Duke Energy

Phase A — Understand the business

Lens 1 · Company Overview

Duke Energy is one of the largest fully-regulated electric and gas utilities in the United States — a Delaware holding company headquartered at 525 South Tryon Street, Charlotte, North Carolina, incorporated May 3, 2005. It serves roughly 8.6 million electric customers and ~1.7 million gas customers across six states (the Carolinas, Florida, Indiana, Ohio, Kentucky, Tennessee). The combined 10-K is filed by eight registrants — Duke Energy plus the regulated operating companies Duke Energy Carolinas, Progress Energy, Duke Energy Progress, Duke Energy Florida, Duke Energy Ohio, Duke Energy Indiana, and Piedmont Natural Gas.

How it actually makes money. This is a textbook regulated-utility model: Duke invests capital into rate base (poles, wires, power plants, gas mains), state public service commissions set an allowed return on equity (ROE) on that capital, and customers pay rates designed to recover the investment plus the allowed return. Earnings grow primarily by growing rate base — not by selling more electricity. Fuel costs are a pass-through (recovered via fuel clauses), so the company is largely insulated from commodity swings. The entire equity story is: how fast can rate base grow, and will regulators let Duke earn its allowed ROE on it?

Business structure — two reportable segments + Other:

• Electric Utilities & Infrastructure (EU&I) — the core; regulated electric utilities in the Carolinas, Florida and the Midwest. FY2025 revenue $29,357M, segment income $5,337M (≈91% of segment income).
• Gas Utilities & Infrastructure (GU&I) — Piedmont + the Ohio/Kentucky gas LDCs + gas storage/midstream/RNG. FY2025 revenue $3,003M, segment income $559M.
• Other — holding-company interest, unallocated corporate, the Bison captive insurer, and the NMC investment. FY2025 segment loss $(985)M (almost entirely holdco interest expense).

Customers / concentration. "No Subsidiary Registrant has an individual customer representing more than 10% of its revenues". The customer base is millions of regulated retail ratepayers — extremely low concentration risk by construction. The new and important wrinkle is large-load data-center customers (see Lens 3/8), which are contracted under long-term Electric Service Agreements with minimum-take provisions.

Contract/payment structure. Revenue is set by tariff, not negotiated contract; the "counterparty" is effectively the state regulator. The exception now mattering for the growth case: large-load customers sign multi-year ESAs with minimum-take protections so Duke earns on the generation it builds for them even if the load ramps slower.

Lens 2 · Supply Chain

For a regulated utility the "supply chain" is two distinct chains — the fuel/commodity chain and the capital-equipment chain. The capital-equipment chain is the one that now governs the growth story, because Duke is trying to build the largest capex program in the industry into a constrained equipment market.

Upstream inputs → Duke → end customer:

1. Fuel & commodity inputs. Natural gas (purchased "almost all… from interstate sources" transported under firm pipeline service agreements), coal, nuclear fuel, fuel oil. Fuel cost is recovered through cost-recovery clauses subject to state-commission approval — so the cost risk is largely passed to ratepayers; the availability risk (pipeline disruption, counterparty default) is Duke's.
2. Capital equipment. Solar panels, transformers, grid components, gas turbines, nuclear components. The 10-K explicitly flags rare-earth export-control risk, tariffs, and trade policy as supply-chain threats to "capital plan execution" and discloses that Duke has "executed longer supply agreements for solar panels" and is using framework agreements with key suppliers to lock equipment and "ensure cost and schedule certainty". This is the named chokepoint: transformer and turbine lead-times against a $103B build.
3. Construction & labor. Large simultaneous multi-plant builds carry "compounding risks"; the 10-K names labor cost, material availability, workforce/equipment productivity, and tariff impacts as cost/schedule risks.

