Energy
PrivateA €50bn interest-rate-shaped bet on German grid-fee returns — the only DAX utility whose earnings compound directly with the RAB, now de-risked by the Dec-2025 WACC ruling but capped by it too; own it for the 8% EPS/dividend algorithm, not for a re-rating.
Research
The verdict
A €50bn interest-rate-shaped bet on German grid-fee returns — the only DAX utility whose earnings compound directly with the RAB, now de-risked by the Dec-2025 WACC ruling but capped by it too; own it for the 8% EPS/dividend algorithm, not for a re-rating.
What E.ON actually is: not a power company. After the 2019–2020 asset swap with RWE (closed Jul 2020), E.ON exited all generation — conventional, nuclear, and renewables — and became a pure regulated-network + energy-retail utility. It is the largest distribution-grid operator in Europe. The business is a regulated-utility toll road, not an energy producer: it owns the wires and pipes that everyone else's electrons and molecules must travel through, and earns a regulator-set return on the capital sunk into them.
Three reporting segments:
Customers: ~50m across 30+ countries (headline figure); the economically-meaningful customer is the regulator, not the household. Employees: ~70,000. CEO: Leonhard Birnbaum (since Apr 2021); CFO: Nadia Jakobi.
Contract structure / revenue quality. Networks revenue is not take-or-pay and not volume-driven in the way generation is — it is a revenue cap set by the Bundesnetzagentur (BNetzA) and equivalent national regulators, calculated as allowed-opex + allowed return on the regulated asset base (RAB). Volumes flow through with a lag but the return is fixed by regulation. This is the most defensive revenue shape in the whole energy complex: cash flows track the RAB, not the power price. Retail is the opposite — thin-margin, churn-exposed, commodity pass-through. customers.csv is empty; no single customer is material (mass-market retail + tariffed grid access).
Map (named stakeholders, `` throughout):
Upstream (what E.ON buys):
The company: distribution-grid operators — Bayernwerk, Avacon, E.DIS, Westnetz (Germany's largest DSO), Hansewerk in Germany; E.ON Sverige (Sweden); E.ON Distribuce (Czech Republic, ~24% of national electricity consumption); networks in Hungary and Romania.
Downstream (who needs E.ON):
Chokepoints: (1) the transformer/switchgear shortage (external, hurts everyone, but caps E.ON's throughput); (2) skilled-labour for grid construction; (3) the regulatory approval pipeline for 110kV expansion, which E.ON explicitly asked BNetzA to accelerate. Single-source dependency runs the other way: renewable generators are single-sourced onto E.ON's grid, which is the moat (Lens 3).
E.ON's moat is the strongest and dullest kind in existence: a legal, geographic, regulated monopoly on the wires. You cannot build a competing distribution grid down the same street; the local grid is a natural monopoly and the regulator grants E.ON's subsidiaries exclusive concessions. Rated on the classic moat sources:
Bargaining power. Over suppliers: moderate and eroding — the transformer shortage flips power to Siemens Energy/Hitachi. Over customers: in Networks, total (they must connect and pay the tariff); in Retail, none. The binding constraint on the moat's value is not competition — it's the regulator. E.ON's monopoly is only as lucrative as the allowed return BNetzA sets. That single dependency is the whole bull/bear axis (Lenses 11–13). The moat is unbreakable but rate-capped — it converts capital into a bond-like annuity, not into pricing power.
Commercial-layer note: kb/energy/wiki/* (bottlenecks, positioning, supply-chain) are all missing, so Lenses 1–3 are web-derived rather than KB-grounded.
segments.csv is empty, so all figures ``:
| Segment | Adj. EBITDA FY2025 | FY2024 | YoY | Investment FY2025 | Read |
|---|---|---|---|---|---|
| Energy Networks | €7.7bn | €6.9bn | +12% | €7.0bn (+20%) | Accelerating — the RAB compounding machine |
| Energy Retail | €1.7bn | €1.8bn | −6% | €0.48bn | Decelerating — normalising post energy-crisis margins |
| Energy Infrastructure Solutions | ~€0.59bn | ~€0.56bn | +5% | ~€0.9bn | Steady, small |
| Group (adjusted) | €9.8bn | €9.0bn | +9% | €8.5bn |
The trend that matters: Networks EBITDA is accelerating (+12%) and now ~79% of group profit, while its capex is up +20%. This is the flywheel — every euro of grid capex enters the RAB and earns the allowed return next period, so EBITDA growth is a function of investment, mechanically. Retail's −6% is not a problem; it's the crisis-era supply margin mean-reverting, and management guides it flat-to-down (€2.5bn total 2026–2030 investment, i.e. deliberately starved). By geography, >60% of the total RAB sits in Germany, with the Germany power-RAB target held at €34bn and the total power RAB targeted at €55bn by 2028. The €400m one-off in 2025 EBITDA was Eastern-European grid-loss/regulatory-account recoveries — non-recurring, which is why 2026 guidance steps down.
The print, all ``:
What drove it: Networks (+12% EBITDA) on RAB growth, plus the ~€400m of non-recurring Eastern-European regulatory recoveries. Retail dragged (−6%).
