CenterPoint Energy, Inc. (NYSE: CNP) — The Best Load-Growth Story in the Sector, Priced at Its Own Ceiling

Independent fundamental research. Report date: 2026-07-03.

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not a solicitation and not investment advice. The analysis in the sections below takes no position and carries no price target; it discusses valuation only as embedded expectations and scenarios.

Verdict: HOLD / own-for-the-algorithm-not-the-multiple / accumulate-on-weakness / not-a-short. Medium conviction. Fair-value zone ~$38–45 (≈18–21× FY27E non-GAAP EPS of ~$2.05–2.10, ~2.3–2.6× book); at ~$44.61 — an all-time high — the stock sits at the very top of fair value, and the multiple is doing none of the work for you from here. You are underwriting ~8% EPS growth plus a ~2% yield ≈ ~10%/yr total return, but only if the richest multiple in the company’s own history holds. Accumulate aggressively only on a rate-driven or credit-driven de-rate into the high-$30s.

CenterPoint owns the single best organic demand story in US regulated utilities. Its Houston Electric wires monopoly sits on top of the fastest-growing large-load corridor in the country — data centers, AI, advanced manufacturing, LNG/energy exports, and 2%-a-year population growth — and management now points to 12.2 GW of “firmly committed” new industrial load and a Houston peak that it expects to rise ~50% by ~2029 and nearly double by the mid-2030s. That underwrites a record $65.5B ten-year capital plan, ~11% rate-base growth, and a 7–9% non-GAAP EPS algorithm the company has extended all the way to 2035 — one of the longest, most visible runways in the sector, and one CNP has actually been hitting. The regulatory toolkit is constructive (~85% of investment recovered through capital trackers with minimal lag), earned ROE (~9.6%) sits right at the allowed return, and the post-Enable, post-midstream portfolio is now ~100% regulated and getting simpler still (Louisiana/Mississippi gas sold; Ohio gas exiting). This is a genuine quality-growth utility, not a value trap.

The problem is the price of that quality, and three asterisks the bull case waves away. First, valuation: at ~23.5× current-year and ~21.5× FY27E non-GAAP EPS, ~14× forward EV/EBITDA, 2.6× book and a 2.06% dividend yield — the lowest in its large-cap peer group — CNP trades at the 95th percentile of its own decade (99.9th on price-to-sales), essentially tied with Entergy for the richest forward multiple in the cohort while offering less yield than any of them. Second, the balance sheet is the binding constraint, not the accelerant: a Baa2/BBB holdco with a Moody’s negative outlook, 92% debt-to-cap, and FFO/debt of ~12.5% against a ~13–14% downgrade threshold — all as it embarks on the largest capital plan in its history, funded by perpetual equity dilution (shares 593M→656M) and debt. Third, this is a company whose Houston grid failed catastrophically in Hurricane Beryl just two years ago — 2.2M customers dark for days, deaths tied to the outage, an $800M lease on mobile generators that were useless when it mattered, an AG fraud probe — and whose political standing, while largely repaired, is not de-risked. And the tell that should give a bull pause: in ERCOT, CNP only owns the wires, so the 12.2 GW of committed load does not directly drive its capex (the large customers pay for their own interconnections); it drives demand charges and indirect, still-to-be-defined future transmission — the load headline and the rate-base build are more loosely coupled than the narrative implies.

The framing is quality-growth-compounder-at-a-full-price / crowded defensive bid — emphatically not a falling knife (the stock is at its all-time high, beta 0.18, max drawdown of just ~7% over the last year) and not deep value (the factor model reads it anti-value, negative-Growth, positive-LowVol, sitting within 1% of its 12-month relative-strength peak). It is a low-volatility, momentum-adjacent defensive compounder that the market has bid to perfection on the AI-power-demand theme and the rate-cut tailwind. Own it for the durable per-share algorithm; do not expect the multiple to keep paying you. Conviction: medium. Flips bullish on a de-rate into the high-$30s (where ~8% growth + a ~2.4%+ yield compounds attractively) or hard evidence that energized — not queued — load is structurally lifting the growth rate above 9%. Flips bearish on a Moody’s downgrade / FFO-to-debt breach forcing a larger-than-planned equity slug, an adverse Texas regulatory turn (an ROE cut or a resiliency-capex clawback), or a higher-for-longer 10-year yield that compresses the entire bond-proxy complex. Catchy version: the best growth in the group, at the thinnest yield in the group — a wonderful utility priced like there’s no such thing as a bad year in Houston.

