CenterPoint Energy

Phase A — Understand the business

Company Overview

CenterPoint Energy is a pure-play, fully-regulated electric and natural-gas delivery utility holding company — a "wires-and-pipes" business, not a merchant generator or commodity trader. It explicitly states it is "first and foremost an energy delivery company" intending "to remain focused on these regulated segments". It makes money the way regulated utilities do: it invests capital into rate base (poles, wires, substations, mains), and state regulators (the PUCT in Texas, the Railroad Commission of Texas for gas, the IURC in Indiana, plus Minnesota/Ohio commissions) set rates that let it earn an authorized return on that capital plus recover its costs.

Structure — three reporting layers, two operating segments:

• Electric segment = Houston Electric (CenterPoint Energy Houston Electric, LLC — transmission & distribution for the greater Houston metro, ~2.9M metered customers; it does not own generation in ERCOT, it delivers power on behalf of Retail Electric Providers) + Indiana Electric (SIGECO — a vertically-integrated utility that does own generation, ~150k customers). FY2025 Electric revenue $4,866M, net income $705M.
• Natural Gas segment = CenterPoint Energy Resources Corp. (CERC) — gas LDC distribution across Texas, Indiana, Minnesota and Ohio. FY2025 net income $570M.
• Corporate & Other — unallocated corporate costs, interest, eliminations: −$223M FY2025.

Customers / suppliers / competitors. As a delivery monopoly, CNP has no direct competitor in its franchise territories — its "competition" is regulatory (can it earn its allowed return?) and capital-markets (cost of debt/equity). Its true counterparties are (a) the regulators who grant rate increases, and (b) the large new loads — data centers, LNG/refining/petrochemical export facilities — queuing to interconnect in Houston. Contract structure is rate-regulated tariff revenue, not take-or-pay; ~85% of projected consolidated investment is recovered through interim capital trackers (DCRF, TCOS, CSIA, GRIP) or forward-test-year rate cases, which materially compresses regulatory lag. 8,794 employees, 3,712 unionized.

The one-line model: the regulated wires monopoly for the fastest-growing major-metro electricity demand pool in America (Houston), monetised through a record $65.5B 10-year capex plan.

Supply Chain

A T&D utility's "supply chain" is its rate-base build-out and its fuel/equipment inputs. Named stakeholders along the chain:

Upstream (what CNP buys):

• Capital equipment & EPC — transformers, switchgear, transmission steel, smart meters, and the contractors who install them. The whole $65.5B plan depends on a transformer/grid-equipment supply chain that is industry-wide constrained (lead times for large power transformers have stretched globally on AI-datacenter demand).
• Natural gas (for the gas LDC + Indiana generation) — CERC buys wholesale gas; it carries $215M natural-gas-and-coal inventory at YE2025. The February 2021 Winter Storm Uri event is the cautionary tale: gas-supply disruption created ~$2B of extraordinary gas-cost recovery.
• Indiana Electric generation fuel — coal (being retired; A.B. Brown securitized) → gas + solar (Posey Solar, 191 MW, acquired from Arevon for ~$357M March 2025, in service Q2 2025).
• TEEEF mobile generation — Houston Electric leases up to 519 MW of mobile/temporary emergency generation under 2021 Texas legislation; the large-unit lease (the "$800M" fleet) runs to 2029. This input is the source of the Beryl scandal (Lens 8/10/13).

The company → end customer:

• Electric: Houston Electric delivers power from ERCOT generators/the grid to ~2.9M premises; it bills Retail Electric Providers (REPs), not end consumers directly (Texas's deregulated retail model). REP default risk is a real line item (bad-debt regulatory assets recorded).
• The demand-side chokepoint that matters: the interconnection queue. Houston-area peak load is forecast +~50% by 2029, ~10 GW of incremental peak by 2031, with 8 GW of data-center load targeted to energize by 2029 plus ~4.2 GW more, 3.5 GW under construction, and 12.2 GW of "firmly committed" new load. CNP reported a 700% increase in data-center interconnection requests in Texas. Refining/LNG exports + AI data centers = 40-60% of the demand growth.

