a low-single-digit share of a vast, fragmented, commoditizing market, with returns on equity (~4% reported, ~7–8% normalized in FY2026) that fall well short of any threshold indicating a durable competitive advantage.

The investment debate is a cyclical-recovery story bolted onto a structural-erosion story. Cyclically, CarMax is climbing out of a multi-year demand air-pocket: comparable-store used-unit sales went +21.9% (FY22) → −14.3% (FY23) → −6.7% (FY24) → +2.2% (FY25) → −2.0% (FY26), with a sharp −9.0% relapse in Q3 FY26 as affordability bit again. Used-vehicle prices are re-firming (Manheim index +3.6% y/y in May 2026) and an off-lease supply trough that bottoms in 2025 should add late-model inventory through 2027 — a real volume tailwind. Structurally, Carvana is taking share aggressively (596,641 units in CY25, +43%, at far higher per-unit gross profit), the no-haggle/omni-channel differentiator has been commoditized, and CarMax’s share of its target segment slipped to ~3.6% from 3.7%.

The single most important fact for any investor is that CAF supplies the large majority of CarMax’s normalized profit — roughly 70–90% of normalized pre-tax income, and 147% of reported FY26 pre-tax earnings (the retail segment was near break-even after charges). Management is deliberately growing CAF: penetration of ~42% toward a 50% target, and a planned push down the credit spectrum (Tier-2 target originations to ~30% of Tier-2 volume from <15%). It is doing so as US subprime auto delinquencies hit a 32-year high (~6.9% 60+ days) and as CAF’s own net charge-offs rise and recoveries fall. CarMax’s prime-heavy book is cushioned today, but the strategy manufactures near-term income (via CECL provisioning timing and securitization gains) while back-loading credit risk onto the one segment that carries the company.

Capital allocation has been below average: procyclical buybacks (peak purchases at $125/share in FY22, a hard pause near $40 in early FY26), a value-destructive Edmunds acquisition now impaired ($141.3M goodwill write-off), and an expensive 2020–2024 omni-channel build whose return is hard to see in flat-to-declining units and a halved FY26 profit. The board terminated long-tenured CEO Bill Nash in November 2025 and installed Keith Barr (ex-IHG hotels CEO, no automotive or consumer-finance background) effective March 2026 — a high-variance bet that CarMax’s problem is brand and customer experience rather than industry structure and capital discipline.

On valuation, consolidated multiples mislead because ~$15.8B of non-recourse CAF debt inflates enterprise value and a charge-laden trough depresses GAAP EPS. On honest reads — normalized P/E ~16–18x, P/B 1.14x (7th percentile of its own ten-year history) — the stock is fairly-to-modestly-cheaply pricing a base-case normalization in which volume recovers and CAF income holds. The asymmetry is the concern: the bear case (Carvana share loss + CAF credit deterioration + no SG&A leverage) de-rates earnings and the multiple simultaneously. This is a leveraged bet on two things going right at once — used-unit volume recovering and CAF credit normalizing rather than deteriorating.

2. Business Overview

CarMax operates two reporting segments: CarMax Sales Operations (used-vehicle retail, wholesale auctions, and “other” — extended protection plans and service) and CarMax Auto Finance (CAF), a consolidated captive lender. The company was founded in 1993 (as a Circuit City unit), IPO’d as a tracking stock in 1997, and was fully separated in 2002. It is headquartered in Richmond, Virginia, with ~27,800 full-time employees and 256 used-car superstores at FY2026 year-end (171 owned, 85 leased), plus seven stand-alone reconditioning/auction facilities.

How CarMax makes money — a per-unit spread plus a finance margin. The economic core is disarmingly simple: CarMax buys used vehicles (increasingly directly from consumers via its instant online appraisal/“we’ll buy your car” channel, as well as at auction), reconditions them at scale, and sells them at a targeted gross profit per unit (GPU) rather than a targeted margin percentage. In FY2026 it earned:

Vehicles that don’t meet retail standards (typically older/higher-mileage trade-ins) are sold through CarMax’s wholesale auctions — a high-velocity, lower-margin channel that also makes CarMax a credible buyer of any consumer’s car, supporting the appraisal/sourcing flywheel. The “other” pool — extended protection plans (EPP), guaranteed asset protection (GAP), and the service department — is the highest-margin slice (~78% gross margin on EPP) and has been the one bright spot, rising from $528/unit (FY24) to $671/unit (FY26), a 27% increase driven by EPP attach and service contribution.

CAF — the profit engine disguised as a feature. CAF provides financing to CarMax’s retail customers across the credit spectrum, retaining prime and a growing slice of near-prime/non-prime paper on balance sheet and funding it through securitizations and warehouse facilities. CAF financed ~42% (net) of CarMax’s retail units in FY2026 and generated $562.7M of segment income. As Section 6 details, this is not a sidecar: CAF is the majority of the company’s normalized pre-tax profit and the swing factor in the equity thesis.

Revenue composition (FY2024–FY2026, $M):

Revenue is overwhelmingly transactional (each car is a one-time sale), not recurring. The recurring-revenue-like element is CAF’s interest income on a seasoning loan book and the service/EPP tail — but even these ride the volume of vehicle sales. Verdict: A well-run but structurally low-margin (≈10.8% consolidated gross margin), capital-intensive, cyclical retailer whose true profit center is a captive consumer-finance balance sheet. Understanding CarMax means understanding CAF.

3. Industry Dynamics

Market size and structure. Roughly 39 million used vehicles change hands in the US each year, of which ~20 million are 0-to-10-year-old vehicles — CarMax’s addressable segment — implying an addressable market worth on the order of $1.0–1.3 trillion at prevailing transaction prices. The market is extraordinarily fragmented: more than 18,000 franchised new-car dealers (who sell the majority of late-model used vehicles through their used and certified-pre-owned operations), plus tens of thousands of independent used-car dealers and a large private-party channel. After three decades as the category’s pioneer and largest player, CarMax holds only ~3.6% of the 0–10-year segment — and that share slipped from 3.7% the prior year.

