Variant Perception

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Where We Disagree With the Market

The market is treating the FY2025 EBITDA beat as validation of the path to $1.18B/2030 — and a $22.35 median target only $0.56 above spot says consensus is comfortable enough to roll the position forward. The forensic file says the "beat" was a $65M ElecLink insurance reclassification with $6M of actual cash collected, and that ex-insurance EBITDA went from a restated $865M (FY2024) to $966M (FY2025) on a USD basis — but minus 1% on a like-for-like native basis. The same multiple is partly paying for take-out optionality from Eiffage that the largest blockholder has explicitly disavowed in writing to the AMF. We disagree with consensus on three measurable fronts: the quality of the FY2025 number, the value of the corporate-action premium embedded in the multiple, and the classification of ElecLink as infrastructure when its revenue trajectory and regulatory tail are merchant in nature. The 23 July 2026 H1 trading update resolves the first; AMF threshold-crossing filings between now and FY2026 results resolve the second; the H2 2026 ElecLink forward-2027 capacity disclosure plus CRE / National Grid / RTE rule finalisation resolves the third.

1. Variant Perception Scorecard

The variant view is specific and testable, not contrarian. Consensus clarity is high on three of the four disagreements: the median sell-side target ($22.35) and the mixed Buy/Hold rating distribution show what the market is paying for, the multiple itself (~16x EV/EBITDA on reported FY2025 EBITDA of $1,009M) shows what duration and optionality consensus is underwriting, and the JPMorgan September 2025 downgrade to Neutral (rising yields cited) shows where conviction has already eroded. Evidence strength rests on filings rather than estimates — the $65M insurance recovery, the $966M ex-insurance EBITDA, the $606M IAS-37 provision growing $129M YoY, the AMF disavowal letter, and Eurostar's "2029-2030 at the earliest" public commentary. The first decisive resolving event sits roughly three months out (the H1 2026 trading update on or around 23 July 2026); the second runs continuously through the FY2026 audit cycle.

2. Consensus Map

What the market appears to believe today, anchored to observable signals — not what we wish the market believed.

The honest reading of consensus is that the market is paying for a long-duration, treaty-protected concession with two embedded options — capacity (path occupancy) and corporate action (Eiffage/Mundys). Neither option is being challenged inside the $22.35 PT. Our disagreement is not with the moat itself — Eurotunnel's 78% of group EBITDA, 53.9% margin, and rights to 2086 are not in dispute — it is with what the equity is being asked to underwrite on top of the moat at the 16x multiple.

3. The Disagreement Ledger

Disagreement #1 — the FY2025 print was a reclassification. Consensus would respond that the company published both numbers, the auditor signed off unqualified, and FY2026 guidance accommodates a normalised base. Our evidence is that the published reconciliation is the disagreement: at $966M ex-insurance, FY2025 underlying went backwards versus the $865M restated FY2024 base in native terms (the USD trajectory is distorted by the FX appreciation between period-end 2024 and period-end 2025), and the FY2026 mid-guide of $988M does not bend the path up on the like-for-like base. If we are right, consensus would have to mark the multiple to underlying $964–988M EBITDA rather than reported $1,009M, and the $0.06/yr dividend escalator becomes the binding constraint. The cleanest disconfirming signal is an H1 2026 EBITDA print on ~23 July 2026 above $506M with ElecLink revenue sustaining and no further insurance one-off — that would close the case for consensus.

Disagreement #2 — the take-out premium is unwinding. Consensus would note that strategic blockholders move on their own timetable and that Mundys exercising to 25% is itself a stake-up move. Our evidence is the direction of disclosure: Eiffage's October 2025 AMF letter is a written disavowal of control intent, the October block at $20.80 carries an 18-month price-protection clause that effectively penalises Eiffage for buying lower than $20.80 on-market over the next 18 months, and Mundys' CEO already sits on the board — strategic minorities buy when they want governance influence, not when they intend to bid in 12-24 months. The clean disconfirming signal is either an AMF crossing above 30% or coordinated concert-party language; neither has appeared.

Disagreement #3 — ElecLink is a merchant book. Consensus would object that ElecLink revenue is up sharply in Q1 2026 and forward-sold cover is high. Our evidence is the two-year revenue trajectory ($579M → $264M, -57%) and the asymmetry of the IAS-37 provision (growing $129M YoY against undefined rules). A 70% segment EBITDA margin in a corridor with documented cable-condition risk and an open regulatory cap is not infrastructure cash flow. The clean disconfirming signal is forward-2027 sales pushing past 50% by the Q3 2026 traffic release and the regulator finalising rules at or below $606M without an auditor flag step-up.