Single-source / chokepoint flags: (1) large-power-transformer lead times (industry-wide); (2) geographic concentration in a single metro and a single regulator — Houston/PUCT is both the moat and the single point of failure; (3) the ERCOT-only, generation-light Texas model means CNP is a delivery toll-taker, insulated from power-price risk but fully exposed to storm/reliability execution in a hurricane-prone coastal city.

Competitive Advantages (moats)

The moat is a regulated geographic monopoly sitting on top of the best demand-growth zip code in the US utility universe. Specifically:

1. Franchise monopoly + regulatory recovery design. No one else can string wires in Houston Electric's certificated territory. More important than the monopoly itself is the recovery mechanism: ~85% of rate base has been through a rate case since 2023 (supporting clarity through 2029), and ~85% of investment is recovered via interim trackers or forward-test-year cases. Forward-test-year + interim trackers = low regulatory lag = the capex actually earns close to its allowed return in near-real-time. This is a structurally better regulatory construct than many peers operating under historical-test-year regimes.

2. Demand tailwind as a moat. Most US utilities are fighting flat-to-1% load growth and have to manufacture rate base. CNP has the opposite problem — it has more economically-justified investment than it can physically build. Houston is the petrochemical/LNG-export capital of the US and a top AI-data-center landing zone. That converts "rate base growth" from a regulatory negotiation into a genuine economic necessity, which is the single most defensible position a utility can hold.

3. Bargaining power. Over customers: high (monopoly, non-bypassable charges). Over suppliers: low-to-moderate (price-taker on transformers/gas in a constrained market). Over regulators: this is the crux — its bargaining power eroded materially after Hurricane Beryl (Lens 8/13). A utility's real moat is the regulator's goodwill, and CNP spent some of it in 2024.

Durability: high on structure, with a reputational dent. The franchise and the Houston demand pool are about as durable as moats get in this sector; what's contestable is whether management keeps the regulatory relationship constructive enough to actually earn the allowed returns on $65.5B of spend.

Segments

FY2025 revenue and net income by reportable segment [all research-layer: filings/10-k-2025-q4.md, lines 1088-1091, 1113]:

Consolidated revenue: utility $9,301M + non-utility $56M = $9,357M FY2025 (vs $8,643M FY24, $8,696M FY23).

What's moving and why: Electric is the growth engine — the Houston Electric rate case (final order March 2025: 9.65% ROE, 43.25% equity ratio) reset rates upward effective 28-Apr-2025, and customer growth contributed ~$26M/yr. Natural Gas net income looks flat because the denominator shrank — CNP sold its Louisiana and Mississippi gas LDCs (closed 31-Mar-2025, ~$1.2B divestiture proceeds). The "current assets held for sale" of $2,669M at YE2025 (vs $1,361M YE2024) signals another large gas-LDC divestiture in progress (CNP has been pruning gas to concentrate on Texas electric — a deliberate portfolio rotation toward the highest-growth, best-regulated assets). Geographically the company is consolidating around Texas + Indiana + Minnesota and exiting the Gulf-state gas footprint.

---

Phase B — Measure performance

Earnings Result (latest print — Q1 2026, filed 2026-04-23)

Drivers. Houston Electric net income rose to $96M (from $84M) on rate-case rates and load. CERC net income fell to $230M (from $305M) — almost entirely the LA/MS divestiture losing a quarter of contribution. Management called out weather/usage −$0.02 (mild winter across TX/IN) and interest expense −$0.04 (new debt issuance to fund the build) as the headwinds against the rate-base tailwind.

Guidance. Reiterated full-year 2026 non-GAAP EPS of $1.89–$1.91 (≈8% growth over 2025 delivered) and the $65.5B plan; ~$6.8B of capex planned for 2026. Tone: confident, load-growth-forward — they raised the Houston load forecast and pulled the +50% peak milestone forward to 2029 (two years early).