Downstream / end customers: millions of regulated retail electric and gas ratepayers, plus a fast-growing slug of hyperscale data-center load. Named anchor: Amazon's planned $10 billion high-tech cloud/AI innovation campus in Richmond County, NC — a site that had been in Duke's Site Readiness Program since 2019. In 2025 Duke won 87 economic-development projects representing >$30B of new third-party capital investment and ~29,000 jobs in its territories.

Chokepoints / single-source dependencies: (1) interstate gas pipeline capacity into the service territories; (2) transformer/turbine/solar-module lead-times and tariff exposure on imported grid gear; (3) — uniquely for Duke — the regulator itself is the binding "supplier" of returns. No amount of equipment matters if a commission disallows the spend (see Lens 10/13). Names or it didn't happen: suppliers locked under framework agreements (solar modules), interstate pipeline operators delivering firm gas, Deloitte & Touche LLP as auditor (Charlotte), Brookfield Super-Core Infrastructure as the new Florida minority financier (Lens 5).

Lens 3 · Competitive Advantages (moats)

Regulated utilities have the most boring and most durable moat in equities: a legal monopoly franchise. Duke's specific edge:

1. Monopoly service territory. Duke is the sole electricity provider across most of its footprint. There is no competitor "stealing customers." The product (electrons) is undifferentiated; the moat is the franchise, not the brand. Switching cost is effectively infinite — customers cannot choose another wires provider.
2. Regulatory-jurisdiction quality is the real moat variable. A utility is worth what its commissions let it earn. Duke's footprint is a mix: Florida and Indiana are generally constructive (forward-looking rates, riders, storm securitization); North Carolina is the swing factor and is currently hostile (see Lens 8/10). This is the single most important moat nuance — Duke's franchise is strong, but the terms of the franchise are being contested in its biggest state.
3. Scale & rate-base size. $114B rate base in 2025 growing toward $180B by 2030 — the scale gives lower financing cost, supply-chain leverage (framework agreements), and the balance sheet to absorb a $103B plan. Bargaining power over equipment suppliers is improving with scale; bargaining power over regulators is not — that asymmetry is the whole risk.
4. The new structural advantage: being where the AI load is landing. The Carolinas/Southeast and Indiana are precisely where hyperscalers are siting data centers (cheap land, Site Readiness Program, available interconnection). 4.5 GW of signed data-center ESAs + ~9 GW late-stage pipeline means Duke has a demand tailwind most utilities lack — load growth that justifies the capex and, with minimum-take contracts, de-risks it.

Bargaining power: strong over suppliers (scale, multi-year framework deals), strong over customers in the legal-monopoly sense, weak over regulators — and regulators are the counterparty that sets the price of everything. Net: a real, wide moat on the franchise, with a live crack in the most important jurisdiction.

Lens 4 · Segments

All segment figures ``. Duke reports by product segment; geography is essentially "U.S. only" ("substantially all assets and revenues… are within the U.S.").

Operating revenue by segment (USD M):

Note the trap: the "$29,357M" widely associated with Duke is the EU&I segment revenue; consolidated total operating revenue (incl. intersegment eliminations) is $32,237M in FY2025 vs $30,357M FY2024. GU&I's +25.6% revenue jump is mostly higher pass-through cost-of-gas revenue (+$429M on higher commodity prices) and the 2024 Piedmont NC rate case — not real margin growth.

Segment income (USD M):

Products & services within segments (FY2025, USD M):

• EU&I: Retail electric $25,504, Wholesale electric $2,423, Other $1,430.
• GU&I: Retail natural gas $2,782, Other $221.

Geographic / operating-company split (EU&I detail, FY2025 revenue): Duke Energy Carolinas $9,713M; Progress Energy (Duke Energy Progress + Duke Energy Florida) $14,509M; Duke Energy Florida $7,105M (within Progress); Duke Energy Indiana $3,544M; Duke Energy Ohio $2,797M.