That single statistic is the industry’s structural signature. By the lens of competitive-strategy analysis, you cannot count the meaningful competitors on one hand, the leader cannot break out of low-single-digit share, and share is not stable — all hallmarks of an industry without durable barriers to entry. A large, growing addressable market is, counterintuitively, the enemy of a scale advantage: fixed costs shrink as a share of total cost, and new entrants can reach efficient operating scale serving customers who are equally available to everyone.

The pricing cycle. Used-vehicle values spiked from a pre-COVID Manheim index base of ~110–120 to a ~257 peak in early 2022 (pandemic supply shock), then normalized, and are now re-firming: the Manheim Used Vehicle Value Index was 212.6 in May 2026 (+3.6% y/y), having touched 215.3 in March — the highest since mid-2023. Prices remain ~75–90% above pre-COVID levels. The driver is supply: pandemic-era leasing collapsed (leasing was just ~22.8% of 2020–2024 volume), so off-lease returns bottom at ~2.4M units in 2025, recover to ~3.2M (2026) and ~3.6M (2027), then the pipeline largely runs dry. This is a double-edged setup for CarMax: tight late-model supply supports per-unit margins (used GPU held at $2,253 despite the cycle) but raises acquisition costs and — combined with ~7–9% auto-loan APRs and elevated prices — pushes monthly payments to levels that throttle unit demand. CarMax is in a volume-constrained, price-supported regime: good for GPU, bad for the unit growth the equity story needs.

Competitive intensity — rising, and partly price-based. The defining competitive fact is Carvana. In CY2025 Carvana sold 596,641 retail units (+43% y/y) at total GPU of ~$7,000–7,500 — multiples of CarMax’s used GPU — and is building out ADESA-based reconditioning capacity toward >1.5 million units annually (≈2.5x its current volume). Carvana is now within striking distance of CarMax’s unit count and is growing while CarMax shrinks. ADESA also gives Carvana an owned wholesale/logistics backbone analogous to CarMax’s, neutralizing CarMax’s historical reconditioning-scale edge. Layer on the franchised mega-groups (AutoNation, Lithia/Driveway, Penske, Group 1, Sonic) pushing used and CPO through their own omni-channel platforms, and the long tail of independents, and the picture is of intensifying, capacity-led rivalry. Online price transparency has commoditized the no-haggle proposition CarMax invented; it is now table stakes.

Auto-finance backdrop (affects CAF). The consumer-credit environment is stressed and bifurcated. Subprime 60±day auto delinquencies hit ~6.90% in January 2026 — a 32-year record — and repossessions reached ~1.73M in 2025, the highest since 2009. CAF’s book is prime-heavy (total past-due 5.11%, 90±day just 0.41%), so the subprime wave is more a risk-and-opportunity frame than a direct hit today — but it throttles consumer demand broadly and signals where the cycle is. Regulation has eased federally: the FTC’s CARS Rule (a vehicle-shopping junk-fee rule) was vacated by the Fifth Circuit (January 2025) and formally withdrawn (February 2026). The tail risk has migrated to activist state attorneys general and CFPB enforcement.

Capital cycle. Used-car e-commerce ran a textbook capital cycle: a 2020–21 capital flood (Carvana’s run, Vroom’s IPO, Shift’s SPAC), then a brutal bust (Shift filed Chapter 11 in 2023; Vroom wound down retail in early 2024; Carvana itself nearly failed and restructured ~$1.2B of debt). The marginal capital has been cleansed — a one-time benefit to survivors. But the favorable, disciplined-under-supply half of the cycle is not in place: rather than survivors exercising pricing restraint, Carvana is aggressively rebuilding capacity it must fill, a forward-looking negative for industry GPU and share stability.

Verdict: Structurally bad-to-mediocre. No durable barriers to entry; a leader stuck at low-single-digit, declining share; a commoditized differentiator; and a capital cycle whose constructive phase is incomplete because the #2 player is on a capacity offensive. The cyclical supports (re-firming prices, off-lease supply recovery) are real but largely offset by intensifying rivalry and an affordability ceiling. The one genuinely franchise-like asset is CAF — but that is a spread/credit business riding the rate and credit cycle, not an escape from a weak industry structure.

4. Competitive Position

The moat question, answered directly: there is no durable competitive moat — at most a narrow, eroding operational-scale advantage that fails the financial test. CarMax is a quality operator, not a quality franchise. Running through the genuine advantage types:

Economies of scale (the only candidate). CarMax has real operational scale: it is the largest single consumer-sourcing/appraisal engine in the country (it bought ~1.1M vehicles in FY2026, an increasing share directly from consumers), with nationwide reconditioning and logistics and three decades of pricing/appraisal data. This is genuine efficiency. But a scale advantage requires that the edge widen with size and translate into superior, defensible returns — and here it does not. SG&A as a percentage of gross profit deteriorated to 87.4% in FY2026 (from ~84%); the per-unit cost-savings program delivered only modest results and missed its FY26 target. Carvana has demonstrated that the online model scales without 256 stores, and ADESA has handed it comparable reconditioning scale. CarMax’s fixed costs are a shrinking slice of a $1T+ pool — scale efficiency, not a scale moat.

Customer captivity / switching costs: absent. A used car is an infrequent, considered, high-ticket purchase; there is essentially no repeat-purchase lock-in on a meaningful cadence. CarMax’s own customer-friendly policies actively remove friction and lock-in: a 30-day/1,500-mile money-back guarantee, multi-day return windows, and an instant online appraisal that customers can take to any competitor. These are excellent for customers and for trust, but they are the opposite of captivity.

Brand / no-haggle pricing: a feature, not a barrier. CarMax pioneered no-haggle pricing and transparent appraisal, and the brand is strong and trusted. But the FY2026 10-K itself acknowledges that competitors “have replicated or attempted to replicate” the no-haggle model and the buy-your-car offer. A differentiator that the market leader admits is being copied is, by definition, not a barrier to entry. Online price aggregation has made transparent pricing universal.