Balance-sheet flags. This is a capital-intensive, levered utility and the cash statement shows it plainly (FY2025):

• Operating cash flow $2,486M; capex $4,870M → free cash flow ≈ −$2,384M. CNP does not self-fund its growth capex — it is structurally FCF-negative and funds the gap with debt + equity. This is normal and expected for a high-growth regulated utility, but it makes the name rate-sensitive and dilution-prone.
• Total debt ≈ $23.6B (short-term $500M + current portion of LT debt + securitization $1,914M + LT debt net $20,566M) against equity of $11,153M → debt/cap ≈ 68%. Heavy, but again typical for the sector and supported by regulated cash flows.
• Receivables +$253M use of cash (load growth) [line 2080]; goodwill $3,550M (down from $3,943M — written down/reallocated with the LA/MS sale) [line 2015].
• Equity securities mark: a −$51M loss on equity securities and a −$49M loss on sale hit FY25 other-income (the equity-securities line is the legacy ZENS/AT&T-Charter indexed-debt complex — noise, not operations) [lines 1952-1953].

Unusual vs. own history: nothing alarming. The story is consistent: rate cases + load lifting Electric, gas divestitures shrinking the gas line on purpose, interest expense the growing drag.

Earnings Calls (sentiment trend)

No transcripts on the research-layer shelf (transcripts/ empty), so this is -sourced from quarterly call coverage.

Tone across the last ~4 calls (Q3 2025 → Q4 2025 → Q1 2026) has been steadily, increasingly bullish on Houston load, and that is the dominant message management keeps escalating:

• Recurring/escalating phrases: "industry-leading growth," "premium growth at premium scale," the 8% EPS algorithm, "firmly committed" load (the number keeps rising — to 12.2 GW), data-center/large-C&I interconnections, "resiliency" and "grid hardening".
• What got louder: the load-growth forecast (peak +50% milestone pulled forward two years to 2029); the capital plan (raised twice — Sept 2025 to $65B, Feb 2026 to $65.5B).
• What they downplay now vs. 2024: the Beryl post-mortem. In late-2024 calls Beryl response and the "42 critical initial actions" dominated; by 2026 the narrative has been successfully pivoted to load growth and resiliency investment (turning the storm liability into a capex tailwind).

Net: management has executed a clean narrative reset from "storm-response problem child" to "AI/petrochemical load-growth darling." The sentiment is genuine and backed by real interconnection data — but the tone is also notably promotional (Lens 9/13 flag).

Comps

CNP + key US integrated/regulated-utility peers. Multiples are, dated; where a clean figure wasn't sourced it is marked n/a rather than fabricated. Forward P/E figures.

ROE figures; CNP's ~9.6% is an cross-checked to filings (FY25 income to common $1,052M over ~$10.9B average common equity).

Read: CNP trades at a clear premium — ~23x forward vs. an integrated-utility peer set clustered at ~17-21x, with only NEE in the same multiple zip code. Trailing P/E is even starker at ~26-27x vs. peer average ~22x and industry ~19x. The premium is for the growth (8% EPS algo, the strongest large-metro load pool in the country). The question Lens 12/13 must answer: is ~8% EPS growth + ~2% yield = ~10% total return enough to justify paying 23x when Sempra (also Texas, also LNG/data-center, arguably cheaper risk) is at ~18x? CNP's yield (~2.1%) is the lowest in the peer group — you are buying growth, not income.

Stock-Price Catalysts (5-yr — what actually moves CNP)

Mostly; the pattern is unusually clear for this name.