Disagreement #4 — the dividend yield is mispriced as defensive. Consensus would point to investment-grade ratings, the March 2025 green-bond refi, and a 60-year concession horizon. Our evidence is that the company's own conservative FCF definition delivers 0.87x cover, the dividend is committed to grow $0.06/yr against an EBITDA line that is flat ex-insurance, and ND/EBITDA at 4.2x leaves limited cushion. The clean disconfirming signal is the FY2026 cash-flow statement reconciling the $59M insurance receivable, capex inside $200–259M, and CFO above $1,000M.

4. Evidence That Changes the Odds

These are not generic facts; each one moves the probability of the variant view materially.

The $966M ex-insurance underlying versus $865M restated FY2024 (-1% LFL in native currency, the USD step is FX-distorted) is the single fact that does the most work in this file. It is not a complicated number — the company itself published it — but it inverts the headline interpretation of the FY2025 result. Every other piece of evidence is a corollary: ElecLink's two-year revenue decline ($579M → $264M) is the operational driver of the underlying flatness; the IAS-37 provision is the regulatory tail that caps the recovery; the AMF disavowal is the corporate-action backstop being unwound; the FY2026 guide of $964-1,011M is management's own forecast that the line stays flat in the year before the bull case has to inflect.

The chart is the disagreement in one frame. Native LFL went backwards from FY2024 to FY2025; in USD the optical step up is mostly EUR appreciation between period-end 2024 (FX 1.04) and period-end 2025 (FX 1.18). Management is guiding FY2026 to $988M, which on a like-for-like basis is flat to down. The amber insurance bar in 2025 is what the headline $1,009M is "pricing in." The $1.18B/2030 target is the bar consensus is paying for; the underlying line has to add $209M of EBITDA in five years from a base that just printed minus 1% LFL.

5. How This Gets Resolved

The resolution path is concentrated rather than diffuse. Two signals (the H1 2026 print and AMF threshold filings) resolve roughly 70% of the variant view inside three months. The remaining signals (ElecLink forward sales, profit-share rules, dividend cover) work over the FY2026 audit cycle and FY2026 results in February 2027. There is no scenario in which all four disagreements remain unresolved 18 months from now.

6. What Would Make Us Wrong

The cleanest way the variant view breaks is if the FY2025 ex-insurance underlying was simply the trough of an in-progress mix shift, not a stalled base. Eurotunnel core EBITDA at $784M in FY2025 (+5% YoY native) is consistent with that reading; ElecLink at full-tilt Q1 2026 ($82M, +112%) suggests the merchant cable is rehabilitating faster than the consolidated FY2025 line implied. If Q2 ElecLink revenue follows the Q1 trajectory, if forward-2027 sales close the cliff at or above 50% by the Q3 2026 release, and if the Eurotunnel core compounds at the +5% it printed in FY2025, the H1 2026 print could land at $506M+ ex-insurance — at which point the FY2026 mid-guide of $988M becomes a conservative anchor and the bridge to $1.18B/2030 looks visible. We would also be wrong on disagreement #2 if Eiffage or Mundys cross 30% and tender at a meaningful premium — the price-protection cap on the October block does not preclude a friendly negotiated transaction outside the on-market route, and Mundys is not bound by Eiffage's October representations.

The dividend disagreement breaks if the $59M insurance receivable converts in cash and ElecLink CFO contribution stabilises — the FY2025 cover gap was driven primarily by ElecLink's revenue decline ($114M of the $364M three-year CFO drop) plus the timing of the insurance recovery. A clean FY2026 cash-flow statement closes that gap mechanically. And on ElecLink classification: if the CRE finalises rules at or below $606M and forward-2027 sales recover to historical patterns, the merchant-interconnector multiple framing weakens and the consolidated 16x becomes defensible.

The serious red-team point is that the variant view depends heavily on the H1 2026 print being clean. If the H1 release adds further insurance items, contingent revenue from the Passenger Shuttle arbitration, or other one-time items that look "real" but are timing, the same forensic logic that supports our view for FY2025 could be deployed against the FY2026 underlying — a mirror trap that consensus could fall into in the opposite direction. Stan's verdict is "Watchlist" for a reason; the moat is durable, the cash flows are real, and the multiple sits inside a defensible range for the assets owned. The variant view is "not yet earned" not "not real."

The first thing to watch is the H1 2026 EBITDA print on or around 23 July 2026 — like-for-like ex-insurance vs the $988M FY2026 mid-guide.