📈 Stock Price Action — Five-Year Event Map

CenterPoint did not round-trip; it re-rated. The stock climbed almost without interruption from roughly $24–25 in mid-2021 (fresh out of the 2020 Vectren/Enable/dividend-cut near-death, when it briefly traded below $12) to an all-time high of ~$45.04 on 2026-06-26, closing $44.61 on 2026-07-02 — barely off the high, inside a 52-week range of ~$35–45. Over five years the shares roughly doubled, a strikingly low-volatility ascent (beta 0.18; 1-year max drawdown only ~7%). The engine was not earnings heroics — GAAP EPS went from ~$1.13 (continuing ops, 2021) to ~$1.61 (2025) — it was a wholesale multiple re-rating: trailing P/E expanded from ~12× to ~24×, and EV/EBITDA from ~10.9× to ~13.0×, as the market re-cast CNP from a broken, midstream-encumbered dividend-cutter into a premium, pure-play Houston load-growth utility. Notably, the stock barely flinched at Hurricane Beryl — the defining operational disaster of its recent history.

The price moves are facts; the attributed drivers are interpretation, cross-referenced to earnings dates, 8-K events, the capital-plan announcements and the rate cycle. No price target, support/resistance, or chart-pattern reading is implied — the opportunity judgment lives in Claude’s Take above.

1. Executive Summary

CenterPoint Energy is a ~$29B-market-cap, ~100% rate-regulated electric-and-gas utility holding company built around one exceptional asset: CenterPoint Houston Electric, the transmission-and-distribution (wires-only) monopoly serving ~2.86 million metered customers across the Texas Gulf Coast, including the city of Houston. Around that core sit a vertically integrated Indiana electric utility (~154k customers, its own generation) and natural-gas distribution utilities in Texas, Minnesota and Indiana (Ohio gas is being sold, expected to close Q4-2026; Louisiana and Mississippi gas were sold in 2025). FY2025 revenue was $9.36B (Electric $4.87B, Natural Gas $4.48B); net income $1.05B; GAAP diluted EPS $1.60 (non-GAAP ~$1.75).

The investment case is unusually clean for a utility: CenterPoint has the best organic volume-growth backdrop in the sector. Houston is the epicenter of the US large-load boom — data centers, AI, advanced manufacturing (Gulf Coast is building much of the physical equipment that goes into data centers), energy and LNG exports, plus ~2%/yr population growth. Management has raised its “firmly committed” new industrial load to 12.2 GW and expects Houston peak demand to climb ~50% to over 30 GW by ~2029 and nearly double by the mid-2030s. That underwrites a record $65.5B ten-year capital plan (2026–2035), ~11%+ rate-base growth, and a 7–9% non-GAAP EPS growth algorithm extended through 2035 — a runway few peers can match, delivered through a constructive, tracker-heavy regulatory framework (~85% of investment recovered with minimal lag) at an earned ROE (~9.6%) that sits right at the allowed return.

