CSX: A Fine Railroad, Priced for a Takeover That Hasn't Happened
Opening Remarks
CSX Corporation is an eastern Class I railroad — the kind of business Warren Buffett actually bought when he acquired BNSF. The problem is not the business. The problem is the price tag, which appears to carry a premium for a merger that has not been approved and may never be. Walking the four gates makes this plain.
Gate 1: Circle of Competence
CSX operates a rail network across 23 eastern states, earning freight revenue from shippers of coal, chemicals, agricultural products, automobiles, and intermodal containers. The eastern freight rail market is effectively a duopoly between CSX and Norfolk Southern. Railroads hold structural fuel and labor cost advantages over trucks for long-haul, heavy freight, and building a new rail network is physically and regulatorily implausible — the routes themselves are irreplaceable. FY2025 revenue was $14.1 billion with operating income of $4.52 billion ($4.69 billion adjusted).
Can we know this business? Information gathering passes without difficulty: SEC filings, earnings releases, STB records, and industry press reconstruct it fully. Predictability passes conditionally. That freight will still move by rail in the eastern US in ten years is within the realm of analysis. What is not: the pending Union Pacific–Norfolk Southern megamerger (accepted for STB review on 2026-05-28), which, if approved, would redraw the eastern competitive map. The industry is knowable; the current restructuring moment involves an element of probabilistic betting.
Verdict: pass, conditionally.
Gate 2: Moat
The moat’s source is efficient scale — the same mechanism behind BNSF. The eastern market supports only two railroads, and entry is essentially impossible. A cost advantage over trucking provides secondary support.
On pricing power: Class I railroads have consistently raised core pricing above inflation over the past twenty years, and negotiation leverage is particularly strong in single-served territories. 2025 saw operating margin decline to 32.1%, but this looks like volume and coal mix deterioration rather than a crack in the moat itself.
Direction: holding steady, but with an exogenous risk attached. If the UP-NS merger is approved, the first single transcontinental railroad in US history would emerge, leaving CSX to compete without a western connection. Both BNSF (Berkshire) and CPKC publicly stated in August 2025 that they have no interest in acquiring CSX. The route monopoly is intact; its relative value may not be.
The AI-era reassessment leans favorable: physical infrastructure moats cannot be eroded by AI, and AI-driven optimization of scheduling and predictive maintenance should reinforce the cost advantage, with aggregate and construction-material volumes from data center building a mild tailwind.
Verdict: pass, with a merger-outcome observation condition attached.
Gate 3: Management
Share repurchases have been durable, with share count steadily declining; as of March 2026, $989 million remained under authorization, plus a newly approved $5 billion increase. Dividends have been raised consistently. The blemish is Quality Carriers — the trucking acquisition was confirmed a failure with a $164 million goodwill impairment in Q3 2025. Small in scale, but it is a documented failure at adjacent-business M&A.
The retained-earnings test broadly passes over five years (market cap growth exceeding retained earnings, including the shrinkage effect of buybacks), though the current market cap likely embeds merger premium, lowering the test’s reliability.
Candor and alignment carry a deduction: in September 2025 the board abruptly replaced CEO Hinrichs with Steve Angel, apparently in response to weak service metrics and the merger environment. There is continuity risk, but no record of capital destruction through repeated peak-buying or dilution.
Verdict: pass, with deductions.
Gate 4: Price
Owner earnings come to roughly $1.667 billion on a net income plus depreciation minus full capex approximation (FY2025: net income about $3.0 billion, D&A about $1.6 billion, capex $2.9 billion, FCF $1.7–1.8 billion). Since most railroad capex is maintenance in nature, treating FCF as a conservative proxy for owner earnings is reasonable. Normalizing for the coal cycle and service recovery, a normalized owner earnings range of $2.5–3.0 billion is adopted.
Applying 12x — appropriate for a low-growth enterprise — yields $30–36 billion. Even generously, at 15x on $3.0 billion of owner earnings assuming real growth, the ceiling is $45 billion.
The market disagrees. Current market cap is $91.8 billion at 30.5x earnings, implying an owner-earnings yield of 1.8–3.3% — well below the 10-year Treasury at 4.3%. There is no margin of safety. The prevailing price appears to embed a substantial “CSX will be acquired” premium tied to the UP-NS merger; the precedent is instructive — when BNSF and CPKC declared no acquisition interest in August 2025, the stock fell sharply. The target buy price is a market cap of $36 billion or below: roughly a 30% discount to the top of intrinsic value, about 60% below the current price.
Verdict: fail.
One data note: the screening feed’s debt-to-equity of 0.0 is an XBRL tagging artifact. Actual FY2025 figures show total debt of about $19.0 billion ($18.2 billion long-term) against $13.6 billion of equity — D/E of roughly 1.4, an industry-standard level for a regulated railroad with stable cash flows, with interest coverage near 5x. The 28.3% ROE should be understood as leverage-produced, not the product of an unlevered, high-return business.
The Verdict
Watch. Gates 1 through 3 pass — this is the genuine article as far as business quality goes, the same efficient-scale moat Buffett paid for at BNSF. Gate 4 fails decisively: a 1.8–3.3% owner-earnings yield against a 4.3% risk-free rate is not a valuation, it is a lottery ticket on being acquired. That is event speculation, not value investing, and the moat’s relative worth may be impaired if the transcontinental railroad comes into being. Reassessment triggers: the STB’s final merger decision (expected in the second half of 2026 through 2027), a merger-expectation collapse in the share price — the August 2025 episode is precedent — and a market cap approaching $40 billion warranting a precise revaluation. Monitoring metrics: intermodal volume share versus NS and trucking, operating margin recovery from 32% toward the 36% range of 2023–24, new CEO Angel’s first capital allocation decision (a large buyback at today’s price would be a further deduction), and whether coal’s revenue decline proves cyclical or permanent.
What Would Change the Verdict
A merger-expectation collapse pushing the market cap toward $36 billion — the buy threshold — with the core moat intact, would put this firmly on the table. Conversely, a conditioned STB approval that leaves CSX structurally exposed to a transcontinental rival, degrading the efficiency-of-scale moat itself, would end the watchlist entry entirely.
This analysis is AI-generated, educational, and not investment advice. Figures may contain errors or be delayed. Disclaimer