Margins: operating margin is not a meaningful lens for a regulated network (revenue is a pass-through-plus-return construct); the signal is the spread between allowed return and cost of debt, which widened modestly as rates stabilised.
Balance-sheet flag — the one that matters: E.ON is deliberately, structurally free-cash-flow negative. An analyst on the call flagged a >€1bn FCF deficit in 2025 and >€2bn expected in 2026; CFO Jakobi's answer: "We are a growth business" — E.ON is funding a RAB expansion larger than its operating cash flow and will stay FCF-negative for years, plugging the gap with debt and green bonds. Net-debt/EBITDA ≈ 4.6x — high, but normal-to-slightly-elevated for a regulated network with an A-/BBB+ profile; the RAB is the collateral.
Market reaction / what's priced: the stock is up ~20% over the trailing year to €19.02, near its 52-week high of €20.39. The market has clearly already rewarded the Networks-growth + Reg5-de-risking story; this is not a beaten-down name.
2026 guidance (the tell): adjusted EBITDA €9.4–9.6bn (i.e. below 2025's €9.8bn), net income €2.7–2.9bn, EPS €1.03–1.11. The step-down is entirely the roll-off of the €400m one-off — on a clean basis 2025 was ~€9.4bn, so 2026 is flat-to-up underlying. Management framed it as "stable earnings at the adjusted prior-year level." A reader who sees "EBITDA guided down" without the bridge will misread it.
transcripts/ empty; FY2025 call read via ; H1/Q3 2025 tone via .
Tone: confident on operations, insistent on regulation, dismissive of forecasting. Across H1 → Q3 → FY 2025 the message hardened around one demand: regulatory clarity on Reg5 (the 5th regulatory period, electricity, from 2029). Recurring phrases:
What shifted: the 2025 arc moved from "we need regulatory clarity or we won't invest" (H1) toward cautious relief as the BNetzA process advanced (the NEST determination landed Dec 2025 — Lens 8). What they stopped saying: the crisis-era language about Retail supply margins; that story is over.
| Company | Ticker | Mkt cap | P/E | Div yield | Net-debt/EBITDA | Business mix | Source |
|---|---|---|---|---|---|---|---|
| E.ON | EOAN.DE | ~€50bn | 14.6x (fwd basis) / 16.4x (on adj EPS 1.16) | 2.97% | ~4.6x [est] | Pure networks+retail | stockanalysis.com, 2026-07-06 |
| RWE | RWE.DE | ~€41bn | ~17.7x | 2.15% | n/a | Generation/renewables (the swap counterparty) | web: company/Investing.com, 2026-03 |
| Enel | ENEL.MI | ~€101bn | 13.5x | ~4.97% | n/a | Integrated (grids+gen), Italy/Iberia/LatAm | web: Morningstar/Investing.com, 2026 |
| National Grid | NG.L | ~£60bn | 19.0x | ~4.0% | n/a | Pure regulated T&D (UK/US) — closest analogue | web: Simply Wall St/CNBC, 2026 |
| Iberdrola | IBE.MC | >€140bn | n/a | ~3% (growing 8%+) | 3.4x | Integrated, networks-heavy | web: Investing.com/Morningstar, 2026 Q1 |
EV/EBITDA (E.ON): ≈ 9.7x on reported FY2025 EBITDA (10.1x on clean ~€9.4bn). Peer EV/EBITDA left n/a (each requires that peer's own net debt, which I did not source individually — refusing to fabricate).
Read: E.ON is cheaper on P/E than National Grid (19x) — the purest comparable — and than RWE (17.7x), but richer than Enel (13.5x), and offers a lower dividend yield (3.0%) than every peer except Iberdrola. The market is paying a slight discount to the pure-network peer (National Grid) — arguably because German Reg5 return risk still isn't fully resolved, and because E.ON's yield is the sector's thinnest (it's reinvesting the cash into the RAB instead of paying it out, targeting up-to-5% annual dividend growth). This is a fairly-valued, not cheap utility — the return comes from the earnings algorithm, not a re-rating.
`` throughout. Pattern over ~5 years:
What the market actually reacts to for E.ON: regulatory determinations first, RAB/investment-plan updates second, demand-growth narrative third. Earnings prints themselves rarely surprise (guidance is tight and the algorithm is transparent). This is a regulated bond-proxy whose beta is the German regulator's pen, not the power price.
CEO — Leonhard Birnbaum (since Apr 2021; on the E.ON board since 2013).
insider-transactions.csv absent. RWE holds 16.67% of E.ON from the swap — a large, strategically-aligned-but-rival anchor shareholder.Regulated-utility accounting is lower-risk than average — revenue is regulator-defined, there's little revenue-recognition judgement, and the RAB is externally audited into the tariff. Areas scanned, / (no filings on disk to cite as ``):
Regulatory findings (required). Read regulatory/regulatory-findings.md: 0 SEC findings — E.ON has no CIK, is not an SEC filer, so no EDGAR LR/AAER search is possible. Non-SEC / EU:
Verdict: No material regulatory or legal red flags — the marquee antitrust risk (the RWE swap) is court-settled in E.ON's favour, verified via web search + regulatory-findings.md (no SEC filings exist) as of 2026-07-06. The genuine forensic homework — the adjusted-to-IFRS reconciliation and the impairment/hedge lines — requires the actual annual report and is deferred to a grounded refresh.