Three things temper the enthusiasm. First, valuation: at ~$44.61 (an all-time high) CNP trades at ~23.5× current-year / ~21.5× FY27E non-GAAP EPS, ~14× forward EV/EBITDA, 2.6× book, and a 2.06% dividend yield — the lowest in its large-cap peer group — placing it at the 95th percentile of its own ten-year valuation history (99.9th on price-to-sales). It is essentially tied with Entergy for the richest forward multiple in the cohort while yielding less than any peer. Second, the balance sheet is the binding constraint on the plan, not its fuel: a Baa2/BBB holding company on Moody’s negative outlook, ~92% debt-to-capital, and FFO/debt of ~12.5% against a ~13–14% downgrade threshold, all funded by perpetual equity dilution (share count 593M→656M since 2021). Third, execution and political risk are real and recent: Hurricane Beryl (July 2024) left 2.2M customers without power for days, with deaths tied to the outage, an $800M lease on mobile generators that proved useless, and a Texas Attorney General fraud investigation — a reputational wound largely, but not fully, healed. A structural nuance often lost in the narrative: because CenterPoint owns only the wires in ERCOT, the 12.2 GW of committed load does not directly drive its capex (customers fund their own interconnections) — it drives demand charges and indirect, still-to-be-defined future transmission. The growth is real; the coupling between the load headline and the rate-base build is looser than it looks.

The analysis that follows takes no position and sets no price target. It concludes that CenterPoint is a high-quality regulated-growth utility whose premium is partly earned by a genuinely superior demand story and partly a rich, crowded defensive bid — with multiple compression, not fundamental deterioration, the dominant risk to a buyer at today’s price.

2. Business Overview

What CenterPoint is. CenterPoint Energy, Inc. (NYSE: CNP; CIK 0001130310; incorporated in Texas; headquartered in Houston; S&P 500) is a public-utility holding company that, following a multi-year transformation, is now essentially a pure-play, fully rate-regulated electric and natural-gas delivery business. It reports two segments — Electric and Natural Gas — plus a Corporate & Other bucket that houses parent-level interest expense and the residual ZENS/legacy items. There is no midstream, no merchant generation of consequence, and no material unregulated earnings: the 2021 exit from Enable Midstream (folded into Energy Transfer) and subsequent monetization of the Energy Transfer units completed the shift the company brands “Our Path to Premium.”

The Electric segment (~$4.87B FY2025 revenue; ~$705M segment net income; the growth engine). This is two very different businesses:

• CenterPoint Houston Electric — a transmission-and-distribution-only (wires) monopoly on the Texas Gulf Coast, operating inside ERCOT. It owns no generation and bears no commodity risk; it earns regulated delivery charges billed to the ~67 retail electric providers (REPs) that serve ~2.86 million metered customers (2.54M residential + 0.31M commercial/industrial). This is the crown jewel — the asset the entire investment thesis rests on.
• Indiana Electric (SIGECO) — a smaller (~154k customers), vertically integrated utility in southwestern Indiana that owns generation (coal transitioning to gas + renewables) and operates within MISO.

The Natural Gas segment (~$4.48B FY2025 revenue; ~$570M segment net income; the cash/ballast business). Local gas distribution utilities (LDCs) serving residential, commercial and industrial customers. The go-forward footprint is Texas + Minnesota + Indiana after two rounds of pruning: Louisiana & Mississippi gas were sold in 2025, and the Ohio gas LDC sale is expected to close Q4-2026. These are classic rate-base-and-replace safety/reliability businesses — slower-growing than the Houston wires, and management is steadily shrinking the gas mix in favor of Texas electric.

How it makes money. Like every regulated utility, CenterPoint earns an allowed return (~9.4–9.8% ROE) on a rate base — the depreciated capital it has prudently invested in poles, wires, substations, pipe and (in Indiana) generation — plus recovery of operating costs and, in most cases, commodity pass-through. Revenue is a function of invested capital × allowed return, not units sold, which is why rate-base growth (~11%) rather than volume is the primary earnings driver even though the volume story is what makes the rate base grow. Roughly 85% of its capital spending is recovered through interim “tracker” mechanisms (DCRF and TCOS in Houston Electric, GRIP in Texas Gas, TDSIC in Indiana) that update rates between full rate cases, sharply reducing regulatory lag.

