Report contents
At a glance
- Business model: CSX sells rail haulage to industrial shippers, utilities, and intermodal intermediaries under multi-year contracts and in the open market, earning roughly volume times price plus a fuel surcharge. Dense, hard-to-duplicate routes give real bargaining power on long, heavy hauls, but that power depends on service reliability, and in intermodal CSX prices only the middle linehaul link.
- Financial quality: Earnings convert to cash reliably, with operating cash flow exceeding net income by 1.35x to 1.60x every year since FY2021. The strain is below that line: free cash flow fell to USD 1.71bn in FY2025 as capital spending reached USD 2.90bn, about 21% of revenue, and leverage with roughly 5.5x interest coverage leaves a thinner cushion than FY2021.
Business model
I. CSX: a mature freight network whose pricing power runs through service
CSX Corporation (Nasdaq: CSX) is a freight railroad. It sells haulage: shippers hand over coal, chemicals, grain, automobiles, or a container of imported goods, and CSX moves it over its own track for a price per carload or per container. The paying customer is not a consumer. It is industrial shippers, utilities, automakers, agricultural exporters, and the intermediaries that assemble intermodal freight, buying transportation in bulk under multi-year contracts and in the open market. Revenue behaves roughly like volume times price, plus a fuel surcharge that passes diesel cost through with a lag.
That makes this a capital-intensive network infrastructure business — mature, sitting inside a cyclical demand envelope. The FY2025 Form 10-K, filed February 12, 2026, reports 1,859,556,799 shares outstanding as of January 31, 2026, and roughly $61 billion of non-affiliate market value at June 30, 2025.
Three freight pools with three different futures
Rail revenue is a portfolio of commodity pools whose demand drivers barely overlap. Coal burns because utilities and export buyers choose it. Intermodal containers move because imports arrive and trucks are tight. Merchandise rides industrial production; automotive rides vehicle build rates. The mix decides the outcome more than the total does, because each pool carries a different contribution margin and a different durability.
The excerpt supplied here is the filing's cover page and table of contents. It does not include segment revenue splits, carload tables, or the operating ratio — the railroad-specific measure of operating expenses divided by revenue, where lower is better. Any specific mix percentage would be invented. Item 1 and the MD&A hold those figures, and the first thing to establish is how much contribution margin sits in coal, which is structurally shrinking and sensitive to export prices, versus intermodal, which is the genuine growth pool and depends on port volumes and highway capacity.
In intermodal, CSX is the middle link
Merchandise and coal economics are largely CSX's to control, because the railroad owns the origin, the destination, and the only practical route. Intermodal is different. Door-to-door container economics split across the box owner, the drayage trucker at each end, and the railroad in between. CSX can price its linehaul, but it cannot capture the whole chain, and its share of the customer's total cost sets a ceiling on rate increases. That is a structural cap on pricing power in the fastest-growing pool, and it is worth tracking separately from the commodity business.
Pricing power is rented from service
A dense, hard-to-duplicate network does create real bargaining power on long, heavy hauls where the alternative is a truck. That power is not unconditional. It is rented from reliability: if transit times stretch, the shipper pays up for trucking and the rate increase stops being collectible.
The tradeoff is price versus volume, and fixed costs sharpen it. Track, crews, terminals, and locomotives are paid for whether or not a particular carload moves, so each incremental carload carries high contribution margin. Discount to defend volume and margin leaks on every unit. Hold price and lose volume, and contribution disappears while little cost does. Because so much of the cost base is fixed, a modest volume decline can erase the benefit of a solid rate gain. Revenue per carload next to carload counts is the evidence that resolves this — not either figure alone.
Growth needs capital committed years early
The binding constraint is timing. Converting truck freight to rail requires terminal capacity and network slots, which take years of permitting, capital, and hiring before the first incremental carload shows up. Expansion versus execution is the live tension: money leaves today against volume that may not arrive, and the FY2025 filing carries a dedicated labor agreements section because crew availability limits how much added capacity actually moves.
Fragility follows from the same structure. A railroad is one connected system, so a single congested terminal or a weather event degrades lanes that had nothing to do with it. Add concentrated end markets, large individual shippers, and Surface Transportation Board oversight of rates and service, and the boundaries of pricing power become visible.
What would overturn this read: revenue per carload falling while volumes hold, which would signal a price war rather than a service advantage; or volumes rising while service metrics deteriorate, which would signal growth bought with a network that cannot sustain it. Both are observable over time. Neither is settled by the material available here.
Questions worth following
- Can CSX keep raising real prices without shedding carloads to trucks?
Pricing is the main lever a mature railroad controls, and its cost base is largely fixed. If rate gains arrive alongside proportional volume losses, the contribution margin given up on each abandoned carload outweighs the increase, and earnings fall even as the headline rate rises.
- Is CSX's volume growth a structural share gain or a cyclical rebound in industrial freight?
Cyclical volume recovers on its own and reverses; structural share gain persists through a downturn and justifies committed capital. The distinction changes how much of current volume growth can reasonably be capitalized into normalized earnings and how much capacity spending the network can support.
