Business

Know the Business — Getlink SE (GET)

Getlink is a single-asset European concession that masquerades as a transport company. Roughly three-quarters of revenue and ~78% of EBITDA come from the Channel Tunnel, where shuttle pricing is dynamic but rail tolls are formula-bound to inflation under a contract running to 2052, and the underlying concession runs to 2086. The market debate that matters is not "ferry vs tunnel" — it is whether a 16x EV/EBITDA premium concession is paying for the 60-year monopoly tail or for the noisy near-term EBITDA from a still-ramping electricity interconnector.

1. How This Business Actually Works

Getlink owns one asset built once for ~€20bn and rents it out three different ways for the next 60 years.

The cost structure is fixed. Once the Tunnel is built and crewed, the next truck, the next Eurostar passenger, the next megawatt-hour is nearly free at the margin.

Eleclink's 70% segment EBITDA margin (excluding the €55m insurance compensation that was 46%) is the merchant-interconnector arithmetic — once the cable is live, the differential between French nuclear-led and British gas-led power prices drops straight to the bottom line, partially clawed back through a regulator-imposed profit-sharing provision (€516m booked at year-end 2025). Europorte's 19.9% margin is a reminder that being in transport without a concession looks very different.

The economic engine is therefore not "shuttles plus a few extras." It is a regulated infrastructure annuity (rail tolls, ~€411m/yr, formula-bound to 2052) wrapped around a contestable shuttle business that uses spare capacity — and bolted onto a merchant electricity asset that swings group EBITDA by ±€100m on power-price normalisation. Path occupancy at 45.6% means the asset has roughly 2x its current revenue capacity locked inside without major new capex; ETCS/ATO signalling could lift design throughput from 20 to 24 paths/hour. Operating leverage on incremental volume is 60–80% — every extra truck slot or rail toll euro flows almost directly to EBITDA.

2. The Playing Field

Getlink has no public peer that does the same thing. The best comparison set blends single-asset concession monopolies (AENA, ADP) for the regulated cash-flow shape and diversified concession-construction groups (VINCI, Eiffage) for the multiple debate, with one ferry operator (DFDS) for the contestable-transport overlay.

Three things the peer table reveals:

Getlink trades at a structural premium to airport monopolies. AENA earns 63.8% margins on five times the revenue and 4.5x the EBITDA, yet trades at 9.6x EV/EBITDA versus Getlink's 16.7x. The premium is paying for two things: concession length (2086 vs airport regulatory cycles every 5 years), and the takeover bid embedded in Mundys' creeping stake. Strip out the M&A optionality and the multiple looks demanding.

The "good" Getlink looks like AENA, not VINCI. AENA's combination of 60%+ margin, 17.8% ROCE, 1.2x leverage, and single-country regulated monopoly is the operating model Getlink should converge to as Eleclink stabilises and Eurotunnel volumes recover. Getlink's ROCE of 7.4% is held down by a still-amortising asset base (~€6.6bn net PPE on €20bn historical capex) and concession-related leverage that no airport carries.

DFDS confirms where the rents are. Cross-Channel ferries operating against a fixed-link monopoly print mid-single-digit margins, 5% ROCE, and 11x leverage. The Brexit-era ferry-capacity additions and Seafarers Wages enforcement are existential for them; for Getlink they are tactical pricing pressures.

3. Is This Business Cyclical?

Yes — but on three different clocks. The rail-toll annuity is acyclical (formula-bound). Shuttle services track UK-EU goods trade and consumer travel. Eleclink follows GB-FR power spreads on a one-to-three-year auction lag. Visible cycles get masked by infrequent shocks.

Three regimes are visible. Pre-2020 plateau (revenue ~€1.0–1.1bn, op profit ~€350–410m) — Eurotunnel-only era, steady volumes, no Eleclink. COVID-Brexit trough (2020–21) — revenue collapsed 28% in 2020 and operating profit dropped 67%, then stayed depressed for a second year as Brexit border frictions compounded the demand shock. Post-2022 step-change — Eleclink commissioning in May 2022 plus the European energy crisis added €350–600m of revenue in a year and re-rated EBITDA.

The non-obvious point: the asset has never been priced under capacity stress. Path occupancy is 45.6% — half the design throughput. So when the truck market contracts 2.4% (as in 2025), Getlink is not absorbing fixed-cost dilution on tight capacity; it is leaving slots empty. The cycle hurts pricing on the truck side and Eleclink margins, but rarely impairs unit economics structurally.