Revenue quality. Essentially all recurring and regulated: monopoly delivery service to a captive customer base under multi-decade franchises, with fuel/commodity costs largely passed through. Demand is inelastic (electricity and heat are non-discretionary), and the customer base is growing in the core Houston territory. The earnings are among the most predictable in the equity market — the entire question is price, not durability.

Verdict. A clean, high-quality, ~100%-regulated delivery utility anchored by a best-in-class growth asset (Houston wires) with a slower gas business being deliberately shrunk. The business model is simple, durable and predictable; the transformation from the messy 2020 holding company to today’s focused pure-play is complete and creditable.

3. Industry Dynamics

Structure. Regulated electric and gas distribution is a legal-monopoly, cost-of-service industry. A utility is granted an exclusive franchise (in Texas, a PUCT Certificate of Convenience and Necessity plus municipal franchises of 30–40 year terms) to serve a defined territory; in exchange, its rates and returns are set by regulators to approximate a fair return on prudently invested capital. There is no direct competition for the wires: “there are no other electric T&D utilities in Houston Electric’s service area” (10-K). Barriers to entry are effectively absolute — a competitor cannot build a duplicate grid, and would not be allowed to.

Profit pool and its ceiling. The profit pool is administratively determined: rate base × allowed ROE, with the equity layer levered by regulator-approved debt (typically ~45–55% equity in the capital structure). This makes the industry’s returns stable but capped — allowed ROEs across CNP’s jurisdictions cluster ~9.4–9.8%, and consolidated ROIC (~5.4%) sits below a typical ~6–7% utility WACC, with the equity return above WACC only because of leverage. Utilities do not out-earn their cost of capital by much; they compound book value at the allowed return and grow the quantity of rate base. The value-creation lever is therefore volume of investable rate base, funded largely externally, not margin expansion.

The ERCOT / Texas context — a double-edged distinction. CenterPoint’s core sits in ERCOT, the isolated Texas grid, and in a deregulated retail market where CenterPoint is a pure wires company. This is favorable in that CNP bears no generation, fuel, or commodity risk and no stranded-asset exposure — it is the pure “toll road.” But ERCOT is also the epicenter of US power-demand growth (data centers, crypto, industrial electrification, LNG), which creates the single best volume tailwind in the sector. The nuance (discussed below): in ERCOT’s cost-allocation model, large-load customers pay for their own interconnection facilities, so surging load translates into demand-charge revenue and indirect future transmission need rather than a one-for-one rate-base increase.

Regulation — constructive but with a political ceiling. Texas (PUCT) has historically been a constructive regulator with strong interim-recovery mechanisms (trackers, securitization of storm costs). Indiana (IURC) and its TDSIC trackers are constructive; Minnesota (MPUC) is the least constructive of CNP’s set (longer lag, multi-year cases). But Beryl demonstrated the industry’s hard political ceiling: when reliability fails visibly, legislators and regulators will push back — the PUCT trimmed CNP’s requested resiliency plan from $5.75B to ~$2.7B, and forced the utility to eat the useless mobile-generator lease. Affordability is a binding real-world constraint; Houston’s charges are ~11% below the national average and the lowest in ERCOT, which is precisely the headroom CNP is spending against.

Marathon capital-cycle lens. In most industries, heavy capital inflow signals mean-reversion risk (returns attract capital, capital competes returns away). Regulated utilities invert that signal: heavy capex is the product, not a warning, because the regulator guarantees a return on it — provided the spend is prudent and the customer can afford it. The genuine risks are therefore not competitive but (a) regulatory (will the regulator keep funding the plan at a fair ROE?), (b) financial (can the utility raise the debt and equity to fund a chronically FCF-negative build without a credit downgrade?), and © political/affordability (will rising bills provoke a clawback?). CenterPoint is more exposed to all three than a slower-growing peer precisely because its plan is so large.

Verdict: a structurally attractive-but-capped industry, with CenterPoint holding the sector’s best volume tailwind and, correspondingly, above-average regulatory-execution and financing risk. The differentiated positive is real Houston growth (rare in the sector); the differentiated risk is Texas political/regulatory execution and perpetual external funding.