• Feb 2020 — dividend cut + Enable crisis (down hard). Pre-pandemic, CNP cut its quarterly dividend ($0.2875 → $0.15) under balance-sheet stress tied to its Enable Midstream stake — the trough that the current management team was hired to fix. (NB: a secondary source claims a separate "46% dividend cut in 2025 to fund a $53B plan" — this conflicts with the filed record, which shows dividends paid of $574M FY25 ($0.88/sh) and a current $0.92/yr rate that was recently raised. Treat the "2025 cut" framing as unreliable/garbled; the real cut was 2020.)
• 2021 — capital-plan pivot. Launch of the utility-focused $40B+ 10-year plan and Enable/midstream exit re-rated the stock from value-trap to regulated grower.
• Feb 2021 — Winter Storm Uri. ~$2B gas-cost spike; ultimately securitized/recovered, so a scare more than a permanent impairment.
• July 2024 — Hurricane Beryl (the defining negative catalyst). ~2.3M Houston customers out, some >1 week; ~80% of the Texas service territory lost power; the $800M TEEEF mobile-generator fleet was largely unused/ineffective. Triggered a Gov. Abbott-directed PUCT investigation (final report 21-Nov-2024: deployment "inefficient," fleet "not right-sized for a Beryl-type event"), a Texas AG investigation into possible "fraud, waste," and a Lt. Gov. demand to claw back the $800M. This is the single biggest reputational/regulatory overhang on the name.
• 2025-2026 — the load-growth re-rate (up). Each upward revision to the Houston load forecast and the $65/$65.5B plan has supported the premium multiple.

What the tape reveals: CNP reacts to (1) regulatory/political risk events (Uri, Beryl, dividend/balance-sheet) far more than to quarterly EPS beats/misses (the Q1 2026 ~1¢ miss barely mattered), and (2) load-growth/capital-plan revisions. It is a regulatory-relationship and demand-narrative stock, not an earnings-surprise stock. The premium multiple is hostage to Texas politics.

---

Phase C — Judge people & books

Management

• CEO Jason Wells — appointed CEO 5-Jan-2024, having joined CNP as EVP & CFO in 2020 and served as President/COO before stepping up. Prior: 13 years at PG&E, rising to EVP & CFO of the ~$60B enterprise; credited with leading $50B+ of capital-markets and strategic transactions. So: a finance-/capital-allocation-bred CEO, not an operator-from-the-trucks. That cuts both ways — superb at the capital-formation and rate-case-strategy game that an $65.5B plan demands; the PG&E pedigree is double-edged given that PG&E is the canonical wildfire/operational-disaster utility (though Wells was CFO, not ops).
• Track record at CNP: the team (Wells + Chairman Dave Lesar, who preceded him as CEO) executed the value-creating turnaround — exited Enable Midstream, sold non-core gas (Energy Systems Group 2023; Louisiana/Mississippi LDCs 2025; another LDC sale in progress per the $2.67B held-for-sale), redeemed the expensive Series A preferred, and rebuilt around a pure-regulated, Texas-tilted growth story. Net income rose from $867M (FY23) to $1,052M (FY25) and the plan ramped from ~$40B to $65.5B. That is a genuine, quantified re-rating of the franchise.
• Skin in the game / capital allocation: insider ownership data not on the research-layer shelf (our figures absent) — n/a. Capital-allocation behavior is the more telling read: disciplined portfolio pruning + funneling proceeds into the highest-return regulated rate base, ~85% of which is covered by constructive recovery mechanisms. ROE ~9.6%; the bet is that the $65.5B deploys at roughly the ~9.65-10.4% allowed ROEs being set in current rate cases.
• Red flags: (1) the Beryl execution failure happened on this team's watch — the $800M generator fleet they leased proved ineffective when it mattered, which is both an operational and a capital-allocation black mark (Lens 13); (2) promotional tone — "industry-leading / premium growth at premium scale" is laid on thick, and the company has a history of needing to walk back over-promises; (3) modest but steady equity dilution to fund the build (WASO 631M→656M FY23→FY25).
• Archetype: professional-manager / financial-engineer team, mid-execution on a credible regulated-growth thesis. Right archetype for the capital-formation challenge; the open question is operational reliability in a hurricane zone.

Forensic Red Flags

Grounded in the FY2025 statements + the regulatory file. Every figure labeled.