Intangibles / proprietary technology: modest and narrowing. CarMax’s appraisal algorithms and data are an asset, but the 10-K flags that third-party dealer software and appraisal tools are eroding the gap, and Carvana/the franchised groups have built competitive digital stacks.

The decisive financial tests. The two cleanest moat tests are market-share stability and excess returns on capital. CarMax fails both: share is ~3.6% and falling, and ROE is ~4% reported / ~7–8% normalized — far below the 15–25% range that signals a real advantage, and even that is flattered by CAF’s balance-sheet leverage. An advantage that cannot be tied to excess returns on capital that would deteriorate in its absence is, by definition, not a moat.

Pressure tests against the competitive set. Carvana proves the format scales and is taking share at superior unit economics. Franchised dealer groups own the late-model/CPO niche and bundle finance and service. The independent long tail keeps anyone below ~4% share. CarMax is now the slower-growing incumbent defending share — the structurally weaker position in a commoditizing category.

Verdict: A crowded, commoditizing market with weak differentiation. CarMax is the best-run operator in a structurally bad-economics industry — a quality management team, not a durable competitive franchise. The lone franchise-like asset is CAF’s captive-channel scoring advantage (it sees CarMax’s customers first and can price risk on its own population), and that is real — but it is a finance edge, not a retail moat, and it comes with credit-cycle exposure.

5. Growth History and Forward Opportunities

History. CarMax’s growth has come from two engines historically: store-base expansion and comparable-store unit growth, amplified by CAF. Both have stalled. Store openings have decelerated to a crawl (240 → 245 → 250 → 256 stores across FY24–FY26; just 5, 5, and 6 openings; FY27 guidance is four). Comparable-store used-unit sales have been negative or barely positive for four straight years:

The FY22 surge was a stimulus-and-supply-shock anomaly; the business has not posted durable positive comps since. FY25’s +2.2% bounce proved fragile, reversing to −2.0% in FY26 with a sharp −9.0% relapse in Q3 as affordability re-tightened and customers traded down to older, higher-mileage vehicles (>40% of Q3 sales, up ~5 points). Total revenue has declined in each of the last two fiscal years.

Quality of historical growth: low. What growth there has been is volume-driven (GPU is held flat by design), cyclical, and increasingly hard-won — FY26 used GPU was cut ~$58 to stimulate Q4 demand, and management has guided to further per-unit margin declines (~−$300/unit in Q1 FY27) to defend volume. This is the opposite of high-quality, pricing-power-driven growth; it is buying units with margin.

Forward opportunities.

• Off-lease supply recovery (cyclical, real): the 2026–2027 rebound in lease returns should ease the late-model inventory constraint and support unit volume — the most credible near-term growth lever.
• CAF penetration (the management focus): growing financed-unit penetration from ~42% toward 50%, with each point worth ~$10–12M of lifetime pre-tax income. This is genuine incremental profit — but it is being pursued via down-spectrum lending into a stressed credit environment, so it adds risk-weighted, not risk-free, growth.
• “Other” gross profit (EPP/service): continued attach-rate and service growth, the highest-margin pool, has compounded ~27% over two years and is a quieter, higher-quality grower.
• Omni-channel maturity: the expensive digital build is complete; if it converts to share gains rather than mere defense, it could re-accelerate units — but the evidence so far is defensive (units flat-to-down through the build).
• New-CEO strategy reveal (June 2026): Keith Barr is expected to lay out a refreshed strategy and long-term targets — a wildcard, not a quantifiable opportunity yet.

Verdict: Low-quality, currently-stalled growth. The realistic path to renewed unit growth is cyclical (off-lease supply) rather than structural (share gains), and the most profitable growth lever (CAF penetration) is being pulled in a way that imports credit risk. The equity story needs volume to inflect; the structural backdrop (Carvana, affordability) makes durable inflection uncertain.

6. Financial Quality

The headline. FY2026 net earnings fell ~51% to $247.3M (diluted EPS $1.68) from $500.6M ($3.21) in FY2025; FY2024 was $479.2M ($3.02). The decline reflects a soft top line (revenue −1.8%), SG&A deleverage, and two charges: a $141.3M Edmunds goodwill impairment (non-cash, non-deductible) and ~$49.8M of restructuring (CEO-change severance, workforce reductions, Edmunds lease abandonment). Normalizing for these, FY26 diluted EPS is ~$2.63 (impairment add-back) to ~$2.89 (also tax-effecting restructuring); management’s “adjusted” figure is ~$2.91. Even normalized, earnings fell ~10–18% year-over-year — this is not purely an optics problem.

Three-year P&L (consolidated, $M):

Do economics improve with scale? No — they are deleveraging. The gross-profit engine is high quality and durable: used GPU is pinned at ~$2,250–2,310 across a violent price cycle, and “other” GP per unit is rising. But because management targets dollar GPU rather than margin, the model cannot price its way to higher per-unit profit — gross profit is essentially volume × a fixed spread. With volume negative and SG&A per unit rising ($1,860 in FY26), SG&A absorbed 87.4% of gross profit, up from ~84%. That is negative operating leverage — the opposite of what a scale advantage should produce.

CAF deep-dive — the company’s true profit center and its biggest quality-of-earnings risk.

Two things stand out. First, CAF is the company. CAF income of $562.7M equals 147% of FY26 reported pre-tax earnings (because the impairment and restructuring sit in the retail segment) and ~70–90% of normalized pre-tax income (FY25 ~87%, FY24 ~89%). The “used-car retailer” earns the large majority of its profit from a consumer lender; the retail segment, after charges, was roughly break-even-to-negative in FY26. Second, the credit metrics are deteriorating into a down-spectrum push. Net charge-offs have risen from 1.93% to 2.34% of average loans in two years; recovery rates have fallen from 53% to 45.5%; yet management plans to raise Tier-2 target originations to ~30% of Tier-2 volume (from <15%) and grow penetration toward 50%. Because CECL books lifetime-loss provisions and securitizations recognize gains up front, this strategy manufactures reported CAF income today while back-loading credit risk onto the segment that carries the entire company. That is the single biggest quality-of-earnings concern.