- Can CSX add intermodal and terminal capacity without degrading network service?
Capacity additions are how the railroad converts truck freight into recurring volume, but they consume years and capital before revenue arrives. If construction and heavier traffic slow the network, service-driven pricing power erodes while the capital stays committed either way.
Financial quality
Financial quality: cash converts reliably, but each dollar of capital earns less
CSX is a railroad, so the accounting is unusually easy to read. Revenue is freight moved, the biggest operating costs are labor and fuel, and the largest non-cash charge is depreciation on track, locomotives, and terminals. That structure produces dependable cash conversion: operating cash flow has exceeded net income by 1.35x to 1.60x every year since FY2021, and nothing in the supplied record suggests accrual games or working-capital stuffing. The earnings are recurring, and they arrive in cash.
The weakness sits one line lower. Free cash flow fell from $3.49bn in FY2022 to $1.71bn in FY2025, while capital spending rose from $2.13bn to $2.90bn, roughly 21% of revenue. Returns moved the other way. ROA slid from 10.1% to 6.6%, and an ROIC estimate (EBIT taxed at the reported effective rate, divided by debt plus equity less cash) fell from about 16% to 11%. More capital, less return.
Five years of annuals: less freight, thinner margins, more spending
| USD bn unless noted | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | Q3 2026 (interim) | Assessment |
|---|---|---|---|---|---|---|---|
| Revenue | 12.52 | 14.85 | 14.66 | 14.54 | 14.09 | 4.02 | Contracting, now recovering |
| Operating income | 5.16 | 5.95 | 5.47 | 5.37 | 4.52 | 1.57 | Weak, improving |
| Operating margin | 41.2% | 40.1% | 37.3% | 36.9% | 32.1% | 39.1% | Weak, improving |
| Net income | 3.78 | 4.11 | 3.67 | 3.47 | 2.89 | 1.05 | Adequate |
| Net margin | 30.2% | 28.0% | 25.0% | 23.9% | 20.5% | 26.1% | Strong for a railroad |
| ROE | 28.4% | 31.9% | 29.8% | 28.3% | 22.0% | Not disclosed | Strong, leverage-assisted |
| ROA | 9.4% | 10.1% | 8.7% | 8.2% | 6.6% | Not disclosed | Adequate, declining |
| ROIC (estimate) | n/a | ~16% | ~15% | ~14% | ~11% | Not disclosed | Adequate, declining |
| Liabilities / assets | 66.7% | 69.9% | 71.6% | 70.8% | 69.9% | 68.5% | High, normal for rail |
| Current ratio | 1.73 | 1.56 | 1.04 | 0.86 | Not disclosed | Not disclosed | Thin |
| Operating cash flow | 5.10 | 5.62 | 5.51 | 5.25 | 4.61 | Not disclosed | Strong |
| Capital spending (OCF − FCF) | 1.79 | 2.13 | 2.25 | 2.53 | 2.90 | Not disclosed | Rising |
| Free cash flow | 3.31 | 3.49 | 3.26 | 2.72 | 1.71 | Not disclosed | Adequate, falling |
| OCF / net income | 1.35 | 1.37 | 1.50 | 1.51 | 1.60 | Not disclosed | Exceptional |
| FCF / net income | 0.88 | 0.85 | 0.89 | 0.78 | 0.59 | Not disclosed | Weak |
Revenue peaked at $14.85bn in FY2022 and declined every year after; FY2025 was 5% below that peak. Operating margin lost more than 800 basis points, from 41.2% to 32.1%. Some of that is cyclical — soft freight demand and weaker coal — and some looks structural, because intermodal and truck-competitive traffic carries lower margins and consumes more equipment capital than the coal and merchandise franchises it replaced. The reported gross margin line decays the same way (42.2% to about 33%), but railroads do not disclose gross profit in the usual sense, and the quarterly version does not reconcile with operating income (in FY2026 Q2 gross profit is reported as equal to revenue), so operating margin is the more trustworthy measure.
The equity return flatters the business. FY2024 ROE of 28.3% rides on liabilities equal to 70.8% of assets; ROA of 8.2% is the more honest figure. Railroads normally carry heavy debt against stable, contracted cash flows, so 69–72% is not alarming by itself. Coverage and liquidity are the parts worth watching.
