National Grid

Phase A — Understand the business

Company Overview

National Grid earns a regulated return on an asset base, in two geographies:

• UK (≈52% of the £70bn investment plan):
• National Grid Electricity Transmission (NGET) — owns/operates the high-voltage transmission grid in England & Wales (the "Great Grid Upgrade"). Regulated by Ofgem under RIIO price controls.
• National Grid Electricity Distribution (NGED) — the former Western Power Distribution, acquired 2021 to make National Grid the UK's largest electricity distributor (~8m connections, Midlands / South West / South Wales). Regulated under RIIO-ED.
• US Northeast (≈48% of the plan):
• New York segment — Niagara Mohawk (NIMO, upstate NY electric + gas) plus KEDNY/KEDLI (downstate NY gas). >1.7m upstate electric customers.
• New England segment — Massachusetts Electric + Nantucket Electric (~1.4m MA electric customers) and Boston Gas. Regulated by state PUCs (NY PSC, Massachusetts DPU) and, for New England transmission, by FERC.
• National Grid Ventures / Other — interconnectors (IFA/IFA2, BritNed, Nemo, North Sea Link, Viking Link) and residual non-regulated bits; being run down as the group narrows to networks.

Revenue mechanics. Revenue is allowed revenue set by the regulator, not a market price — cost-of-service (US) and RAB×allowed-return + incentives (UK RIIO). Commodity costs are largely pass-through, so the reported top line swings with gas/power prices and is a poor gauge; the meaningful drivers are RAB / rate base, allowed return on equity, and totex efficiency. Group revenue this pass: n/a (and structurally low-signal for a RAB name).

Latest full year (FY2026, to 31 Mar 2026, reported 14 May 2026):

• Underlying operating profit £5,680m, +£459m / +9% constant-FX.
• Underlying EPS 78.0p, +8% constant-FX.
• Total dividend 48.49p, +3.8% (UK CPIH policy).
• Record capexCapital expenditureMoney spent on long-lived things — buildings, machines, servers — rather than on running costs. £11.6bn; plan upgraded to "at least £70bn" over five years.

Supply Chain

For a network builder the "supply chain" is the equipment + civils + connection queue — and here National Grid has named its chain, which de-genericises this lens.

Upstream (the ~£59bn HVDC framework awards, Mar 2025):

• HVDC converter stations (~£24.6bn framework): GE Vernova, Hitachi Energy, Mitsubishi Electric, Siemens Energy — four suppliers, 5-year + 3-year initial terms.
• HVDC subsea/land cables (~£21.3bn framework): Prysmian Group, NKT, LS Cable & System, Sumitomo Electric, Taihan Cable & Solution, Hellenic Cables/Jan De Nul consortium — six suppliers.
• Civils (~£12.8bn) across the framework.
• Early projects consuming this: Eastern Green Link 4, Sealink, LionLink.

Chokepoints / single-source risk:

• Large power transformers are the acute global bottleneck — ~30% supply shortfall and ~128-week lead times in 2025. HVDC converter capacity (only 4 credible global vendors, notably Siemens Energy / Hitachi Energy / GE Vernova) is similarly thin — National Grid competes with every other TSO and offshore-wind developer worldwide for the same order books.
• Skilled labour + planning/consenting — the domestic constraint; UK transmission projects historically stall for years in consenting.

Downstream ("customers"): not really discretionary buyers — connection-seeking generators (offshore-wind developers: Ørsted, SSE Renewables, Equinor, etc.), interconnector counterparties, and, increasingly, large loads — datacenters/hyperscalers and electrified industry. In the US the end customer is millions of captive ratepayers via state utilities. The binding downstream fact is the connection queue, covered in Lens 3/12.

Competitive Advantages (moats)

The moat is a legal monopoly franchise — the strongest and most brittle moat there is.

