CPRT: More free cash, less operating cash

The 60-second read

Five things to know

1. Free cash flow and operating cash tell different stories. Capital spending fell $231.6M, more than offsetting a $195.3M decline in cash generated by operations.
2. The geographic split matters. US quarterly operating profit fell $45.3M. International profit improved only $1.6M, despite double-digit sales growth.
3. Repurchases cushion EPS, not the income statement. Annual cash repurchases were $1.633B; quarterly diluted average shares fell 4.7%, while net profit declined 17.4%.
4. The balance sheet remains a strength. Cash and held-to-maturity securities total $4.490B. Restricted and foreign balances mean this is not an immediately distributable dividend.
5. The quote already assumes improvement. Our cash model requires roughly 13.7% annual five-year growth to explain the after-hours price, versus 7% in our base model.

Three scenarios, one probability-weighted number

Judgmental 12-month outcomes, not statistical probabilities or company guidance.

Signposts: what would change our mind

These are analyst monitoring thresholds, not management forecasts. Expected reporting months are estimates. The next packet should grade these same tests before revising the thesis.

The tape

An after-hours rebound is different from the regular-session move.

The saved after-hours trade was$33.61 at 20:34:35 UTC on September 10, compared with the regular-session close of $30.75. The computed change is+9.3%. This is one thin-session observation, not a final next-day return or a price at which every investor could transact. The earlier trading-day decline belongs to a different window and is not presented as the market’s reaction to these results.

The chart uses completed daily observations; the after-hours trade is described separately. That avoids drawing a smooth line through unlike price series. Historical earnings-window returns below use the first following trading session versus the release-date close. They are a volatility reference, not proof that the earnings release caused every part of the move. No completed following session exists yet for the current report.

The print

A small revenue beat does not erase the profit deterioration.

Quarterly sales were$1,152.4M, up2.4%. The saved vendor comparison is $1,143.743M revenue and $0.3832 EPS expected, versus reported $1,152.439M and $0.35. That is a computed0.8% revenue beat and-8.7% EPS miss. The release reports GAAP EPS; we do not invent an adjusted measure or call an unverified consensus figure a formal company target.

Services and purchased vehicles have different revenue recognition. A consigned vehicle principally produces service fees; a vehicle bought and resold can produce a much larger reported revenue amount while earning a modest spread. Therefore a better consolidated sales line is not sufficient evidence of stronger auction economics. We focus on service revenue, facility costs and the resulting operating margin together.

Annual revenue grew only 0.4%, against a materially stronger FY2025 comparison. This quarter should be read within that broader slowdown. We neither annualize the seasonally weaker fourth quarter into a permanent earnings run rate nor treat the full-year average as evidence that the latest margin compression is harmless. Both views would discard relevant information.

Segments and operating drivers

International sales growth has not fully replaced US profit.

The United States produced $930.334M quarterly revenue and $312.151M operating profit. A year earlier those amounts were $926.263M and $357.468M. International revenue increased to $222.105M from $198.834M, but operating profit rose only to $56.753M from $55.123M. International growth offsets some sales weakness, while the much larger US operation still determines most of the group’s earnings.

The latest detailed 10-Q attributed nine-month US service weakness partly to the unusually strong prior-year hurricane comparison, while revenue per car provided an offset. It also identified foreign-exchange benefits in international results. Those explanations belong to the nine months ended April 30. The current release does not quantify a separate fourth-quarter vehicle-volume, pricing and currency bridge, so we do not transplant the interim percentages into this quarter.

The network connects sellers with a global buyer base, and physical sites provide storage, processing and logistics that a website alone cannot replace. That combination supports the long-term case. It also carries fixed costs: when volumes soften, site expenses do not fall proportionately. International expansion can deepen the network while initially diluting margins. We therefore want growth that converts into profit, rather than simply more countries or a larger revenue total.

The profit bridge

Operating pressure and lower investment income both matter.

The $68.9M decline in attributable earnings includes $43.7M lower operating profit,$10.2M less net interest income and $19.8M less other income. Lower tax expense offsets $5.2M; the minority-interest change accounts for the remainder. Calling the entire decline a tax problem would miss the operating deterioration. Calling it entirely operational would ignore a meaningful loss of below-the-line support.

Quarterly diluted average shares decreased from 977.778M to 932.140M. Repurchases therefore softened the EPS decline relative to the decline in attributable earnings. That is a real per-share benefit, but it does not mean the auction business earned more. We also avoid multiplying annual EPS by the current share count to reconstruct net income: annual EPS uses its own weighted-average denominator and is rounded.