4. Competitive Position

The moat: a regulated-franchise monopoly. In Greenwald’s taxonomy, CenterPoint’s advantage is the strongest type — a government-granted local monopoly combining demand captivity (customers have no alternative wires provider) and economies of scale (a single grid serving millions is vastly cheaper per customer than any duplicate), codified and protected by regulation (CCN + franchises). No competitor can enter Houston Electric’s territory; the barrier to entry is legal, not merely economic. Market share is ~100% and structurally stable — the textbook “wide moat.”

But it is a wide, shallow moat. The same regulation that guarantees the monopoly caps the return at ~9.65% allowed ROE. A moat that would let an unregulated firm earn super-normal profits instead delivers a stable, mid-single-digit-to-~10% return on equity. The moat is real and its financial signature is unmistakable — remove the franchise and the business evaporates — but it does not translate into pricing power or excess returns. It translates into durability and low risk of loss, which is what utility investors buy.

Where CenterPoint’s competitive position is genuinely differentiated — the demand backdrop. Most utility moats are identical in kind (everyone has a monopoly); they differ in the quality of the territory. CenterPoint’s territory is its edge:

• Houston is the best large-load growth corridor in the US. Management’s Q1-2026 update points to 12.2 GW of firmly committed new industrial load (up from 7.5 GW one quarter earlier), spanning >12 customers and ~20 projects, with 3.2 GW already ERCOT-approved and ~8 GW targeted to energize by end-2028. Drivers are diversified — data centers/AI, advanced manufacturing (Gulf Coast builds the equipment inside data centers), energy/LNG exports, electrification, and ~2%/yr population growth. This diversity is a real differentiator versus peers whose growth rests on one or two hyperscaler contracts.
• The affordability flywheel. Because CNP has kept Houston delivery rates roughly flat since 2014 by spreading fixed costs over a growing base, incremental large load lowers per-customer bills (management estimates ~$4B of aggregate residential/commercial savings over ten years from utilizing existing capacity) — which in turn buys political headroom for more investment. This is a genuine, self-reinforcing competitive advantage of the territory.

The critical nuance that tempers the moat’s cash conversion. In ERCOT, CenterPoint provides only T&D, and large-load customers pay for the switchyard/substation modifications needed to interconnect (management: “I wouldn’t look at this as necessarily a direct impact to the CapEx plan”). So the 12.2 GW does not mechanically inflate rate base. What it does is: (1) generate incremental demand charges — management sizes this at ~$6M/month per 1 GW of industrial load, a direct earnings and affordability tailwind; and (2) create indirect future transmission need (replacement capacity, intra-regional lines, and the 765-kV import buildout coming ~2031–2032) that will enter the capital plan later, pending a transmission study due 2H-2026. The load story is real and powerful, but the rate-base translation is indirect and partly still to be defined — a subtlety the headline “12.2 GW committed” obscures.

Head-to-head. Versus the closest comp, Entergy (Gulf South, contracted hyperscaler load, vertically integrated so load does drive its generation + wires capex), CenterPoint offers a more diversified load base and a cleaner wires-only risk profile, but weaker direct capex capture per gigawatt of load and a weaker balance sheet. Versus Atmos (gold-standard gas LDC, A-rated, 11.45% pipeline ROE), CNP has faster potential growth but materially lower credit quality and a more complex, higher-execution-risk plan. Versus slower peers (ED, PPL, AEP), CNP has a clearly superior growth runway.

Verdict: a durable, wide-but-shallow regulated monopoly whose genuine differentiation is the quality and diversity of its Houston growth territory — the best in the sector — partly offset by an indirect load-to-rate-base translation and a below-peer balance sheet. The moat protects against loss; the territory drives the upside.