• Structure: exclusive licensed operator of the transmission network in England & Wales and of distribution/utility territories in the Midlands/SW/Wales and in upstate NY / Massachusetts. There is no competitor; you cannot build a parallel grid.
• Bargaining power over customers: total (captive ratepayers) — but capped by the regulator, which is the real counterparty. National Grid's returns are set, not negotiated.
• Bargaining power over suppliers: improving via the £59bn frameworks (long-term volume commitments lock in scarce converter/cable capacity ahead of rivals) — a genuine, if modest, advantage in a supply-constrained decade.
• Durability: the franchise is effectively perpetual, but the return on it is repriced every ~5 years (RIIO) or per rate case (US). So the moat protects the asset base but not the profitability of the asset base — the state can, and periodically does, cut the allowed return (see RIIO-T3 below-ask; FERC New England ROE cut, Lens 10).
• Reinforcing tailwind: electrification + AI load turns the network from a mature utility into a scarce, growth-constrained asset — the queue (>700GW UK; 9.5GW proposed NY datacenter load) means National Grid's core asset is the binding constraint on other people's growth, and it is paid a regulated return to relieve it. That is a rare position.

Segments

FY2026 underlying operating profit £5,680m:

Geographic mix: UK ≈ £2.8bn (~49%), US ≈ £2.6bn (~45%), NGV/other ~6% — a roughly 50/50 UK/US earnings split. This is the defining structural feature: half the book is UK RAV, CPIH-indexed, real-return RIIO regulation; half is US nominal-ROE, cost-of-service rate-base regulation. The two halves diversify regulatory and FX risk but also mean two sets of regulators to satisfy. Trend & cause: New York accelerating (the Aug-2025 3-year NIMO rate plan reset allowed revenue up); New England stalled (FERC Opinion 594 refunds, Lens 10); UK transmission growing on higher RIIO-T2 allowances and stepping up into RIIO-T3 (Apr 2026).

---

Phase B — Measure performance

Earnings Result (FY2026, reported 14 May 2026)

[all web: National Grid FY2026 results statement, via Investegate / DirectorsTalk / TipRanks / Investing.com, 14 May 2026]

• Underlying operating profit £5,680m, +9% constant-FX — driven by (1) new Niagara Mohawk rates, (2) recovery of prior-year storm & environmental costs, (3) higher UK allowances, (4) cost efficiency.
• Underlying EPS 78.0p (+8% cc); statutory EPS materially lower (~60p range ) on disposals/exceptionals — a wide underlying-vs-statutory gap to watch (Lens 10).
• Dividend 48.49p total (+3.8%) — CPIH-linked policy intact post-rebasing.
• Capex a record £11.6bn; group asset growth +10.9%, regulated asset growth +11.7%.
• Balance sheet: net debt £44.2bn (+£2.8bn YoY), the increase funding capex and offset by £2.8bn of divestment proceeds (NG Renewables + Grain LNG). UK RAV £36,986m (£37.0bn); US rate base reported ~£29.5bn; total assets in gearing calc £71,983m; regulatory gearing 61%.
• Guidance: FY2026/27 underlying EPS +13–15% off the 78.0p base (the RIIO-T2→T3 revenue step-up), on top of the medium-term 6–8% EPS CAGR (from the FY25 73.3p baseline) and 8–10% asset growth.
• Market reaction: muted/positive — the result was in line and the £70bn plan + RIIO-T3 acceptance had been pre-flagged (2 Mar 2026). Consensus stayed ~Hold (Lens 7).
• Unusual vs own history: the +13–15% one-year EPS step-up is atypical for a utility and worth stress-testing — it front-loads the RIIO-T3 uplift; the through-cycle number is the 6–8%.