4. The Metrics That Actually Matter

Five numbers explain ~80% of value creation here. Most of the conventional ratio set is noise.

Path occupancy is the closest single number to a fundamental for the Tunnel. At 45.6%, the asset has a ~2x revenue call option locked inside without new construction — but the call only gets exercised if Eurostar competitors actually deliver trains in 2030–2031 and rail freight rebuilds.

Eleclink capacity sold forward is the one number that tells you EBITDA volatility 18 months early. Auctions happen monthly through the Joint Allocation Office; 79% of 2026 already sold by end-2025 means the 2026 print has narrow downside.

What the conventional dashboard misses: ROCE looks pedestrian (7.4%) because the denominator is a fully-depreciated 30-year-old asset re-inflated by lease accounting and IFRS 16. EPS looks lumpy because of Eleclink profit-sharing provisions (€516m balance) and inflation indexation on debt. Use unit-level operating metrics first; whole-group ratios second.

5. What Is This Business Worth?

The right lens is sum-of-the-parts on three economically distinct assets, not a consolidated EV/EBITDA multiple. Getlink reports as one business; it is valued as three.

The Eurotunnel core is a regulated infrastructure concession that should trade on long-duration concession yield with a multiple anchored to AENA/ADP and a duration premium for the 2086 tail. Eleclink is a merchant-style power-infrastructure asset whose rents follow auction-cleared GB-FR spreads — multiple should reflect cash-flow visibility in the next 2–3 years, not a steady-state assumption. Europorte is a sub-scale rail-freight operator that probably trades on book value or modest EV/EBITDA.

The SOTP discipline matters because consolidated multiples blur three different cash-flow shapes. A reader who anchors to "16x EV/EBITDA looks expensive vs AENA at 9.6x" without separating Eleclink ends up wrong twice — first because Eleclink's near-term EBITDA is artificially low after the cable outage and energy normalisation, and second because the concession tail to 2086 mathematically deserves a duration premium over airport regulatory horizons that reset every 5 years.

The catalyst the multiple is paying for: Mundys (ex-Atlantia, taken private 2023) holds 19% rising to 25% via tranched share-swap. Crossing 30% triggers AMF mandatory bid rules. The market is partially pricing a takeout, which is a reason for the premium — but also a reason to be cautious about extrapolating it.

What would make the stock cheap or expensive:

• Cheap: Eurostar volume disappoints into 2030; truck market contracts another 5%+; Eleclink cable fault repeats; Mundys settles at 25% with no bid; rates rise into a refinancing window.
• Expensive: Path occupancy moves toward 60% as Virgin/Trenitalia/Evolyn launch HSR services; truck market stabilises and Shuttle pricing recovers; Eleclink locks 80%+ of 2027–2028 at firm spreads; Mundys triggers a mandatory bid.

6. What I'd Tell a Young Analyst

This is not a transport stock. It is a single-asset European infrastructure concession with a takeover overhang and a noisy power-trading bolt-on. Forget the ferry comparison set; live with AENA, ADP, and motorway peers in your head, and treat ROCE/ROE less seriously than path occupancy and forward Eleclink auction prices.

Three things to watch first. Path occupancy in the annual URD — it is the lock on the asset's growth optionality and almost no one tracks it. Mundys' stake and any 13D-equivalent filings — control creep is the entire near-term re-rating story. Eleclink monthly auction clearing prices and forward sold percentages at the Joint Allocation Office — leading indicator on the most volatile P&L line by 12–18 months.

Three traps to avoid. Do not extrapolate the 2022–23 EBITDA peak; that was a once-in-a-generation energy crisis flattering Eleclink. Do not read the 51% group EBITDA margin as the steady-state — it is a blend of a 56% concession asset and a 70% merchant asset that will normalise toward the low-50s. Do not trust the consolidated EV/EBITDA versus airport peers without doing the SOTP — the duration premium is real and material, but only if the 2086 concession is not revisited and Mundys' ownership intentions are friendly.

The thesis-changing observation would be a meaningful loss of truck market share to ferries (below 33%) on a flat or growing market, or any sign that the rail-toll formula gets renegotiated. Either would attack the regulated annuity that the multiple ultimately rests on.