Balance sheet — sound on a recourse basis; the scary number is non-recourse. Total debt of $18.05B looks alarming against $5.89B of equity, but ~$15.8B is non-recourse securitization/warehouse debt secured solely by the auto-loan pool (investors have no recourse to CarMax’s other assets). Recourse/corporate debt is only ~$2.2B (revolver $840.8M, term loan $499.3M, senior notes $400M, financing obligations $483.6M) against $122.8M of cash. Corporate leverage is moderate — but it rose in FY26 (the revolver was drawn from zero to $840.8M, partly to fund buybacks), cash fell from $247M to $123M, and management flagged leverage as “slightly above our targeted range,” which is why it paused repurchases in Q4. Book value per share is $41.53; shares outstanding fell 7.5% (153.3M → 141.8M) on FY26 buybacks. Inventory is $4.14B (~5.5x turns).

Quality-of-earnings flag on cash flow. FY26 operating cash flow of $1.78B looks far stronger than the $247M of net income — but ~$909M is a one-time “proceeds from sale of auto loans” (a September 2025 non-prime securitization structured off-balance-sheet, ~$930M sold for a $26.9M gain), and the headline is further flattered by the loan book contracting ~$1.3B (the auto-loans-held-for-investment line sits inside operating activities). Normalize these out and core operating cash is ~$875M; in FY24, when the loan book grew ~$980M, reported free cash flow was actually negative. The presentation makes FY26 cash generation look much more robust than the underlying business delivered.

Verdict: Economics are not improving with scale — they are deleveraging on soft volume. The durable GPU engine and the sound recourse balance sheet are genuine strengths, but reported and even normalized earnings are down, the headline cash flow is flattered by a one-time loan sale, and the company’s profit is concentrated in a captive lender being steered down-spectrum into a deteriorating credit cycle. High operational quality; low current earnings quality.

7. Capital Allocation

Capital allocation is the bridge from business value to shareholder value, and here the bridge is weak. CarMax pays no dividend, so the entire capital-return judgment rests on buyback timing — which has been poor.

Buybacks — procyclical, a textbook capital-cycle error.

The pattern is buy-high, hesitate-low, stop-at-the-bottom: the largest per-share prices were paid at the FY22 peak ($125.49, now deeply underwater); buybacks slowed to a $91M trickle in FY24 precisely as the stock had derated; the company finally leaned in during FY26 at an attractive $53.76 (the one well-timed tranche) — and then paused entirely in January–February 2026 at ~$40, within pennies of the multi-year low, to defend a leverage ratio. The dollar-weighted average purchase price across the period sits well above today’s $47. Roughly $1.31B of the $2.0B authorization remains unused. The lone redeeming feature is the FY26 volume; the pause squanders even that.

M&A — Edmunds, a value-destructive peak-cycle deal. CarMax acquired Edmunds (the automotive shopping/data site) in June 2021 for ~$404M — the same window in which it overpaid for its own stock. In Q4 FY26 it wrote off $141.3M of Edmunds goodwill (citing a decline in market capitalization, pressured performance, and downward forecast revisions) and abandoned the Edmunds lease (a ~$20M occupancy charge). A goodwill impairment plus lease abandonment is a quantified admission that the deal has not earned its cost of capital — and management’s incentive compensation added the impairment back to pre-tax income, shielding executives from the consequences of their worst capital-allocation decision.

Footprint and the omni-channel build. Store growth has effectively stopped (FY27 guidance: four new stores), and capex is being cut to ~$400M (from ~$541M) — an implicit concession that the growth-capex cycle is over. The expensive 2020–2024 omni-channel and technology build coincided with flat-to-declining units, falling comps, and a halved FY26 profit. The most charitable read is that the investment was defensive (necessary to match Carvana) rather than value-creating — a case where a capability available to all confers durable advantage on no one.

CAF capital and funding. CAF is funded the right way — non-recourse, term-matched securitizations ($13.50B ABS term + $2.33B warehouse, with $4.02B of unused warehouse capacity) that ring-fence risk and are capital-efficient. The judgment flag is the down-spectrum lending push: a deliberate risk-for-spread decision on a book already ~2.7x equity, late in the consumer-credit cycle.

Incentives and alignment. The FY26 proxy reveals a meaningful governance weakness: executive incentives reward EBIT (55%), market share (15%), and operational execution (30%) in the annual bonus, and pre-tax income in long-term PSUs — but contain no ROIC, ROCE, or economic-profit metric anywhere. For a business whose thesis hinges on capital discipline (a finance book at 2.7x equity plus procyclical buybacks), rewarding EBIT growth and stock price — i.e., size and momentum — is the very incentive structure that tends to drive over-investment. Mitigants: pay-for-performance genuinely bit in FY26 (the annual bonus paid at 30% of target; a prior PSU cycle paid 0%), and the plan is shifting toward TSR-linked market stock units to reduce dilution. Insider ownership is thin (~1.0% for all directors and officers); the register is institution-dominated (BlackRock ~11.8%, AQR ~5.6%, Vanguard ~5.3%).

Management change. The board terminated long-tenured CEO Bill Nash in November 2025 (he also left the board) and, after an interim period, installed Keith Barr (former CEO of InterContinental Hotels Group, 2017–2023) effective March 16, 2026. Barr has a strong brand/loyalty/digital-transformation background but no automotive-retail or consumer-finance experience — a notable gap for a company that is, in profit terms, mostly a subprime-tilting lender. The hire signals the board views CarMax’s problem as brand and customer experience, not industry structure or capital allocation. His package is conventional ($1.25M base, 175% target bonus, $1M RSU + $1M option sign-on, $3.5M option + $3.5M PSU annual, 2x severance) and not egregious — but it, too, lacks an ROIC linkage.

Verdict: Below average. Procyclical buybacks (bought high, paused low), a value-destructive and impaired acquisition (with management shielded from it), an expensive omni-channel build of unclear return, no returns-on-capital metric in compensation, and ~1% insider ownership. The mitigants — sound non-recourse CAF funding, compensation that did pay out low, and a forced turn toward discipline (capex cut, FY26 buyback volume) — keep this from “egregious,” but management has not demonstrated the capital-allocation skill that converts a decent business into shareholder value.