Earnings Calls (sentiment trend)

• Tone shift = leadership shift. New CEO Zoë Yujnovich (in seat since autumn 2025) framed FY26 as "the momentum we're building" and put execution at the centre — a deliberate pivot from her predecessor's strategy-reshaping era to a delivery narrative.
• Recurring phrases: "investable and workable" (their RIIO-T3 test), "at least £70bn", "record investment", "comfortable headroom" (ratings).
• What they now emphasise: in UK transmission, ~£31bn of RIIO-T3 investment (a ~150% step-up), connecting up to 35GW of generation and 19GW of new demand, and confidence in an overall return on equity above 9% across the control (i.e. the 6.12% real allowed CoE grosses up to >9% nominal). Regulatory gearing trending to the high-60s% by 2030/31 (from 61%), with "comfortable headroom" vs rating thresholds and unused hybrid-debt capacity — the explicit answer to "will you raise equity again?"
• What they stopped saying: the M&A/portfolio-reshaping language of 2022–24 (gas sale, WPD, ESO) is largely done — the story is now organic RAB growth, not portfolio surgery.

Comps

Peer set = large regulated electricity T&D networks. Multiples are `` with source/period or n/a; none are fabricated. All figures 2026 unless noted.

Read: on P/E, National Grid (~13–16x underlying; higher ~21x on statutory) sits below US regulated peers (Duke ~21x, NextEra ~25x) and roughly in line with UK peer SSE (~16x fwd) — the standard UK-utility discount to US utilities. Yield ~4% is at the top of the group (only Terna/SSE close). The closest pure structural comps are Terna (pure transmission) and Iberdrola (grids-led) — not the US integrateds. Valuation vs asset base: EVEnterprise valueWhat it would cost to buy the whole company: its market value plus its debt, minus the cash you would get with it. Often very different from market cap. ≈ £64bn equity + £44.2bn net debt ≈ ~£108bn, against combined UK RAV £37.0bn + US rate base ~£29.5bn ≈ ~£66bn → ~1.6x EV/RAB — a real but not extreme premium, consistent with RAV indexation + the funded £70bn growth pipeline. Analyst consensus Hold, avg target ~1,315p vs ~1,280p price, range 1,070–1,440p (JP Morgan Buy 1,440p ).

Stock-Price Catalysts (moves >5%, ~5y)

• 23 May 2024 — the £7bn rights issue: −~10% on the day, ~another −10% ex-rights. THE defining move: a surprise 7-for-24 raise at 645p (34.7% discount to TERP), +29% share count, sprung on results day after prior guidance implied no need. This single event reset the register and the narrative.
• 2022 rate-shock drawdown — NGG total return −12.7% in 2022 as gilt/Treasury yields spiked; classic bond-proxy de-rating.
• 2023 recovery — NGG total return +18.8% as rate fears eased.
• 2024 flat (−1.3% total return) — the rights-issue overhang.
• RIIO milestones — Ofgem draft/final determinations (2025) and the RIIO-T3 acceptance (2 Mar 2026) removed a multi-year regulatory overhang.
• Pattern: the market reacts to (1) long rates (bond-proxy beta) and (2) equity-issuance / regulatory-return surprises far more than to quarterly operating beats. Earnings are pre-set; the swing factors are the discount rate and the allowed return. That is the whole game for this name.