The cause behind the trend: EU&I's acceleration is driven by rate-case pricing (+$951M), Florida storm-recovery revenue (+$753M), weather-normal volume (+$223M) and riders (+$161M) — partially offset by a $1,119M decline in fuel revenue (pass-through, lower fuel rates). The takeaway: the real earnings engine is regulated price increases on a growing asset base, exactly what you want to see, with the caveat that the largest single driver (rate cases) is the thing now under legal attack in NC.

The deteriorating "Other" loss (-$985M) is the cost of the balance sheet — holding-company interest expense rising with debt issued to fund the build. This is why the Florida/Brookfield equity-funding deal (Lens 5) matters: it slows the growth of that drag.

---

Phase B — Measure performance

Lens 5 · Earnings Result

FY2025 (10-K) and Q1 2026 (10-Q) — the actuals ``:

vs consensus. FY2025 EPS of $6.31 = +7% YoY and "finished above the midpoint of guidance". The Q4-2025 quarter itself was roughly in line / a hair light: Q4 EPS $1.50 vs ~$1.49 consensus (Investing.com framed it as a beat; Nasdaq framed Q4 as lagging estimates) — call it in-line. Q1 2026 EPS $1.97 beat (driven by data-center load + favorable weather + infrastructure recovery).

What drove it. Pure regulated mechanics: rate-case pricing across multiple jurisdictions, Florida storm-cost recovery, higher transmission revenue, weather-normal volume growth. Offsets: higher O&M (+$1,229M at EU&I — storm amortization, NC litigation/environmental costs, higher employee costs), higher depreciation on a bigger asset base, and higher interest expense. Note GAAP EPS equaled adjusted EPS in 2025 ($6.31 = $6.31) — no special-item gap, a sign of clean reporting this year (2024 had a $5.71 GAAP vs $5.90 adjusted gap from SC rate-case impairments).

Guidance / tone. 2026 adjusted EPS guidance $6.55–$6.80 (reaffirmed at Q1); 5–7% long-term EPS CAGR through 2030 off the $6.30 2025 midpoint, with management expressing confidence in the top half of that range by 2028. Tone is confident and growth-forward — a notable shift for a utility historically pitched as a bond proxy.

Balance-sheet flags. Total long-term debt (incl. current maturities) ≈ $86.9B at YE2025; unsecured debt $35,585M at a 4.63% weighted average rate, plus secured and ~$39.8B of first-mortgage bonds. This is a heavily levered balance sheet (as all regulated utilities are) — the central financial question is funding the $103B plan without a credit downgrade or massive equity dilution. The Florida/Brookfield $6B (Lens 5 note below) is the answer to part of that.

Market reaction. Despite the Q1 2026 beat, the stock dipped on the print — the market is trading the regulatory overhang (NC rate case), not the EPS. That tension is the whole investable setup.

Florida Progress / Brookfield (the financing lever): In Aug 2025 Duke agreed to sell 19.7% of Duke Energy Florida to Brookfield Super-Core Infrastructure for $6.0B all-cash, keeping 80.3% and full control. First closing $2.8B on March 3, 2026; remaining tranches ($200M by 12/26, $500M by 6/27, $1.5B by 12/27, $1.0B by 6/28) total ~$6B by mid-2028. Duke is also selling Piedmont's Tennessee business (expected close 3/31/2026). Both are non-dilutive (or less-dilutive) equity-substitute funding for the capex plan.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the research shelf (transcripts=0), so this is web-grounded ``.

The clear multi-quarter narrative shift: Duke has repositioned from "stable regulated yield play" to "AI-load growth utility." Across the Q4 2025 and Q1 2026 calls, management's dominant theme is data-center load growth — the 4.5 GW of signed ESAs and ~9 GW late-stage pipeline are the headline metric they lead with, and they repeatedly stress the minimum-take contractual protections that de-risk the associated capex. The second recurring theme is capital-plan scale and execution — the $103B plan (raised $16B), framed as the largest in the industry, with explicit confidence in the top half of the 5–7% growth band by 2028.