8. Changes and Headwinds — Last Two Years

Leadership upheaval (the dominant change). The most consequential development is the abrupt, for-cause-style termination of CEO Bill Nash in November 2025 after ~9 years in the role and three decades at the company, an interim stint by director David McCreight, and the March 2026 arrival of Keith Barr from the hotel industry. A CEO transition of this nature at a company already struggling with volume and competition introduces material execution and strategy-discontinuity risk. Barr’s first strategic plan and long-term targets are expected in June 2026.

The Edmunds write-down. The $141.3M goodwill impairment and lease abandonment in Q4 FY26 closed the book on a peak-cycle acquisition and contributed to the halving of FY26 EPS.

CAF’s strategic shift down-spectrum. Over the past two years CAF has moved from a predominantly prime book toward “full credit spectrum” lending, with a stated plan to grow Tier-2 originations and push penetration toward 50% — a strategic change that raises the company’s structural credit exposure.

Competitive pressure from a resurgent Carvana. Carvana’s return from near-bankruptcy to 43% unit growth and aggressive capacity expansion is the central competitive headwind, reversing the post-2022 reprieve when Carvana was retrenching.

Demand and affordability headwinds. Elevated vehicle prices plus ~7–9% APRs have throttled units (comp-store used units −9.0% in Q3 FY26), forcing CarMax to cut GPU to defend volume — a margin-for-volume trade-off that management has signaled will continue into FY27.

Buyback pause and rising leverage. Drawing the revolver to fund repurchases, then pausing buybacks as leverage exceeded target, is a self-inflicted balance-sheet tightening that removed a support for the stock at the lows.

Regulatory easing (a modest tailwind). The vacatur and withdrawal of the FTC CARS Rule is a minor relief for dealer practices, partially offset by persistent state-AG and CFPB enforcement risk.

Verdict: On balance these changes weaken the thesis. The leadership disruption, the impairment, the down-spectrum credit shift, and the competitive resurgence outweigh the cyclical supply tailwind and the regulatory easing. The next 12 months hinge on an unproven CEO executing a volume turnaround while managing rising credit risk in the profit engine.

9. Risk Analysis

The dominant risk cluster is the intersection of CAF credit and competitive volume loss. Because CAF is the profit engine and it is being pushed down-spectrum into a 32-year-high delinquency environment, a credit cycle that turns would hit precisely the segment that carries the company — and it would likely coincide with weak retail volume (same macro cause). The bear case is therefore correlated, not diversified: multiple, mutually-reinforcing risks fire together. The reassuring counterpoint is the absence of existential risk — recourse leverage is modest and the finance book is ring-fenced, so a total loss is very unlikely; the realistic downside is a value-trap derating, not a wipeout.

10. Valuation Discussion (Embedded Expectations)

Why consolidated multiples mislead. CarMax’s $18.05B of total debt makes EV/EBITDA look alarming (~15–24x depending on source), and its charge-laden GAAP EPS of $1.68 makes trailing P/E look expensive (~28x). Both are artifacts. ~$15.8B of the debt is non-recourse CAF funding term-matched against a self-liquidating loan book; consolidating a finance company’s balance sheet into a single EV/EBITDA mashes together a retailer (where EBITDA is meaningful) and a lender (where interest is cost-of-goods and the right lens is P/E or price-to-book). And GAAP EPS is depressed by the non-deductible impairment and restructuring. The honest reads are normalized P/E ~16–18x (on ~$2.63–2.89 normalized EPS) and P/B 1.14x — the latter sitting in the 7th percentile of CarMax’s own ten-year history, alongside a P/S in the 7th percentile. The stock is cheap on book and sales, “expensive” only on trough earnings.

Sum-of-the-parts (illustrative, not a target). Valuing the two businesses separately:

The current ~$5.43B cap sits in the middle of this range. The decisive insight: most of CarMax’s equity value rests on CAF plus a normalized-volume recovery in retail, not on today’s near-break-even retail P&L. That concentration is also the central risk.

Comparable multiples (public market-data aggregators, June 5 2026; reconciled to filings).

There are two valuation worlds: franchised dealers at 7–12x earnings (cyclically bottomed on normal earnings) and Carvana at a growth premium. KMX trades at a P/E premium to franchised peers only because its E is at a charge-distorted trough; normalized, it is ~16–18x — a premium to franchised dealers, a discount to Carvana. CVNA’s 3.24x P/S vs KMX’s 0.21x quantifies the market’s growth-vs-stability bet.

Embedded expectations. Reverse-engineering ~$47.15 at a 15–18x blended multiple implies the market is underwriting normalized EPS of ~$2.6–3.1 — almost exactly the normalized estimate here, and below FY25/FY24’s $3.21/$3.02. So the price embeds normalization, not heroics: that the charges don’t recur, that used-unit volume stabilizes off the off-lease trough, that CAF income holds near $560M despite rising losses, and that SG&A re-leverages. The market is pricing the trough correctly (P/B and P/S in the 7th percentile say it is not extrapolating depression indefinitely). What it may be underweighting is (i) CAF credit normalization as penetration pushes down-spectrum, (ii) Carvana’s structural share/GPU advantage, and (iii) the timing of SG&A re-leverage if volume recovers slowly. The sell-side mean target of ~$41.69 (below the current price; third-party color, not a target here) is internally consistent with a market that sees full normalization as already priced while near-term fundamentals still deteriorate.

Scenario analysis (3-year horizon to ~FY29; all figures illustrative, assumption-based).

The bear case is the value-trap geometry: earnings and the multiple de-rate together (~40–55% downside). The base case is modestly above today (the market is paying roughly fair for normalization). The bull case roughly doubles the equity if volume and CAF and SG&A all deliver.