5. Growth History and Forward Opportunities

History (2021–2025). Revenue grew from $8.35B (2021) to $9.36B (2025), a modest ~2.9% CAGR distorted by divestitures and commodity pass-through; the more meaningful metric, non-GAAP EPS, compounded at a steady ~8%/yr, and CNP has consistently delivered at the mid-to-high end of its guided range. GAAP EPS moved from ~$1.13 (continuing ops, 2021) to ~$1.61 (2025). Crucially, the growth was not revenue-driven — it was rate-base-driven: capex ran $4.4B (2023) → $4.6B (2024) → $5.4B (2025), and rate base compounded at a double-digit rate, with per-share earnings growing ~8% after dilution.

The forward algorithm. Management’s model is explicit and, unusually, extended to a fifteen-year horizon: ~11%+ rate-base growth → 7–9% non-GAAP EPS growth, guided at the mid-to-high end through 2028 and 7–9% annually through 2035. FY2026 non-GAAP EPS is guided to $1.89–$1.91 (~8% over 2025). The mechanics: rate-base growth (~11%) minus equity dilution (~1.5–2 pts/yr) minus rising interest expense ≈ ~8% EPS. CNP “rebases” guidance off each year’s actual result, so the compounding is off a moving, delivered base — a discipline that has built credibility.

Where the growth comes from:

1. Houston Electric load + resiliency. The $65.5B ten-year plan is dominated by Houston electric transmission and distribution, including the post-Beryl System Resiliency Plan (~$2.7B 2026–2028) and the transmission needed to move and import power for the load boom. This is the core.
2. Indiana upside. Management is in advanced talks for a large-load customer in southwest Indiana — potentially a ~$1B incremental capex opportunity (simple-cycle → combined-cycle conversion + transmission unlocking ~1.5 GW), targeted within 2027–2029, with more possible.
3. Plan increases. Management explicitly frames the $65.5B as a base, with >$10B of identified incremental opportunity plus a 2H-2026 transmission study likely to add projects — i.e., the growth guidance has a built-in upward bias if load materializes.
4. Cash-tax tailwind. A favorable corporate-AMT change removes ~$150M/yr of cash taxes (plus refunds), which management says can fund ~$1B of incremental capex with no incremental equity — a rare non-dilutive growth lever.

Quality of the growth — mostly high, with two caveats. The demand backdrop is genuinely organic and diversified (not financial engineering, not a single fragile contract), which is high-quality. Caveat 1: much of the “12.2 GW committed” and the “doubling by mid-2030s” is CenterPoint’s own forecast — self-serving (bigger load justifies bigger rate base and capex), and a chunk of the pipeline is queue positions / MOUs rather than energized load. The 3.2 GW ERCOT-approved is the firm number; the rest is conviction. Caveat 2: per-share growth is diluted-down rate-base growth — the ~8% EPS is what’s left after ~2 points of annual share issuance, and it depends on continuous capital-markets access on acceptable terms.

Verdict: high-quality, genuinely organic growth with the sector’s best runway — tempered by a self-forecasted load pipeline that is only partly firm and a per-share result structurally throttled below rate-base growth by chronic dilution.

6. Financial Quality

Income statement. FY2025: revenue $9,357M; operating income $2,110M (22.5% operating margin); EBITDA $3,640M (38.9% margin); net income $1,052M (11.2% net margin); GAAP diluted EPS $1.60. Interest expense of $903M (up from $838M in 2024 and $701M in 2023) is the fastest-rising cost line — the direct consequence of a debt-funded rate-base build in a higher-rate world, and a structural headwind to EPS. The effective tax rate is low (~15.6%) owing to production tax credits and amortization of excess deferred taxes — normal for a utility. Electric (Houston) is the earnings engine (~$705M segment net income and rising), Natural Gas the ballast (~$570M), and Corporate & Other a persistent ~$220M parent-interest drag.