What they're now saying that they weren't 18 months ago: "data center," "minimum take," "top half of the range." What's getting more airtime defensively: North Carolina regulatory/rate-case language and supply-chain/tariff risk to "capital plan execution." The CEO voice changed too — this is Harry Sideris's first full cycle of calls after taking over from Lynn Good in April 2025 (Lens 9). Net sentiment: confident and growth-forward, with a visible undercurrent of regulatory caution that the market is weighting more heavily than management's optimism (hence the post-beat stock dip).

Lens 7 · Comps

Peer set = large-cap U.S. regulated electric utilities. Multiples are ``, June 2026, P/E and yield from stockanalysis/companiesmarketcap aggregations; EV/EBITDA and 5-yr ROE not cleanly sourced → n/a rather than fabricate.

Read. Duke trades at 18.9x — the cheapest P/E of the large-cap regulated group, slightly under AEP (18.7x) and well under SO (23.8x) and NEE (20.8x), while offering an above-peer 3.44% yield (only Dominion's 4.32% is higher, and that reflects a higher payout/lower growth). For a utility with the largest capex plan in the industry and a genuine data-center demand tailwind, a discounted multiple is the anomaly the bull case is built on. The discount is the market's price for North Carolina regulatory risk and balance-sheet leverage.

Industry structural event to weigh: NextEra and Dominion announced a combination on May 18, 2026 to create "the world's largest regulated electric utility". This re-rates the comp set, signals scale consolidation in regulated utilities, and arguably makes Duke's standalone scale + Southeast data-center positioning more strategically valuable (and itself a theoretical consolidation participant, though its size makes it more acquirer than target).

Lens 8 · Stock-Price Catalysts

What actually moves DUK (mostly ``, 2024–2026):

1. North Carolina rate-case / regulatory headlines — the dominant catalyst. Duke's pending NC request (reported ~15% increase) faces resistance from state officials, and a NC Court of Appeals ruling found the utility commission wrongly approved a 2024 Duke Energy Carolinas rate hike. Regulatory news now drives the stock more than earnings — the Q1 2026 beat still produced a stock dip, and recent analyst target cuts (Mizuho, Evercore) cite "regulatory uncertainty". DUK fell ~6% over a three-month window on this overhang.
2. Data-center ESA announcements. Each large signed agreement (and the Amazon $10B NC campus) is a positive catalyst — they convert the load-growth narrative into contracted backlog. 4.5 GW signed / 9 GW pipeline is the metric the market rewards.
3. Hurricanes / storm cost. 2024's Debby/Helene/Milton hit ~3.5M customers; cumulative restoration O&M reached ~$789M by Sept 2025. Storms are a near-term EPS/cash drag but a rate-base and securitization opportunity (recovery + $112M Helene securitization in the rate case) — the market reaction depends on recovery clarity.
4. Capital-plan / guidance updates. The $103B plan and 5–7% growth reaffirmation are structural supports; raising the plan $16B was a positive.
5. Asset-sale / financing news. Brookfield $6B Florida deal and Piedmont-TN sale were received as credit-positive funding events.
6. Rates/macro. As a high-yield, capital-intensive bond proxy, DUK is sensitive to the 10-year Treasury and Fed path — falling rates help the multiple and lower financing cost; rising rates do the reverse.

Pattern: the market reacts most violently to regulatory outcomes in North Carolina and to rate sensitivity — far more than to the quarterly print itself. This is the defining feature of the name: earnings are predictable; the regulator and the bond market are not.