Verdict: At ~$47, KMX is fairly-to-modestly-cheaply discounting a base-case normalization — but only after looking through trough GAAP earnings and disaggregating the CAF balance sheet. It is neither obviously cheap nor obviously rich. It is a leveraged bet that (1) CAF income holds while credit normalizes rather than deteriorates and (2) used-unit volume recovers off the off-lease trough. The asymmetry is unfavorable in the bear case. No price target, no recommendation.

11. Variant Perception

Consensus view. The Street sees CarMax as a high-quality category leader at a cyclical earnings trough, with off-lease supply recovery and CAF penetration as the recovery levers — but is cautious enough on near-term volume and competition that the mean price target sits below the current price. In other words, consensus is “good company, fully-priced-for-now, prove the recovery.”

The strongest bull case. CarMax is the best-capitalized, best-run operator in a fragmented industry at a genuine cyclical bottom. Off-lease supply recovers through 2027, restoring late-model inventory and unit volume; the completed (now-cheaper) omni-channel build finally converts to share; SG&A re-leverages hard on returning volume; CAF penetration marches to 50% with each point adding ~$10–12M of lifetime profit; the heavy short interest (11.6% of float) and a credible new-CEO strategy reveal spark a re-rating. Normalized EPS recovers above $3.20 and the multiple re-rates to a quality 18x — the bull’s ~$88–104/share. The stock is cheap on book (1.14x, 7th percentile) precisely because the market is over-anchored on a charge-distorted trough.

The strongest bear case. CarMax is a structurally disadvantaged, low-return retailer (ROE ~7–8% normalized) losing share to a faster, higher-GPU competitor in a commoditizing category — and its profit is really a subprime-tilting loan book being pushed down-spectrum at a 32-year delinquency high. CAF net charge-offs accelerate as the 2025–26 down-spectrum vintages season; provisions spike; the segment that is 70–90% of profit cracks just as retail volume stays weak on affordability. SG&A never re-leverages, capital allocation stays procyclical, and the unproven hotelier-CEO mis-executes. Earnings and the multiple de-rate together to the bear’s ~$21–25/share. The “cheap on book” argument is a value trap because book itself is at risk if CAF credit turns.

The 3–5 assumptions that matter most:

1. CAF credit: normalizes vs. deteriorates as penetration pushes down-spectrum. (Most load-bearing — CAF is the company.)
2. Used-unit volume inflects durably positive off the off-lease trough, enabling SG&A re-leverage.
3. Carvana does not permanently out-take share at structurally superior unit economics.
4. The charges ($141M impairment, restructuring) are genuinely non-recurring (disclosed as such — high confidence).
5. The new CEO executes a coherent strategy without value-destructive reinvestment.

Falsification evidence: The bull case is falsified by CAF net charge-offs continuing to climb past allowance coverage and recovery rates falling further, or by another year of negative comps as Carvana’s share gains. The bear case is falsified by two-plus consecutive quarters of positive used-unit comps with stable CAF credit metrics and stabilizing share.

12. Fact vs. Interpretation Table

13. Open Questions

1. CAF credit trajectory: Will net charge-offs on the 2025–26 down-spectrum vintages stabilize within the 2.78% allowance, or accelerate through it and force provision spikes? This is the swing variable for the entire thesis.
2. True CAF segment ROIC: CAF carries no allocated indirect cost in the filings, so its standalone return on allocated equity is not directly knowable — what is the genuine economic return on the capital tied up in the loan book?
3. New-CEO strategy (June 2026): What will Keith Barr’s long-term targets and capital-allocation philosophy be — and will an outsider with no auto/credit background prioritize the right levers?
4. Volume inflection durability: Is the modest sequential comp improvement (Q3 −9.0% → Q4 −1.9%) a genuine off-lease-driven recovery or a price-cut-bought, temporary bounce?
5. SG&A re-leverage: At what unit-volume level does SG&A/GP return toward the historical ~84%, and is that volume achievable against Carvana?
6. Carvana’s GPU advantage: Why is CVNA’s total GPU (~$7,000+) so far above CarMax’s, and is it structural (mix, finance attach, ADESA) or a temporary harvest that will compress?
7. Buyback resumption: At what leverage level and price will repurchases restart — and will management again be procyclical?

14. What Must Be True

Bull case — what must be true:

• Used-unit volume inflects durably positive off the off-lease supply recovery, driving SG&A re-leverage back toward ~84% of gross profit.
• CAF income holds near/above $560M with stable-to-improving credit (NCOs plateau, recoveries stabilize) even as penetration grows toward 50%.
• CarMax stabilizes its ~3.6% share against Carvana rather than continuing to cede it.
• Falsification test: If, over the next four quarters, used-unit comps remain negative or CAF net charge-offs rise above ~2.6% with recovery rates still falling, the bull thesis is broken.

Bear case — what must be true:

• CAF credit deteriorates materially as down-spectrum vintages season, forcing provision increases that hit the profit engine.
• Carvana continues to take share at superior unit economics, keeping CarMax volume flat-to-negative and preventing SG&A re-leverage.
• Earnings and the multiple de-rate together toward franchised-dealer levels (~12x on lower normalized EPS).
• Falsification test: If CarMax posts two-plus consecutive quarters of positive used-unit comps with stable CAF credit metrics (NCOs flat/declining, recoveries stabilizing), the bear thesis is broken.

The two falsification tests are deliberately symmetric and observable within ~12 months: the debate resolves on (a) used-unit comps and (b) CAF charge-off/recovery trends. Watch those two series above all.

15. Source Appendix

Primary sources (full citations in the Source Appendix below):

• CarMax FY2026 Form 10-K (filed 2026-04-15, year ended Feb 28, 2026); FY2025 10-K (2025-04-11); FY2024 10-K (2024-04-15). SEC EDGAR, CIK 0001170010.
• CarMax Q3 FY2026 Form 10-Q (period ended Nov 30, 2025) and prior 10-Qs.
• CarMax 8-Ks (2023–2026): earnings, CEO termination (Nov 2025), Keith Barr appointment (2026-02-10), buyback disclosures.
• CarMax DEF 14A proxy (2026-05-12): executive compensation, incentive metrics, beneficial ownership.
• CarMax Form 4 insider filings (2023–2026); director open-market purchases (Oct 2025).
• Cox Automotive / Manheim Used Vehicle Value Index (May & mid-April 2026).
• Carvana Q4/FY2025 results release (2026-02-18).
• Bankrate, Wolf Street (subprime auto delinquency data, Q1 2026).
• Holland & Knight / Federal Register (FTC CARS Rule vacatur and withdrawal).
• Public Q4 FY2026 earnings-call transcript (2026-04-14, via Motley Fool).