• Cash flow vs. earnings: net income $1,052M but OCF $2,486M and capex $4,870M — earnings quality is fine (OCF > NI, normal for a D&A-heavy utility; D&A alone was $1,530M). The "flag" isn't quality, it's the structural FCF deficit (≈−$2.4B) funded by perpetual debt+equity issuance. Sustainable only as long as capital-markets access stays open and rates don't spike — a real but sector-normal risk.
• Receivables outrunning revenue: receivables/unbilled were a $253M use of cash vs. revenue +$714M — directionally consistent with load growth, not a recognition red flag.
• Regulatory assets — the utility-specific watch item: $3,005M non-current + $170M current regulatory assets. These are costs the company has capitalized in the expectation regulators will let it recover them (Uri gas costs $70M, TEEEF lease costs ~$201M, storm restoration, bad debt). If a regulator ever disallows a chunk — a live risk given the Beryl politics around the TEEEF leases — it converts to a write-off. The TEEEF regulatory assets ($78M short-term + $123M long-term lease costs deferred) are the ones to watch given the clawback pressure. CNP states it has recorded no impairment on the TEEEF assets as of YE2025 — i.e., it is still assuming full recovery of generators that politicians want clawed back. That is the sharpest forensic flag in the file.
• Goodwill: $3,550M (down from $3,943M with the LA/MS sale) — sits against $11.2B equity; not trivial, but supported by regulated cash flows; no impairment taken.
• SBC / non-GAAP gap: the ~$0.08 wedge between Q1 2026 GAAP EPS ($0.48) and non-GAAP ($0.56) is the standard utility adjustment set (M&A/divestiture, mark-to-market on the indexed-debt/equity-securities complex); worth noting that the company guides and is valued on non-GAAP, which runs ~15-17% above GAAP this quarter — investors should anchor on GAAP for through-cycle earnings power.
• Indexed-debt/ZENS complex: the −$51M equity-securities loss / +$55M indexed-debt-securities gain are offsetting legacy noise, not operations.

Regulatory findings (required sub-section).

• SEC Litigation Releases: None. No LR naming CenterPoint Energy in the 2021-06-29 → 2026-06-29 window.
• SEC AAERs: None in the same window.
• Non-SEC enforcement (web): The material item is not a federal financial-enforcement action but state utility/political scrutiny of the Hurricane Beryl response: a Gov. Abbott-directed PUCT investigation (final report 21-Nov-2024 — deployment "inefficient," fleet "not right-sized"), a Texas Attorney General investigation into possible "fraud, waste" in the storm response, and Lt. Gov. Patrick's demand that the PUC claw back the $800M TEEEF lease. No criminal/financial-fraud finding has resulted to date; the exposure is rate-case/disallowance risk and reputational, not securities-law.
• 10-K Item 3 / Note 14 (Legal Proceedings): CNP discloses material litigation including February 2021 Winter Storm Uri proceedings (Note 14(c)) and the general litigation risk language. Uri-related civil litigation against multiple ERCOT-area utilities remains outstanding.
• Verdict: No accounting-fraud or SEC-enforcement findings (clean on the books per EDGAR EFTS LR/AAER + 10-K Item 3 as of 2026-06-29). The genuine regulatory risk is state-level cost-disallowance (TEEEF/Beryl, Uri) — a recoverability risk sitting inside the $3.0B regulatory-asset balance, not a financial-reporting-integrity risk.

---

Phase D — Project & stress-test

Forward Projection (FY2026 / FY2027 / FY2028 non-GAAP EPS)

Built bottom-up from FY2025 delivered non-GAAP EPS and the company's own algorithm. FY2025 delivered ≈ $1.75 non-GAAP.

The growth engine is mechanical: rate base compounds as the $65.5B plan deploys (~$6.5B/yr average), most of it earning ~9.65-10.4% allowed ROE with low lag, funded ~50/50 debt/equity. EPS growth = rate-base growth × allowed ROE − interest drag − ~1% dilution. Management's stated algo is 7-9% through 2028, 7-9% thereafter to 2035.