Built bottom-up from management's own algorithm (guidance is unusually reliable for a regulated name). Anchors: FY2025 adj EPS €1.16; 2026 guide EPS €1.03–1.11 (mid €1.07); 2030 target EPS ~€1.45, EBITDA ~€13bn, net income €3.8bn; investment €48bn 2026–2030 (€40bn to Networks). Implied 2026→2030 adjusted-EPS CAGR ≈ 7.9%/yr, EBITDA CAGR ≈ 8.2%/yr.
Three fiscal years forward (2027–2029), adj. EPS, all `` with arithmetic:
| Year | Base (7.9% CAGR off €1.07) | Bull | Bear (~3%/yr, inflation-only) |
|---|---|---|---|
| 2026 (guide) | €1.07 | €1.11 | €1.03 |
| 2027E | €1.15 | €1.22 | €1.10 |
| 2028E | €1.25 | €1.34 | €1.14 |
| 2029E | €1.34 | ~€1.40 | €1.17 |
Brier forecast (logged conceptually — not committing in the unattended --watchlist loop): "EOAN FY2029 adjusted EPS ≥ €1.30, p≈0.55" — coin-flip-plus, hinging entirely on the Reg5 WACC parameter. No forecast.ts create run (per --watchlist rules and web-only grounding).
Bull case. E.ON is the purest listed proxy for the single most capital-hungry, most inevitable, most regulator-protected theme in Europe: the electricity-grid buildout. Electrification + renewables + data centres force a generational grid CapEx supercycle, and E.ON mechanically converts that CapEx into a growing RAB and a growing allowed-return annuity — an 8% EPS/EBITDA/dividend algorithm through 2030, backed by a legal monopoly, with the affordability problem now socialised by the German state (€6.5bn 2026 grid-fee subsidy) so the buildout can't be voted down. The Dec-2025 NEST ruling raised the allowed return and the regulator publicly committed to being "more conducive to investment." Demand vastly exceeds what E.ON can connect — the growth is supply-constrained, not demand-constrained. And there's a free call option: €5–10bn of headroom management will deploy the moment Reg5 returns justify it. You are buying a regulated bond that grows its coupon ~8%/yr.
Bear case (permanent-impairment candidates).
Pre-mortem (it's H2 2027, thesis broke — what happened?): BNetzA's Reg5 draft (published ahead of 2029) sets an allowed equity return the market judges inadequate given 4–5% bund yields; E.ON cuts the €48bn plan by ~20%, 2030 EBITDA target quietly drops toward €12bn, the dividend-growth algorithm slips to ~3%, and the stock de-rates from ~15x to ~12x → a ~20% drawdown without any operational failure — a pure multiple-and-growth reset driven by one regulatory number.
Multiples too high? No — fairly valued. 14.6x fwd P/E / ~9.7x EV/EBITDA is a discount to the pure-network comp (National Grid ~19x) and reasonable for ~8% growth; the thin 3.0% yield is the only rich-looking metric (justified by reinvestment). Not a bubble, not a bargain.
Contrarian view (what the market refuses to see): consensus treats E.ON as a boring bond-proxy and anchors on the regulatory-uncertainty overhang. The contrarian read is that the Dec-2025 NEST ruling already broke the uncertainty in E.ON's favour (allowed return rising, "+1.4% revenue," regulator explicitly pro-investment), and the €6.5bn state grid-fee subsidy removed the affordability-backlash tail — so the biggest bear pillar is quietly resolving while the stock still trades at a discount to National Grid. The market is under-pricing the option value of the held-back €5–10bn headroom.
Dismantling the bull:
Single scenario that permanently impairs: a structurally low Reg5 allowed return (below E.ON's cost of capital) that persists — turning the RAB from a compounding annuity into a value-neutral treadmill. Plausibility: moderate-and-falling, given the Dec-2025 ruling raised (not cut) returns and the regulator's stated pro-investment stance — but not zero, because the exact Reg5-2029 WACC is still open.
A de-risked regulated-utility play on the data-center power buildout — the PSCW's April-2026 verbal approval of the VLC/Bespoke tariffs converts a $37.5B capex plan into a rate-base annuity, but at ~20x forward EPS the re-rating is mostly priced and the upside now lives in 2028 acceleration, not the multiple.
A regulated-utility levered call on the Georgia data-center build-out — the cleanest large-cap way to own AI power demand, but priced as if the affordability politics and equity dilution won't bite; own the growth, respect the ~24x multiple.
A real turnaround that has already been paid for — six straight quarters of margin repair and the return to positive operating cash flow are genuine, but at ~$60 the stock prices in a clean, AMPTC-independent recovery the filings explicitly say does not yet exist (ex-45X credits, SolarEdge is still gross-loss-making), so the asymmetry from here is poor.