All quantitative figures reconciled to the FY2026 10-K. Third-party aggregator data and sell-side targets are signals only, and no price target appears outside the Author’s Take.

Appendix A — Diligence Questionnaire

Supplemental to the research article. Fact/Interpretation/Assumption labels applied where it matters. Where a question does not map to a used-vehicle-retailer-plus-captive-lender, the correct sector analog is given.

General

What thoughtful questions have other investors asked about this company? The recurring institutional debates are: (1) Is the earnings recovery cyclical or is CarMax structurally impaired by Carvana? — i.e., is the post-2022 unit weakness a demand air-pocket that off-lease supply will heal, or permanent share loss. (2) How much of the company is really a subprime lender? — given CAF is ~70–90% of normalized pre-tax profit, many investors treat KMX as a financial, not a retailer, and scrutinize CECL adequacy and the down-spectrum push. (3) Was the omni-channel spend value-creating or merely defensive? (4) Why pause buybacks at the lows? (5) Can a hotel-industry CEO turn around an auto retailer/lender? These map directly to the variant-perception section.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? (Interpretation) A low. FY26 GAAP EPS $1.68 (normalized ~$2.63–2.89) is well below FY25 $3.21 and FY22-era peaks; comps have been negative/flat for four years; used-vehicle volume is constrained by affordability and an off-lease supply trough that bottoms in 2025. Own-history valuation percentiles confirm it: P/E in the 95th percentile (depressed E) but P/B and P/S in the 7th percentile.

Driven by the external environment or internal actions? (Interpretation) Mostly external (rates, affordability, used-price cycle, off-lease supply) compounded by internal choices (GPU cuts to defend volume; the Edmunds impairment; CEO disruption). The competitive pressure from Carvana is a structural external factor, not purely cyclical.

How stable are revenues? (Fact) Not very — revenue declined in each of the last two fiscal years; each car is a one-time transaction. The most stable streams are CAF interest income (on a seasoning book) and EPP/service (“other” GP), but both ride vehicle-sale volume.

Outlook for products/services? (Interpretation) Volume outlook hinges on off-lease supply recovery (2026–2027 tailwind) versus affordability and Carvana share loss (headwinds). Management guides to further per-unit margin declines into FY27 to defend volume.

How big will this market be — growing, shrinking, domestic or international? (Fact) ~39M US used vehicles/year (~20M in CarMax’s 0–10-year segment); a large, roughly GDP-paced, domestic-only market (~$1.0–1.3T). Growing slowly in units; CarMax’s share of it is the issue, not the market size.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? (Interpretation) More. Carvana’s resurgence (+43% units, 2.5x capacity build), franchised-group omni-channel pushes, and commoditized online pricing all intensify rivalry.

How profitable is the business (ROIC, ROE)? (Fact) Low: ROE ~4% reported / ~7–8% normalized FY26; net margin ~0.9%. True ROIC is obscured by the consolidated finance book; the retail segment was near break-even after FY26 charges. Well below any moat threshold.

How profitable is the industry — how many competitors, what barriers to entry? (Fact/Interpretation) Structurally low-margin and hyper-fragmented: >18,000 franchised dealers plus tens of thousands of independents; the leader holds ~3.6% after 30 years. Barriers to entry are weak-to-absent.

Can the business be easily understood? (Fact) The retail model is simple (buy/recondition/sell at a fixed dollar GPU). The complication is CAF — a securitizing consumer lender whose CECL accounting and credit-tier strategy require real diligence.

Can it be undermined by foreign low-cost labor? (Fact) No — used-car retail/reconditioning is inherently domestic and service-local. Not a labor-arbitrage-exposed business.

Do brands matter? (Interpretation) The CarMax brand (trust, no-haggle) matters for customer acquisition but is a feature competitors have replicated, not a barrier. It supports, but does not protect, the franchise.

What is the nature of competition? (Interpretation) Increasingly price- and convenience-based (online appraisal, financing attach, delivery), with Carvana competing on growth/price and franchised groups on CPO/late-model inventory.

Customers’ switching costs? (Fact) Essentially none — infrequent purchase, generous return windows, and a portable instant appraisal all reduce lock-in.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? (Interpretation) The CAF customer-scoring data/relationship and the brand are valuable intangibles not capitalized. Conversely, the Edmunds goodwill that was on the balance sheet has now been written to zero.

Off-balance-sheet liabilities? (Fact) Operating leases (~$522M liability, largely capitalized under ASC 842). A September 2025 non-prime securitization was structured for off-balance-sheet treatment (~$930M loans sold) — worth monitoring as a recurring lever.

How conservative is the accounting? (Interpretation) Mixed. The dollar-GPU discipline and CECL allowance are reasonable, but CECL provisioning and securitization gain-on-sale give management estimation latitude, and the FY26 headline operating cash flow is flattered ~$909M by a one-time loan sale — a presentation that overstates underlying cash generation.

How CapEx-hungry is the business? (Fact) Moderately — ~$465–541M/year recently (stores, reconditioning, technology), guided down to ~$400M in FY27 as store growth slows. The larger capital sink is the finance book, which absorbs balance-sheet capital as it grows.

Capital Allocation & Management

How much FCF does the business generate; how does management use it; what is the philosophy? (Fact/Interpretation) “FCF” is muddied by the finance book swinging through operating cash flow; core retail FCF is modest (~$875M normalized FY26, sometimes negative when the loan book grows). The sole use is buybacks (no dividend). Philosophy has been growth-then-buyback, but buyback timing has been procyclical.