Arithmetic shown for base: $1.90 × 1.08 ≈ $2.06; $2.06 × 1.08 ≈ $2.22. The path is unusually legible because regulated utilities convert capex → rate base → EPS on a known formula; the variance is almost entirely (a) allowed ROE outcomes and (b) interest expense, both of which the bear case stresses.

Per --watchlist rules, no Brier forecast logged in this loop. Candidate to log on promotion to our position log: "CNP FY26 non-GAAP EPS ≥ $1.89, resolves 2026-12-31, p≈0.80."

Bull vs Bear

Bull case. CNP owns the regulated wires of the single best electricity-demand-growth metro in America at exactly the moment AI data centers + LNG/petrochemical export expansion are colliding in Houston. Peak load +~50% by 2029 (pulled forward two years), 12.2 GW firmly committed, 700% jump in data-center interconnection requests — this is not a manufactured rate-base story, it's economically forced investment. The $65.5B plan, ~85% covered by constructive recovery mechanisms, converts that demand into a visible 7-9% EPS / ~10% total-return algorithm through 2035 with a regulated, recession-resistant cash-flow base. Management has already proven it can high-grade the portfolio (Enable exit, preferred redemption, gas divestitures) and raise the plan twice. The contrarian read bulls lean on: the Beryl overhang is transient politics that, ironically, increases the resiliency capex CNP gets to earn on — the scandal becomes a tailwind.

Bear case (permanent-impairment lens). Three risks that could break it:

1. Regulatory/political backlash converts to dollars. The Beryl response shredded CNP's standing with Texas politicians and the PUCT. The $800M TEEEF fleet sits on the balance sheet as a ~$201M recoverable regulatory asset that CNP is still assuming full recovery on — a disallowance, a punitive ROE in the next case, or legislated affordability caps would hit both EPS and the multiple. A utility's entire value rests on the regulator letting it earn; CNP burned goodwill.
2. The multiple is the risk. At ~23x forward / ~26x trailing, CNP prices in flawless execution of the 8% algo. Peers with arguably similar Texas/data-center exposure (Sempra ~18x, AEP ~21x, Dominion ~17x with a 4%+ yield) are cheaper. If load timing slips, a rate case disappoints, or rates stay high, the de-rate from 23x → 19x is a ~17% headwind that swamps a year of EPS growth.
3. Funding/rate sensitivity. Structurally FCF-negative (≈−$2.4B FY25), CNP must continuously issue debt and equity. Higher-for-longer rates raise the interest drag (already −$0.04/qtr in Q1 2026) and the cost of the equity it dilutes shareholders with.

Pre-mortem (18 months out, thesis broke): A 2027 Houston Electric rate case lands with a sub-9.5% ROE and a partial TEEEF disallowance; the Texas legislature, still smarting from Beryl, imposes affordability constraints; data-center interconnections slip on transformer lead times; the stock de-rates to ~18x on $1.95 EPS → ~$35, a ~20% drawdown, and the "premium growth" narrative cracks.

Are multiples too high? Probably modestly, relative to the risk. The growth is real, but at ~23x you are paid little for the regulatory-tail and execution risk that this specific name carries more of than its peers. Contrarian view the market is refusing to see: CNP is being valued as a clean AI-data-center compounder (like NEE) while still carrying an unresolved Texas-political/cost-disallowance tail that NEE/AEP don't — the market has already "moved on" from Beryl that the regulator and legislature have not.

Devil's Advocate (short-seller)

Dismantling the bull case.