---

Phase C — Judge people & books

Lens 9 · Management

1. Track record. Harry Sideris became President & CEO on April 1, 2025, succeeding Lynn Good after her 11+ year tenure. Sideris is a 29-year Duke veteran, president since April 2024, who ran the electric and gas utilities — operations, customer service, economic development, regulatory/legislative affairs, and grid/generation strategy. This is an inside, operations-and-regulatory-bred successor — exactly the profile you want for a company whose #1 risk is regulatory execution. Good's legacy was transforming Duke into a fully-regulated utility (exiting Commercial Renewables, merchant generation) — a de-risking that defines today's clean utility P&L.
2. Tenure & skin in the game. Sideris is new in the seat but deeply tenured in the company. Insider ownership at large-cap utilities is structurally small as a % of a ~$96B cap; insider-transactions.csv is not on the shelf, so precise insider holdings are n/a. Lead independent director Ted Craver became independent chair (splitting Chair/CEO — a governance positive).
3. Capital-allocation history. The defining decisions: (a) exiting Commercial Renewables (recognized large disposal losses — $1,725M loss on disposal in 2023, +$14M 2024 ) to become fully regulated; (b) selling 19.7% of Florida to Brookfield for $6B and Piedmont-TN to fund capex without max equity dilution; (c) committing the $103B 2026–2030 plan, the industry's largest. ROE/segment-income trend is favorable (EU&I segment income $4,223M → $5,337M over three years). The capital-allocation story is disciplined regulated reinvestment funded creatively — a credit-conscious posture.
4. Red flags. No related-party or promotional-behavior flags surfaced. The honest critique is strategy-pivot cost: the Commercial Renewables exit destroyed real value on the way out (billions in disposal losses) before leaving a cleaner business — a reminder that the prior era's diversification was a mistake shareholders paid to unwind. The current risk is over-promising on the $103B plan if NC regulators won't fund the returns.
5. Founder vs professional manager. Pure professional-manager / lifer-operator archetype (Sideris, 29 yrs internal). Appropriate for a regulated monopoly in heavy-build mode — you want an operator who knows the commissions, not a visionary founder. The implication: execution and regulatory relationships over bold strategic reinvention.

Lens 10 · Forensic Red Flags

Acting as a forensic analyst over the FY2025 10-K and Q1 2026 10-Q.

• Revenue recognition — low risk. Revenue is tariff-set and regulator-approved; fuel is a tracked pass-through. The GAAP-equals-adjusted EPS in 2025 ($6.31 = $6.31) signals no special-item games this year.
• Regulatory assets/liabilities — the structural soft spot of every regulated utility. Large deferred balances (storm costs, coal-ash closure, fuel under/over-recovery) sit on the balance sheet as regulatory assets on the assumption commissions will allow recovery. If a commission disallows a deferred cost, it impairs immediately. This is not fraud — it's the mechanism — but it's where the accounting meets the regulatory risk (Lens 13). Example live items: Florida storm recovery, ~$417M NC coal-ash deferred closure cost sought for recovery, ~$112M Helene securitization.
• Coal-ash / CCR retirement obligations (AROs) — Duke's signature long-tail liability. The 10-K repeatedly addresses Coal Combustion Residuals closure under the Coal Ash Act and federal CCR rules; closure cost estimates are explicitly "uncertain and difficult to estimate" and recovery runs through normal ratemaking. A multi-billion ARO whose ultimate cost and recoverability are both uncertain — watch for upward ARO revisions and any recovery disallowance.
• Goodwill — annual test as of Aug 31, 2025; all reporting units' fair values "materially exceeded" carrying value except the GU&I reporting unit of Duke Energy Ohio, which Duke flags could take an impairment if economic conditions or peer equity valuations deteriorate. This is the one named goodwill watch-item.
• Cash flow vs earnings — the perennial utility tell. Reported segment income ($4,911M) and net income ($5,071M) sit atop $14.0B of capex — free cash flow is deeply negative by design (rate-base-growth utilities outspend operating cash flow and fund the gap with debt + equity). That is the model, not a red flag, but it is why leverage and the Brookfield/equity funding plan are the real risk axis, and why a credit downgrade would be materially negative.
• SBC / non-GAAP — not a material distortion here; the adjusted-vs-GAAP gap is essentially zero in 2025.
• Receivables/inventory outrunning revenue — no flag surfaced; regulated utility working capital is stable and clause-recovered.