---

Phase C — Judge people & books

Management

• Zoë Yujnovich — CEO (from ~autumn 2025). External hire; ex-Shell (Integrated Gas & Upstream Director, an ExCo role), previously Shell Australia Chair and Rio Tinto (President/CEO, Iron Ore Company of Canada). Read: the board deliberately imported a large-capital-project / heavy-industry operator, not a regulatory insider — a signal that the £70bn build is now an execution problem, not a strategy problem. The oil-&-gas pedigree is mildly incongruous with a decarbonisation-mission utility, but her megaproject-delivery skill set fits the actual bottleneck. First non-lifer CEO — worth watching for a step-change in delivery discipline (bull) or culture friction (bear).
• John Pettigrew — CEO Jun 2016 → retired 16 Nov 2025 after 35 years at the company (graduate joiner 1991). His era reshaped the portfolio: bought WPD (2021, ~£7.8bn ) to lead UK distribution; sold gas transmission (National Gas) to Macquarie/BCI; sold Rhode Island (Narragansett) to PPL; separated the ESO into publicly-owned NESO; agreed the US-renewables and Grain-LNG exits; and launched the £60bn plan + £7bn rights issue. Coherent strategic repositioning to a networks pure-play — but the surprise rights issue is a permanent governance blemish (Lens 13).
• Andy Agg — CFO (since ~2019/2020, at NG since 2008, ex-PwC). Continuity; steered the balance sheet through the raise.
• Skin in the game: low insider ownership, as is typical for a large UK utility run by professional managers (not founders) — n/a exact, not sourced. Alignment is via LTIPs, not stock.
• Capital allocation: overwhelmingly reinvestment into RAB — value-accretive so long as allowed return (>9% nominal RIIO-T3 / US ~9–10% ROEs) exceeds cost of capital (~5.6% WACC), which it currently does. Disposals recycled into the core. Not a buyback story — they issued equity. The judgement call bulls and bears split on: is management a disciplined RAB compounder, or a serial acquirer/divestor generating churn and one dilutive surprise?

Forensic Red Flags

Accounting watch-points (RAB-utility specific):

1. Underlying vs statutory EPS gap — underlying 78.0p vs statutory ~60p; recurring adjustments (exceptionals, disposals, remeasurements, deferred tax, timing/"regulatory" items) flatter the headline. Track the reconciliation each year.
2. Capitalisation policy — heavy capex means aggressive vs conservative cost capitalisation swings current earnings and future depreciation; in a RAB frame, over-capitalised costs still enter the asset base and earn a return, so the incentive tilts toward capitalising.
3. FCF is structurally deeply negative by design — capex £11.6bn/yr vastly exceeds operating cash generation; the equity story rests entirely on RAB growth × allowed return, funded by debt + (occasionally) equity, not on free cash flowFree cash flowCash left after paying to run and maintain the business. Unlike profit, it is hard to flatter with accounting choices.. This is normal for the model but means classic "cash conversion" screens will flash red and should be read in that light.
4. Leverage & indexation — net debt £44.2bn, gearing 61%→high-60s%; large index-linked debt book means RPI/CPIH cuts both ways (indexes up RAV and index-linked interest). Rate sensitivity is a genuine forensic exposure, not just a valuation one.
5. Regulatory assets/liabilities (over/under-recovery) smooth reported vs cash revenue — a legitimate but opaque timing lever.

Regulatory / legal findings:

• SEC/EDGAR: none possible — no CIK; the pre-ingested regulatory/regulatory-findings.md records 0 SEC LR/AAER findings (National Grid is not a US domestic filer).
• FERC — Opinion No. 594 (Mar 2026): base ROE for New England transmission owners cut retroactively 10.57% → 9.57% after 15 years of litigation; ~$1.5bn of industry refunds ordered by 20 May 2027 (Eversource ~$880m, Avangrid ~$203m; National Grid a NETO with a material-but-smaller share). This is why the New England segment was flat in FY26. NETOs asked FERC (30 Apr 2026) to raise base ROE to 11.39% — live, two-way.
• Massachusetts DPU — $15m fine for 2023 service-quality/billing-system failures (gas customers not billed on time, late self-reporting); refunded to customers May 2025. Plus AG $7m overcharge settlement and a DOJ $5.38m natural-resource-damages payment.
• Ofgem — £8m penalty (finalised Mar 2025) against National Grid Gas for misreporting gas-pipe maintenance data (2005/06–2007/08); plus a historic £41.6m gas-metering abuse-of-dominance fine (CAT-upheld, 2008) and an ESO SLC16 licence penalty.
• 10-K Item 3 equivalent: standard utility exposures — manufactured-gas-plant (MGP) environmental remediation, storm-cost recovery disputes.
• Verdict: a recurring drumbeat of modest US billing/service-quality and UK reporting frictions plus a real New-England-ROE haircut — none individually existential, but a pattern of consumer-protection run-ins in a politically sensitive sector. Watch the FERC appeal and the 2026 MA gas rate case.