• Where revenue concentrates / what breaks it: geographic single-point-of-failure. CNP's growth thesis is ~entirely Houston Electric. One bad hurricane season + one hostile PUCT/legislature reaction can re-rate the whole company. It already happened once (Beryl). The "moat" (a single regulator's goodwill) is exactly what cracked in 2024.
• Why the moat is weaker than bulls think: a regulated monopoly is only as good as its allowed return and the regulator's willingness to grant it. CNP spent its regulatory capital on Beryl. The $800M generator debacle gives every consumer advocate and politician a cudgel in the next rate case. Bulls treat 9.65% ROE as a given; it's a negotiation with a counterparty CNP just angered.
• Most dangerous competitor bulls underestimate: not a rival utility — it's the affordability/political constraint. Texas politicians can cap, claw back, or disallow. The "competitor" is the customer's electricity bill and the PUC's response to it.
• Worst capital-allocation move: leasing an $800M mobile-generation fleet that didn't work when the city needed it — and still carrying the costs as a recoverable asset. That is the textbook misaligned-incentive / capitalize-and-pray red flag +.
• What must hold for today's price: (a) the 8% EPS algo lands every year through 2028; (b) rate cases clear at ~9.65-10.4% ROE; (c) no material TEEEF/Uri disallowance; (d) ~$6.5B/yr of capex stays financeable without a rate shock; (e) the multiple stays ~23x. Miss two of these and the stock is a high-teens-P/E name.
• If growth disappoints 20-30%: EPS algo to ~5-6%, multiple to ~18x → ~$33-35, a 20-25% drawdown. The premium is the whole downside.
• Single scenario that permanently impairs: a catastrophic, poorly-handled second major storm that triggers legislated structural changes to Houston Electric's recovery model (or, tail-risk, a forced restructuring of the franchise). Plausibility: low-moderate — Houston will get more hurricanes; whether the next response is botched is an execution bet on this team, which failed the last test.

Management Questions (ordered by information value)

1. On the TEEEF/mobile-generation fleet: you carry ~$201M of lease costs as a recoverable regulatory asset and have taken zero impairment — what is your probability-weighted view of full recovery given the PUCT findings and the clawback pressure, and at what point do you reserve?
2. What is the worst-case ROE you are underwriting in the next Houston Electric rate case, and how much of the $65.5B plan still earns the 8% algo if you land 50-75 bps below the last 9.65% settlement?
3. Of the 12.2 GW "firmly committed" load, how much is contractually binding with financial commitments (take-or-pay, deposits) vs. interconnection requests that can evaporate — and what's the cancellation experience so far?
4. How exposed is the capex schedule to large-power-transformer and switchgear lead times, and what slips if equipment delivery stretches another 6-12 months?
5. You're structurally FCF-negative by ~$2.4B/yr — what is your equity-issuance assumption embedded in the 8% EPS algo through 2035, and at what cost of equity does the algo break?
6. What did Beryl change, concretely, about how you and the PUCT/legislature interact — and what have you done to rebuild regulatory goodwill beyond the "42 actions"?
7. Which remaining gas LDC is in the $2.67B held-for-sale, what's the expected after-tax proceeds, and how are they redeployed — buyback, debt paydown, or Texas rate base?
8. How should investors think about the ~15% GAAP-to-non-GAAP EPS wedge through the cycle — what's the run-rate of adjustments once divestitures are done?
9. With the lowest dividend yield in your peer group (~2.1%), what is the dividend-growth policy relative to the 8% EPS algo, and is buyback ever competitive with rate base for capital?
10. Indiana Electric's coal-to-gas/solar transition: what's the residual stranded-cost and securitization exposure, and how does IRA/tax-credit policy risk affect it?
11. What is the interest-rate assumption underpinning the plan, and how much EPS sensitivity per 100 bps of sustained higher financing cost?
12. Where are you on undergrounding / hardening Houston's most outage-prone circuits, and what's the cost-benefit the PUCT has actually approved vs. what you've proposed?
13. Uri civil litigation (Note 14(c)) — what's the residual exposure not covered by securitization/insurance?
14. As a finance-bred leadership team, how do you ensure operational/storm-execution rigor — what changed in the ops org and accountability after Beryl?
15. What would make you slow the capital plan rather than raise it again?

---