Regulatory findings (required sub-section) — sources: SEC EDGAR EFTS (LR + AAER), web search, and 10-K Item 3 / commitments:

• SEC Litigation Releases: None. "No LR found for this company in the search period" (2021-06-21 → 2026-06-21).
• SEC AAERs: None. "No AAER found" in the search period.
• Non-SEC enforcement (web): No material new federal enforcement action (FTC/DOJ/FDA/CFPB/consent-decree/fine) surfaced for the current period. Historically, Duke Energy and subsidiaries have faced significant environmental matters — Clean Water Act/coal-ash litigation and state settlements predating this window — which now run as the regulated coal-ash closure program rather than active enforcement penalties. Treat coal-ash as a regulatory/liability item, not an active enforcement penalty.
• 10-K legal/environmental disclosure: The dominant disclosed matters are coal-ash/CCR closure obligations (Coal Ash Act, federal CCR rules, related agreements) and ongoing rate-case proceedings — Duke's own disclosure of material ongoing regulatory and environmental matters. The NC Court of Appeals ruling against the 2024 Duke Energy Carolinas rate approval is the live adverse legal development.
• Conclusion: No accounting-fraud or SEC-enforcement findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K disclosure as of 2026-06-21. The genuine forensic watch-items are (1) Duke Energy Ohio GU&I goodwill, (2) coal-ash ARO size/recoverability, and (3) regulatory-asset disallowance risk in North Carolina — all regulatory/estimation risks, not integrity red flags.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

Built bottom-up from FY2025 actual adjusted EPS $6.31 and management's framework. Fiscal year = calendar year; FYE Dec 31.

Inputs (labeled):

• FY2025 actual adjusted EPS $6.31.
• Management FY2026 guidance $6.55–$6.80 (mid $6.675).
• Long-term 5–7% EPS CAGR through 2030, "top half" confidence by 2028.
• Rate base $114B (2025) → $180B (2030), ~9.6% CAGR — rate-base growth > EPS growth because equity issuance + the Florida minority dilution share some of the rate-base growth.
• Share count ~778M, rising modestly with ATM equity to fund capex; Brookfield/Piedmont proceeds reduce the equity Duke must otherwise issue.

EPS scenarios ``:

• Base: EPS to ~$7.60 by FY2028 (~6.5% CAGR off $6.31). At a re-rated ~19–20x that supports ~$144–152; at today's 18.9x, ~$144.
• Bull: top-half delivery + a re-rate toward the peer ~20.8x (NEE) on recognized AI-load growth → $7.86 × ~20x ≈ $157.
• Bear: NC disallowances + higher-for-longer rates compress growth to ~4% and de-rate to ~16–17x → $7.08 × ~16.5x ≈ $117 (roughly where the low analyst targets sit, $112–$125).

Forecast log: per --watchlist rules, I am not running forecast.ts create in the breadth loop. The base call to log when promoted: "DUK FY2026 adjusted EPS ≥ $6.65, p≈0.80, resolves 2026-12-31, tags duke-energy,deep-dive."

Lens 12 · Bull vs Bear

Bull case. Duke is a regulated utility with an actual growth catalyst — a rarity. The Southeast/Carolinas + Indiana footprint is exactly where hyperscale AI data centers are landing (Amazon's $10B NC campus, 4.5 GW signed ESAs, 9 GW pipeline), and the load comes with minimum-take contracts that let Duke earn on the generation it builds even if load ramps slowly. That demand justifies the largest capex plan in the industry ($103B), which compounds rate base from $114B to $180B (~9.6%) and underwrites 5–7% EPS growth with "top-half" confidence by 2028. Management funded it cleverly — $6B from Brookfield for a Florida minority + Piedmont-TN sale reduce equity dilution and protect the credit. Yet the stock trades at 18.9x, the cheapest large-cap regulated multiple, with a 3.44% yield. The contrarian re-rate: as the data-center load converts from "pipeline" to "in rate base earning a return," DUK should close the gap to NEE/SO and the market should pay for utility growth, not just utility yield. Catalyst-rich: each ESA, each constructive rate order, each Brookfield tranche.