Seven quarters, and the 2026 turn is real but partly fuel
| USD bn unless noted | Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | Q1 2026 | Q2 2026 | Q3 2026 | Assessment | Trend |
|---|---|---|---|---|---|---|---|---|
| Revenue | 3.42 | 3.57 | 3.59 | 3.51 | 3.48 | 3.94 | 4.02 | Adequate | Improving |
| Operating income | 1.04 | 1.28 | 1.09 | 1.11 | 1.25 | 1.51 | 1.57 | Strong | Improving |
| Operating margin | 30.4% | 35.9% | 30.4% | 31.6% | 35.9% | 38.3% | 39.1% | Strong | Improving |
| Net income | 0.65 | 0.83 | 0.69 | 0.72 | 0.81 | 1.00 | 1.05 | Strong | Improving |
| Total assets | 43.20 | 42.93 | 43.28 | 43.68 | 44.23 | 44.73 | Not disclosed | Not applicable | Stable |
| Total liabilities | 31.02 | 30.55 | 30.52 | 30.52 | 30.65 | 30.64 | Not disclosed | Not applicable | Stable |
| Cash | 1.14 | 0.39 | 0.61 | 0.67 | 0.96 | 1.01 | Not disclosed | Thin | Recovering |
| Operating cash flow | 1.26 | 0.64 | 1.34 | 1.39 | 1.27 | 1.33 | Not disclosed | Strong | Stable |
| Free cash flow (as supplied) | 0.54 | -0.14 | 2.83 | 0.71 | 0.73 | 0.75 | Not disclosed | Not reliable | — |
Eight quarters are not available; the record covers Q1 2025 through Q2 2026, and the newest point comes from the October 21, 2026 third-quarter release, which disclosed only headline figures. Fiscal quarters align with calendar quarters.
The trajectory is the clearest evidence in the file: operating margin went 30.4% → 35.9% → 38.3% → 39.1%, and latest-quarter volume rose 5% with intermodal up 7%. Two things temper it. Q3 2025 included a $164m non-cash goodwill impairment, so 54% EPS growth against that reported base is flattered; measured against the company's adjusted $818m, net earnings rose about 29%. Second, CSX attributes part of the 12% revenue increase to higher fuel surcharge revenue. Surcharges largely pass through fuel cost, adding dollars without adding much margin, and they can make percentage margins harder to read depending on the recovery lag. Volume and price are the durable components.
Leverage and liquidity: normal for a railroad, thinner than it was
Interest expense crept from $0.72bn in FY2021 to $0.84bn in FY2025 while EBIT fell from $5.67bn to $4.59bn — coverage of about 7.9x became 5.5x. Long-term debt has held near $18bn, but short-term debt doubled to $1.70bn by Q2 2026 from $0.71bn a year earlier, and cash of $1.01bn sits against $4.15bn of current liabilities. The current ratio fell from 1.73 in FY2021 to 0.86 in FY2024. Railroads deliberately run negative working capital — receivables turn in roughly 34–36 days — so a sub-1 current ratio is not a red flag by itself. Cash falling while distributions continue is.
That is broadly what happened. Financing outflows, which include dividends, buybacks, and debt repayment, exceeded free cash flow in four of the last five fiscal years (FY2021: $4.11bn out against $3.31bn of FCF; FY2024: $3.06bn against $2.72bn). Cash fell from $2.24bn at the end of FY2021 to $0.67bn at the end of FY2025. Equity did rise in FY2025 despite that, because earnings exceeded the outflow — but the cushion is thinner than it was.
What would change the read
Three checkpoints. Whether the capital spending is growth or maintenance: if capital intensity holds near 21% of revenue while volume grows, the marginal investment is earning its keep; if it keeps climbing on flat carloads, that is a treadmill. Whether the 2026 margin gains survive a fuel-price swing or a weaker pricing cycle, which fuel-surcharge and unit-cost disclosure would settle. And the goodwill line: the Q3 2025 impairment cut goodwill from $240m to $80m, then goodwill reappeared at $260m in Q1 2026 against $80m at year-end 2025 — worth understanding before assuming the writedown cycle is finished. I would also treat the supplied Q3 2025 free cash flow of $2.83bn as a data error; it does not reconcile with $1.34bn of operating cash flow and a $0.69bn investing outflow.
Questions worth following
- Can CSX turn heavier capital spending into better returns rather than lower free cash flow?
Capital spending has risen to about 21% of revenue while ROA fell from 10.1% to 6.6% and free cash flow halved from its FY2022 peak. The answer decides whether CSX is funding growth or simply paying more to move the same freight.
- Is CSX's 2026 margin recovery built on durable pricing and volume or on fuel surcharges?
Operating margin recovered from 30.4% in Q1 2025 to 39.1% in Q3 2026, and the company credits higher volume, pricing, and fuel surcharge revenue. Fuel surcharges pass through cost and add little margin, so durability depends on which component does the work.
- Can CSX cover its dividend and buybacks from free cash flow if capital spending stays near 20% of revenue?
Financing outflows exceeded free cash flow in four of the last five fiscal years and cash fell to $0.67bn at the end of FY2025, while short-term debt doubled to $1.70bn. If free cash flow stays near $1.71bn, the shortfall has to come from borrowing or the balance sheet.
Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.
Continue reading the full report
- Key risksWhat could break the investment case
- Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
- SummaryOur overall judgement and what to watch next
Free to sign up · 20 research points · unlocking this report uses 10
20 points cover 2 full reports. No card needed.
Research reflects its generation date. Preview the first two modules.
More on CSX Corporation
CSX Corporation
Business, moat and key questions
Open company page