For the full year, operating profit fell $44.1M and attributable income fell $68.2M. The detailed source dataset retains every interest, other-income, tax and minority line rather than reducing the explanation to one adjusted percentage. No separate issuer non-GAAP earnings reconciliation is supplied in this release; depreciation and stock compensation remain visible operating costs.

Earnings quality

Profitable and cash-generative, with a weaker operating trend.

Operating cash exceeded consolidated net income for the year, and stock compensation is relatively small as a share of sales. Those are strengths. They coexist with weaker cash generation in dollars, rising receivables and a lower operating margin. A scorecard should preserve that tension rather than assign a clean bill of health because one cash-conversion ratio exceeds 100%.

Copart advances cash and incurs pooling costs before some vehicles are sold. Receivables therefore do not represent only conventional billed sales awaiting collection. The days metric is an analytical warning light, not proof of loose credit or fabricated revenue. Likewise, using total revenue to calculate vehicle-inventory days would mix a fee business with purchased vehicles. The denominator here uses vehicle cost specifically.

The FY2025 annual report records effective internal controls and an Ernst&Young audit opinion. That reporting evidence is separate from the ongoing DOJ investigation into anti-money-laundering practices. We do not portray the investigation as a restatement or assume a fine that has not been quantified. It remains a legal and operating risk that can affect future cash flows and the required valuation discount.

Cash and capital allocation

The free-cash improvement came from reinvestment, not more operating cash.

Operating cash fell from $1,799.750M to $1,604.492M. Cash purchases of property and equipment fell from $568.990M to $337.363M. The result is $1,267.129M free cash flow, up $36.369M. The entire increase is explained by a larger capex reduction than the operating-cash decline. That is better cash retention today, but it does not establish a stronger recurring earnings engine.

Capital spending includes land, facility improvements, equipment and software. Some outlays create capacity for years, so a lower year can follow a period of expansion without damaging the business. Conversely, permanently projecting a low capex year can overstate distributable cash if sites later need more maintenance or capacity. The model uses actual current cash generation and makes its growth assumption explicit; it does not assume all prior investment was unnecessary.

The cash pile is substantial, but it should not be counted twice. Our operating cash valuation removes estimated after-tax interest income before valuing the business, then adds the balance-sheet cash and securities less redeemable minority interests. Lease costs already reduce operating cash, so the model does not subtract the full operating-lease liability again. Foreign cash and potential repatriation friction remain qualifications; the resulting bridge is not a claim that every dollar can immediately be paid to US shareholders.

Valuation

The rebound requires a cash-growth assumption, not just a quality label.

Using the saved after-hours price and 932.140M diluted-average shares gives modeled equity value$31,329.5M. Deducting the$4,473.2M cash/minority bridge leaves$26,856.3M operating value, or16.3times FY2026 EBIT. This denominator is a conservative share proxy, not the vendor’s smaller point-in-time market capitalization. The same share basis is used across all three routes.

The 50%/25%/25% weighting produces$28.85, rounded to a $29 base with a $26–$32 range. The methods share operating assumptions and are not three independent pieces of market evidence. In the DCF,75.2% of operating value comes from the terminal period. That dependence is why the discount rate, reinvestment quality and durability of growth deserve more attention than an extra decimal place in the target.

Holding the 9.5% discount rate and 3% terminal assumption fixed, the market price requires about13.7% annual growth in normalized operating cash for five years. Our base uses 7%. This reverse calculation is not a forecast that growth will be 13.7%; it tells us what must be justified under this particular model. Higher margins, faster fee growth or more efficient capital use could deliver it. A smaller discount rate could also support the price, but that is a different risk judgment.

The SELL 3/5 call is a valuation judgment at the stated quote, not a claim that Copart has lost its moat. High uncertainty and a modest discount to the trading price keep conviction moderate. A price below roughly $25 with stable fundamentals would warrant a fresh review; improving service growth, cash generation and operating margin could instead raise fair value. A lower price caused by new legal restrictions would not automatically become an attractive entry.

Wall Street context

Freshly fetched does not mean freshly revised after earnings.

The maintained Street snapshot contains 12Buy,8Hold and 1Sell ratings, a $36.67 mean target and a $25–$45 range. The aggregate lacks a reliable update date for every underlying opinion. We therefore label it an undated vendor aggregate, not 21analysts who reacted to this release. Our current $29 value is more cautious, but the comparison is not evidence that those analysts reviewed the same numbers or used the same valuation horizon.

The annual estimate feed contains FY2027 average revenue of about $4.829B and EPS $1.6752, with nine EPS contributors. FY2028 has about $5.085B revenue and EPS $1.81729 from five contributors. These are vendor consensus observations, not company guidance. There is no saved sequence of comparable post-print snapshots from which to infer an estimate-revision trend, and the current release supplies no numerical next-year guidance to place beside them.