Bear case (permanent-impairment risks).

1. North Carolina turns structurally hostile. This is the thesis-killer. A commission that disallows rate-base spend or grants sub-allowed ROEs doesn't dent a quarter — it permanently lowers the earnings Duke can extract from its biggest state. The NC Court of Appeals already voided a 2024 Duke Carolinas rate hike, and the pending ~15% request faces political resistance. If "build $103B" meets "but you can't earn your allowed return on it," the entire growth algorithm breaks. The market is pricing exactly this (post-beat stock dip, analyst downgrades on "regulatory uncertainty").
2. Balance-sheet / rates. $87B long-term debt funding a deeply FCF-negative model means a credit downgrade or a higher-for-longer rate regime raises financing cost on every dollar of the $103B plan and compresses the multiple of a bond-proxy stock. Equity issuance to plug the funding gap dilutes the EPS growth the bulls are paying for.
3. Data-center demand disappoints / contracts renegotiate. If the AI-capex cycle cools, hyperscalers slow-walk or restructure ESAs (even with minimum-take, enforcement and timing are real), and Duke is left having pre-built generation into softer load — stranded-asset and disallowance risk compound.

Pre-mortem (18 months out, thesis broke): It's late 2027. North Carolina handed Duke a punitive rate order (low ROE, disallowed storm/coal-ash deferrals), Moody's/S&P put the credit on negative outlook as capex outran funding, and a couple of data-center ESAs got pushed right as the AI build paused. EPS growth slipped to ~4%, the multiple de-rated to ~16x, and the "growth utility" re-rate reversed back to "leveraged bond proxy with regulatory headaches." The stock is ~$115.

Are multiples too high? No — they're arguably too low for the growth profile if NC cooperates. The 18.9x discount to peers is the literal price of the North Carolina binary. This is a regulatory-outcome stock wearing a utility costume.

Contrarian view of what the market refuses to see: The market is so fixated on the North Carolina rate-case overhang that it is under-pricing the durability and contracted nature of the data-center load (minimum-take ESAs in constructive Florida/Indiana jurisdictions, not just NC). The growth is more diversified across jurisdictions than the NC-centric bear narrative implies — and the Brookfield deal proves sophisticated infrastructure capital will fund this rate base at attractive terms even when public-equity holders are nervous.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case.