---

Phase D — Project & stress-test

Forward Projection (underlying EPS)

Base off FY2026 actual 78.0p and management guidance (FY27 +13–15%; medium-term 6–8% CAGR). No forecast logged (our model create skipped — unattended watchlist pass).

• Base logic: FY27 takes the guided RIIO-T3 revenue step-up; FY28–29 normalise to the 6–8% through-cycle CAGR, powered by ~10% asset-base growth net of ~50–60% gearing and rising financing cost.
• Bull inputs: high-end delivery + stronger CPIH/RAV indexation + US rate-case wins + totex outperformance incentives.
• Bear inputs: higher-for-longer rates lift refinancing cost on £44bn+ debt (interest eats EPS), New-England/US ROE compression continues, capex slips/overruns disallowed, another equity raise dilutes.
• Sanity vs price: base FY29 ~101.5p at a 15–16x utility P/E ⇒ ~£15.2–16.2 (1,520–1,620p) intrinsic in ~3y before dividends vs ~1,280p now — i.e. a mid-single-digit-plus-~4%-yield total-return algo if the multiple holds. The multiple holding is a bet on rates.

Bull vs Bear

Bull. A legal-monopoly network with a funded, regulator-accepted, accelerating ~10–12% asset-base growth plan (£70bn, RIIO-T3 signed 2 Mar 2026, no CMA appeal → overhang cleared). Half CPIH-inflation-protected UK RAV, half US nominal-ROE rate base — diversified regulation. FY27 EPS +13–15% is a visible near-term catalyst; ~4% yield + 6–8% CAGR = a ~10% total-return algorithm. And the once-in-a-generation kicker: electrification + AI datacenter load makes the grid the scarce asset — >700GW UK connection queue, 9.5GW proposed NY datacenter load, transformer/HVDC scarcity — and National Grid is paid a regulated return to relieve the bottleneck it sits on. This is a growth utility the market still prices as an income utility. A rates-down cycle re-rates it hard.

Bear. It is a £44bn-net-debt bond proxy with gearing climbing to the high-60s%. Higher-for-longer long rates deliver a double hit — valuation compression (yield/DCF) and rising refinancing cost on a vast debt book. The equity has already been tapped once (£7bn, 2024) and a second raise cannot be excluded if capex overruns or gearing pressures ratings. Execution risk on a doubling of capex is real: transformers at 128-week lead times, only four global HVDC-converter vendors, labour and planning constraints — and in a regulated frame, overruns can be disallowed. And returns are political: RIIO-T3's 6.12% real was below National Grid's ask (they accepted anyway), while FERC just cut New England ROE 100bps retroactively — regulators push back on returns exactly as consumer bills (and the political heat) peak.

Pre-mortem (18 months out, thesis broke): long rates stayed high; a FY27 EPS print landed at the low end or below; a US rate case (MA gas 2026) or the FERC appeal went against them; gearing crept toward rating thresholds and the market began pricing a second rights issue. The stock de-rated from ~16x to ~13x — a ~1,050–1,100p handle.

Are multiples too high? No — ~13–16x underlying is fair, not stretched, and below US peers. The risk is de-rating from fair to cheap on rates/politics, not popping a bubble.

Contrarian view (what the market refuses to see): two-sided. Bull-contrarian — the market is anchored on "boring bond proxy" and under-weights that this is now the most-levered listed pure-play on the electrification/AI-power supercycle with a signed and accelerating plan. Bear-contrarian — the market treats the £70bn as if it all earns its allowed return, ignoring that affordability politics + execution disallowances can quietly turn "RAB growth" into "RAB growth at a haircut," because the buildout and the consumer backlash arrive together.