Significant acquisitions recently? (Fact) Edmunds (June 2021, ~$404M) — now impaired ($141.3M goodwill write-off + lease abandonment in FY26). A clear value-allocation misstep.

Buying back shares? (Fact) Yes — FY26 $631.9M / 11.75M shares at $53.76 (good value), but paused in Jan–Feb 2026 at ~$40 on above-target leverage, after overpaying at $125.49 in FY22.

Issuing large amounts of new shares to insiders? (Fact) No large dilution; equity comp is modest and the share count is falling. FY27 plan shifts options to TSR-linked MSUs to reduce dilution.

Compensation policy of directors/management? (Fact/Interpretation) Bonus = EBIT (55%) + market share (15%) + operational (30%); LTI on pre-tax income, moving to PSU/MSU. No ROIC/ROCE metric — a governance weakness for a capital-intensive lender. FY26 paid out low (30% bonus), so it did bite; but the Edmunds impairment was added back, shielding management.

Motivations of management? (Interpretation) Incentives reward EBIT growth, share, and stock price (size/momentum) rather than returns on capital. Insider ownership is thin (~1%). The new CEO (Barr, ex-IHG) signals a board bet on brand/experience over capital discipline.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? (Fact) No — a US common stock (NYSE: KMX), standard 1099 treatment. No dividend.

Dividend policy? (Fact) None; CarMax has never paid a dividend. All capital return is via buybacks.

How profitable is the business? (Fact) Thinly, at present — net margin ~0.9%, ROE ~4% (FY26). Normalized ROE ~7–8%.

Is net income diverging from cash from operations? (Fact) Yes — FY26 OCF ($1.78B) vastly exceeds net income ($247M), but ~$909M is a one-time loan sale plus loan-book contraction; normalized core OCF (~$875M) is far closer to a normalized earnings base. The divergence flatters, rather than confirms, earnings quality.

Risks & Downside

What factors would cause the stock to decline? (Interpretation) CAF credit deterioration (the dominant risk), continued share loss to Carvana, failure of volume to inflect, SG&A staying deleveraged, a recession hitting big-ticket discretionary demand, or further capital missteps.

Risk of a catastrophic loss? (Interpretation) Low-to-moderate. The realistic downside is a value-trap derating (bear case ~$21–25/share) if CAF credit turns while retail stays weak — earnings and multiple compress together.

Chance of a total loss? (Interpretation) Very low. Recourse corporate debt is only ~$2.2B; the ~$15.8B of finance debt is non-recourse and ring-fenced against the loan pool; there is no existential leverage.

Recent News & Events

Has the business environment changed recently? (Fact) Yes — Carvana’s resurgence, a 32-year-high in subprime auto delinquencies, re-firming used prices, and an off-lease supply trough all materially shape the environment. Recent news flow is quiet (no high-importance items), consistent with a between-catalysts tape ahead of the June 2026 strategy reveal.

Significant acquisitions? (Fact) None recent beyond Edmunds (2021, now impaired).

Change in accounting policies? (Fact) No major change; CECL and ASC 842 in effect. The off-balance-sheet securitization structuring (Sept 2025) is a funding/structuring choice to monitor.

Recent changes — new markets, facilities, management? (Fact) The dominant change is leadership: CEO Bill Nash terminated (Nov 2025), Keith Barr installed (March 2026). Store growth has slowed to ~4 openings planned for FY27; capex cut to ~$400M; CAF strategy shifting down-spectrum.

Appendix B — Source Appendix

Primary sources prioritized over secondary; recent over stale. All quantitative figures reconciled to the FY2026 10-K. Accessed June 7, 2026 unless noted. Third-party aggregator data and sell-side targets are signals only.

A. SEC Filings (primary — SEC EDGAR, CIK 0001170010)

B. Company Disclosures & Transcripts

• CarMax FY2026 Annual Report (CEO letter + 10-K) — public via SEC EDGAR / company IR.
• CarMax Q4 FY2026 earnings call transcript, 2026-04-14 (public via Motley Fool) — management framing on comps, GPU, CAF, buyback pause, FY27 outlook. Treated as hypothesis, validated against filings.
• CarMax Investor Relations: https://investors.carmax.com

C. Competitor & Industry Sources

• Carvana Co. Q4/FY2025 results release, 2026-02-18: https://investors.carvana.com — 596,641 retail units (+43%), GPU, ADESA capacity build.
• Cox Automotive / Manheim Used Vehicle Value Index, May 2026 & mid-April 2026: https://www.coxautoinc.com/insights/ — index 212.6 (+3.6% y/y).
• Edmunds Q1-2026 Used Car Report; Presidio Group; Digital Dealer — off-lease supply trough (2025–2027) data.
• WardsAuto — Carvana 2025 profitability/units.

D. Auto-Finance & Credit Data

• Bankrate — subprime auto delinquency surge (~6.90% 60+ days, 32-year high), Q1 2026.
• Wolf Street — Q1 2026 auto-loan delinquency/balance data; subprime vs prime split.
• Cox Automotive — ~1.73M repossessions in 2025 (highest since 2009).

E. Regulatory

• Holland & Knight — Fifth Circuit vacatur of FTC CARS Rule (January 2025): https://www.hklaw.com
• Federal Register — FTC CARS Rule withdrawal, effective 2026-02-12: https://www.federalregister.gov/documents/2026/02/12/2026-02866/

F. Quantitative Helpers (cross-check; reconciled to filings)

• Public market-data aggregators — price/market cap/multiples/comps (KMX, CVNA, AN, LAD, PAG, GPI, SAH); own-history valuation percentiles (P/E 95th, P/B 7th, P/S 7th). Aggregated third-party data; reconciled to the 10-K.
• SEC EDGAR XBRL — filing enumeration and concept verification.

G. Analytical Frameworks

• Greenwald & Kahn, Competition Demystified (barriers to entry; moat-type taxonomy; market-share-stability and ROIC tests).
• Chancellor / Marathon Asset Management, Capital Returns (supply-side capital-cycle analysis; asset-growth discipline).