• Where it structurally breaks: the business is only as good as its worst large commission, and North Carolina — Duke's biggest state — is actively turning against it. The whole "growth utility" thesis assumes Duke earns its allowed ROE on a $103B build. Strip that assumption (Court of Appeals already voided a 2024 NC rate hike; ~15% NC request facing political resistance) and you don't have a growth utility — you have a maximally-levered utility spending $103B at sub-allowed returns. That's value-destructive, not value-creative.
• Concentration risk that's hiding in plain sight: retail customer concentration is trivially low, but regulatory concentration is extreme — North Carolina (Carolinas) is the single largest earnings jurisdiction (Duke Energy Carolinas $9.7B + Duke Energy Progress NC exposure). One state's commission can re-price a huge fraction of the company. The "no customer >10%" stat masks the real concentration: one regulator.
• Why the moat is weaker than bulls think: the franchise monopoly is durable, but the return on that franchise is set annually by politicians' appointees. In a high-bill, high-inflation, post-storm political environment, the path of least resistance for regulators is to deny rate increases — and Duke's bills are rising fast (rate cases + storm recovery + coal-ash recovery + data-center build all hitting customers). Rate fatigue is a real, growing political constraint.
• Most dangerous thing bulls underestimate: the funding stack. A deeply FCF-negative company spending $103B needs continuous debt + equity access at favorable cost. The Brookfield deal is clever because they needed it — it's a tell that straight equity/debt funding for this plan is getting expensive. A rate spike or a downgrade turns the growth plan into a dilution machine. And the Florida minority sale itself gives away 19.7% of the economics of the best (most constructive) jurisdiction to Brookfield — bulls cheer the cash, but Duke just sold a fifth of its best business.
• Worst capital-allocation history: the Commercial Renewables exit destroyed billions ($1.7B disposal loss in 2023 alone ) — management's prior strategic bet was wrong and shareholders paid to unwind it. That's the same team now sizing a $103B bet on data-center load forecasts.
• What must hold for today's price: NC delivers roughly allowed returns; rates don't spike; the credit holds investment grade through the build; data-center ESAs convert on schedule. Break any one and the 18.9x looks generous, not cheap.
• If growth disappoints 20–30%: EPS CAGR to ~4% + de-rate to ~16x → ~$115–117, ~7% below spot with the dividend as the only cushion. The downside is real but cushioned by the regulated floor — this is not a zero, it's a dead-money-with-derating risk.
• Single permanent-impairment scenario: a sustained adverse North Carolina regulatory regime (multi-cycle sub-allowed ROEs + deferral disallowances) — plausible and already partly materializing. Not catastrophic, but genuinely thesis-breaking for the growth case.

Lens 14 · Management Questions (ordered by information value)

1. On the pending North Carolina rate case and the Court of Appeals remand — what allowed ROE and capital-structure outcome are you actually underwriting in the $103B plan, and what is your contingency if NC delivers materially below your allowed return?
2. Of the $103B 2026–2030 capex, how much is contractually de-risked by minimum-take data-center ESAs versus speculative on forecast load — and what are the minimum-take terms (floor %, duration, credit quality of the counterparties)?
3. What does the funding stack for the $103B plan look like year by year — debt vs ATM equity vs asset-sale proceeds — and what equity issuance is embedded in your 5–7% EPS growth guidance?
4. After the Brookfield Florida minority sale and Piedmont-TN, are further minority-stake or asset sales contemplated, and at what point does selling economics of your best jurisdictions undermine the growth story you're funding?
5. What credit metrics (FFO/debt) are you committed to defending through the build, and what is your hard floor before you'd cut capex rather than risk a downgrade?
6. Which is the binding constraint on realizing the data-center load — interconnection/transmission, generation supply chain (turbines/transformers), or regulatory approval of the associated spend?
7. What is the current best estimate of total remaining coal-ash/CCR closure cost, the recoverable portion, and your exposure if a commission disallows any of it?
8. On the Duke Energy Ohio GU&I goodwill you flagged as near its carrying value — what specific conditions would trigger an impairment, and how large could it be?
9. How exposed is the capital plan to tariffs and rare-earth/transformer supply constraints, and how much of the equipment is locked under framework agreements at fixed cost/schedule?
10. What gives you confidence in the "top half" of the 5–7% range by 2028, and what would push you to the bottom half?
11. How do you manage rate fatigue — bills rising from rate cases, storm recovery, coal-ash recovery, and data-center build simultaneously — before it becomes a political ceiling on approvals?
12. Given storm frequency (Helene/Milton/Debby), how should we think about normalized annual storm O&M and the reliability of securitization recovery going forward?
13. Where is your nuclear fleet in its license/relicensing and capex cycle, and what role do SMRs or new gas play in meeting data-center load this decade?
14. With NextEra–Dominion combining, how do you view scale consolidation in regulated utilities — is Duke a participant in any form, or strictly a standalone operator?
15. As a 29-year insider now in the CEO seat, what is the single capital-allocation mistake of the prior era you are most determined not to repeat?

---