Devil's Advocate (short-seller)

• The concentration bulls ignore is regulatory, not customer. Millions of captive ratepayers, but ~100% of revenue is set by four bodies (Ofgem, NY PSC, MA DPU, FERC). A single adverse determination hits returns directly — and both just happened (RIIO-T3 below-ask; FERC −100bps retroactive). The "moat" is a licence the state re-prices at will.
• The most dangerous "competitor" is the sovereign. The UK already took the system operator into public ownership (NESO, Oct 2024). Any drift toward more public control of networks, a "windfall"-style intervention, or a punitive RIIO-4 is a tail no monopoly-moat model captures.
• Cost of capital is the other silent competitor. If risk-free rates stay elevated, the RAB model's equity value erodes regardless of operational delivery — you can build perfectly and still de-rate.
• Worst capital-allocation move: the surprise £7bn rights issue — ~10% of value gone overnight, 29% dilutionDilutionIssuing new shares, so each existing share owns a smaller slice of the same company., after guidance implied no need. A governance/communication failure that permanently raised the equity-story risk premium. Add serial portfolio churn (WPD in / gas out / RI out / renewables out / Grain out) = fees + integration risk.
• Assumptions that must hold for today's price: rates don't spike; RIIO-T3 + US rate cases deliver ~9% ROE in practice, not just on paper; £70bn deploys near budget; no second equity raise; CPIH indexation persists.
• If growth disappoints 20–30% (EPS CAGR ~4% not ~7%): de-rate to ~13x and a lower terminal value ⇒ ~20–30% downside, a high-900s/low-1,000s p handle.
• Single permanent-impairment scenario: coordinated political action on allowed returns plus a rates-driven refinancing squeeze. Plausibility: low — because the state needs the grid built for its own net-zero/AI-industrial goals, which self-limits how hard it can squeeze. Most-likely bad case is de-rating, not impairment.

Management Questions (ordered by information value)

1. Bridge the 6.12% real RIIO-T3 allowed CoE to your ">9% return on equity" claim — how much is nominal grossing-up vs incentives/ASTI vs totex outperformance, and what outperformance is baked into FY27–29 guidance?
2. At what capex-inflation / interest-rate / disallowance scenario do you need fresh equity again, and what will you pre-commit publicly to avoid a repeat of the 2024 surprise?
3. Gearing is heading to the high-60s% by 2030/31 — what is the hard ceiling before ratings are at risk, and how much genuine headroom (incl. hybrid capacity) exists at each rating threshold?
4. What is the realistic deliverable capex given transformer (128-wk lead times) and HVDC-converter (four-vendor) constraints — is £70bn a spend *target_ or a deliverable number, and what slips first?
5. On the FERC New England ROE cut and appeal — quantify National Grid's refund exposure and the earnings sensitivity to the 9.57% vs 11.39% range.
6. How much of the AI/datacenter connection demand (US 9.5GW proposed; UK Gate-2 pipeline) do you expect to convert to rate-based investment, over what horizon, and at what returns?
7. RIIO-T3 was below your ask and you accepted — what did you concede, and what is your posture into RIIO-4?
8. US rate cases (MA gas 2026, next NY resets) — what ROE and equity-thickness are you filing for, and what's the risk of adverse outcomes as bills rise?
9. What is the through-cycle underlying-to-statutory EPS reconciliation you expect, and when does the disposal-driven noise clear?
10. New CEO: what specifically changes operationally under an execution-first mandate — any hard delivery/cost-efficiency metrics you'll be judged on?
11. Interconnectors / NG Ventures — keep, grow, or exit, and does the "networks pure-play" thesis mean further disposals?
12. Currency/US exposure now ~half of earnings — hedging policy and the FX sensitivity of guidance?
13. Dividend: is CPIH-linked growth a floor or a policy that flexes if gearing tightens?
14. Supply-chain framework (£59bn) — how much price/volume certainty do the converter/cable frameworks actually lock, and where are you still exposed to spot inflation?
15. What is the single scenario that most worries you about the 2026–2031 plan?

---