The conflicting target item is a useful reminder that metadata can be wrong even when the price field looks precise. Our packet retains the raw record for review while avoiding a confident broker attribution. Similarly, the profile’s employee and CEO fields conflict with dated company filings. Primary records govern those identity facts; a freshly downloaded aggregator field does not outrank them.

Management and governance

Capital allocation needs the same scrutiny as operating execution.

Jane Pocock was appointed President effective August 1,2026 after leading the UK operation. The August board announcement quotes Jay Adair as Chief Executive Officer. The vendor profile still names an earlier CEO, so it is not used as current leadership evidence. A leadership transition can bring useful operating discipline, but it does not itself establish that US margins or capital returns will improve.

The FY2025 annual report says no single customer accounted for more than 10% of revenue, yet insurers supplied 81% of processed vehicles. Customer diversification therefore does not remove industry concentration. Changes in insurers’ repair-versus-total-loss decisions, claim volumes, salvage prices and transport economics can affect many sellers at once. The physical network and buyer liquidity provide protection, but they do not make the business independent of those drivers.

• How much of current US weakness is volume, fee mix and prior catastrophe comparison?
• Which capex reductions reflect completed capacity rather than deferred work?
• What return threshold governs repurchases against new sites and technology?
• What explains international G&A growth relative to incremental profit?
• What concrete control milestones can management disclose about the DOJ matter without prejudicing proceedings?

These questions are designed to separate measurable operating changes from reassurance. We do not award management a guidance beat for a quarter with no numerical guide. The relevant test is whether subsequent filings show better revenue conversion, cash generation and returns on retained capital. The dated signposts above preserve that accountability for the next episode.

Risks and the opposing case

A durable franchise can still be an expensive stock.

The strongest bull case is that the current slowdown is a normal digestion period after unusually strong demand, catastrophe activity and investment. A network of sites, seller relationships and global buyers is difficult to reproduce quickly. If the US fee business stabilizes and international locations mature, a high incremental margin on renewed volume could lift earnings without proportionate capital spending. The cash balance gives management flexibility to invest through a weak period.

The strongest bear case is that investors extrapolate historical quality while the cost base becomes less flexible, reported revenue benefits from low-margin vehicle sales and cash flow is supported by reduced investment. Buybacks can improve EPS optics while reducing the cash cushion, and international growth may require more overhead than expected. An unresolved regulatory issue adds a risk that a conventional earnings multiple cannot quantify precisely.

We land between a broken-business claim and an automatic buy-the-dip story. Current profitability and liquidity deserve value. The after-hours price nevertheless requires more cash growth than our base assumptions deliver. That disagreement can be resolved by future evidence: service growth, operating-margin recovery, recurring cash conversion and disciplined capital allocation. The scenarios show both upside and downside rather than disguising uncertainty in a single target.

Catalysts and the next review

The useful catalysts are evidence releases, not promised price moves.

This packet uses the earnings release and already filed periodic reports. The announced call occurs later than the saved research snapshot; no transcript or management answer from that call is invented. If the call supplies material new guidance or a quantified operating bridge, it belongs in a clearly dated update, not retroactively in the current source set. The same applies when the FY2026 annual report arrives.

At the next quarter, we will compare the new data with the signposts before changing the call. A rising stock is not by itself evidence that the valuation assumptions improved, and a falling stock is not by itself evidence that the business weakened. The questions are whether recurring cash earning power changed, whether risk changed, and what price now compensates for both. Keeping those separate makes the quarterly research expandable and testable.

Data, sources and methods

The saved model keeps the calculation trail visible.

The dataset contains all source statement rows as well as standardized metrics. Quarterly cash flow is computed by subtracting the preceding year-to-date figure; fourth-quarter values use the annual less nine-month total. FY2024 and FY2025 fourth-quarter geographic results use the same annual-minus-nine-month method. Calculated sums reconcile to the annual statements. Differences in terminology such as yard versus facility operations are preserved and mapped explicitly.

Dollar financial amounts are in millions unless a table says per share. Shares are in millions for valuation calculations. FCF means operating cash less cash purchases of property and equipment; it is not the company’s net change in cash. Operating owner cash further deducts economic SBC and estimated after-tax interest to avoid valuing cash twice. Receivable and inventory days are ending-balance proxies with 91-day denominators, not audited operating KPIs.

The valuation is an analyst model rather than a company forecast. Growth, discount rate, tax normalization, multiples and scenario probabilities are stated assumptions. No absolute fourth-quarter unit or pricing measure is invented where the release does not provide one. No legal-loss estimate, post-print analyst revision trend or FY2026 employee count is invented either. The complete saved source index records retrieval times and SHA256 checksums so the evidence can be rechecked.