Earnings quality — clean. The GAAP-to-non-GAAP gap (GAAP $1.60 vs non-GAAP ~$1.75 for 2025) is driven mainly by mark-to-market noise on the ZENS (Zero-Premium Exchangeable Subordinated Notes due 2029, indexed to legacy AT&T/Charter shares) and divested-operations items — genuine non-operating noise, not aggressive add-backs. Cash from operations ($2.49B) exceeds net income ($1.05B) by roughly the depreciation charge (~$1.5B) — a clean relationship with no accrual red flags. Large regulatory-asset/liability balances are normal for a rate-regulated utility; earnings are regulator-set, not market-set. Net income tracks cash reasonably; there is no quality-of-earnings problem here (a welcome contrast to the “FCF that’s really float” traps elsewhere).

Free cash flow — structurally negative, by design. CFO of ~$2.49B against capex of ~$5.4B leaves the business deeply FCF-negative before the ~$0.57B dividend. This is not a warning sign — it is the defining feature of a ~11% rate-base grower: the capital build vastly exceeds internal cash generation, and the gap is funded with new debt and equity. (Beware third-party data feeds — including ROIC.ai — that label CFO as “free cash flow”; CNP’s true equity FCF is negative and will remain so throughout the plan.) The corollary is perpetual capital-markets dependence: the thesis requires continuous access to debt and equity on acceptable terms, which ties the equity story directly to the balance sheet and the rate environment.

Balance sheet — stretched, at the aggressive end of “normal utility.” FY2025 shareholders’ equity was $11,153M (book value ~$17.00/share, matching the AZI figure and confirming ROIC.ai’s P/B ~12× and “ROE” of 58% are data glitches — ignore them). Consolidated debt is ~$23B (ST $2,414M + LT $20,566M); total debt/cap ~92%, total debt/EBITDA ~6.3×, EBITDA/interest ~4.0×. Cash is a token $38M — the company runs on revolvers and commercial paper. Goodwill is ~$3.5–4.9B (Vectren legacy). Floating-rate debt is only ~$1.5B (~6.5% of the total) — well-managed. The critical figures: the holding company is rated Baa2 (negative outlook) / BBB / BBB (Moody’s/S&P/Fitch), with Houston Electric higher at A2/A/A; FFO/debt was ~12.5% at Q1-2026 against a Moody’s downgrade threshold of ~13–14%. Management targets the high end of a 150bp cushion by year-end, aided by the cash-tax refund and ~70% of 2026 financing already completed. This is the binding constraint on the whole story: a low-BBB, negative-outlook holdco executing the largest capital plan in its history has little room for error before a downgrade forces either slower spending or more dilution.

Returns. Computed ROE ~9.6% (NI-to-common ~$1,048M / average common equity ~$10,910M), essentially at the allowed return — evidence of good regulatory execution and minimal lag (better than Entergy, which under-earns its allowed ROE). ROIC ~5.4% and ROA ~2.3% are structurally sub-WACC, as for all regulated utilities; the equity return clears the cost of equity only via leverage. The economics do not “improve with scale” in a margin sense — allowed ROE is fixed by regulators — but they compound: more rate base at a steady ~9.6% return, plus ~85% of it recently re-cased (low lag), is a reliable earnings-growth machine so long as the capital keeps flowing.

Verdict: high earnings quality and reliable at-allowed returns, offset by a genuinely stretched, negative-outlook balance sheet and structurally negative free cash flow — the financial profile of a fast-growing regulated utility, with the credit metrics, not the operations, as the point of vulnerability.

7. Capital Allocation

The framework. CenterPoint is a textbook regulated-utility capital allocator: the overwhelming use of capital is rate-base investment ($5.4B in 2025; $6.8B budgeted for 2026; $65.5B over 2026–2035), funded by internally generated cash (~$2.5B), new long-term debt, and equity issuance, with the residual returned as a dividend. There are essentially no buybacks (a growth utility issues, it does not repurchase) and no discretionary M&A of consequence — capital allocation is the capex plan and the financing of it.

Portfolio reshaping — coherent and value-accretive. Over five years management has executed a disciplined simplification toward the highest-growth, fully-regulated core:

• Enable Midstream exit (2021): CNP’s 53.7% stake was contributed to Energy Transfer for ~201M ET units, subsequently monetized — completing the exit from commodity-exposed midstream that had caused the 2020 distribution cut and near-death.
• Louisiana & Mississippi gas (2025): sold to Bernhard Capital Partners (Delta Utilities) for ~$1.2B, ~32× earnings — a rich multiple for slower-growth LDCs, redeployed into higher-growth Texas electric. (Note: this was not an LS Power deal, contrary to some secondary reporting.)
• Ohio gas (closing Q4-2026): further gas pruning.
• Energy Systems Group (2023): the non-core energy-services subsidiary sold to an Oaktree affiliate for ~$157M.
• Indiana coal transition: moving off coal (with a 2025 partial backtrack keeping Culley Unit 3 online past 2027, partly under a DOE emergency order) toward gas + renewables.

The through-line — sell slower-growth/non-core assets at rich multiples, redeploy into the Houston electric growth engine — is sound, disciplined capital allocation and is the right response to the opportunity set.

Financing & dilution — the standing overhang. Funding a ~11% rate-base grower requires perpetual external capital. Share count has risen from ~593M (2021) to ~656M (2025), ~+11%, via ATM and forward-sale equity programs (a new ~$1.0B ATM/forward program was established in May-2026). This dilution is the unavoidable cost of the growth — it subtracts ~1.5–2 points/yr from rate-base growth to arrive at the ~8% EPS algorithm — but it is also a standing risk: if the credit metrics tighten or the equity de-rates, the required issuance becomes more expensive and more dilutive. The cash-tax (AMT) tailwind that funds ~$1B of capex without equity is a genuine, if modest, offset.

Dividend. The 2020 cut (from $0.29 to $0.15/quarter, ~−48%) — forced by the Enable distribution collapse — is the scar that still shapes policy. Since then the dividend has been rebuilt steadily and conservatively, raised to $0.23/quarter (~$0.92 annualized) in April-2026, at a ~52–55% payout of non-GAAP EPS, and grown roughly in line with earnings (~6–8%). The sub-peer yield (2.06%, the lowest in the cohort) is a valuation artifact — the payout ratio is normal; the price is high — not a stinginess signal.

Management & incentives. CEO Jason Wells (in seat since Jan-2024; also Chair) leads; predecessor Dave Lesar drove the 2020 turnaround. Incentive design is pure rate-base-growth alignment: short-term incentives are heavily weighted to non-GAAP Adjusted EPS (with safety/operational/customer modifiers; paid out at 159% of target for 2025), and long-term PSUs split ~35% relative TSR / ~35% cumulative Adjusted EPS / ~30% carbon-reduction. This correctly aligns management with the grow-the-rate-base-per-share flywheel, though it also incentivizes the aggressive capex/issuance model and can pay out well (Wells’s 2025 comp of ~$12.1M nearly doubled, drawing local criticism given Beryl) even in a reputationally difficult year — a governance-optics negative, not a substantive misalignment. Insider activity shows routine grant/vesting patterns (a May-2026 Form 4 cluster consistent with the annual equity cycle), with no evident discretionary open-market purchases — no strong insider conviction signal either way.

Verdict: intelligent, disciplined capital allocation — a coherent multi-year simplification into the highest-growth regulated core, financed prudently within the constraints of a stretched balance sheet. The principal critique is not the quality of the allocation but its dependence on perpetual, dilutive external funding and a credit profile with little slack.

8. Changes and Headwinds — Last Two Years

1. Hurricane Beryl (July 8, 2024) — the defining recent event. A Category-1 storm knocked out power to ~2.2M of CenterPoint’s ~2.8M Houston customers, with multi-day-to-multi-week restoration in extreme summer heat and deaths attributed to the prolonged outage. It triggered a political firestorm (Governor, Lieutenant Governor, legislature), a PUCT investigation (report due December 2024), a Texas Attorney General fraud/waste probe, and a third-party after-action review with 77 recommendations. The most damaging detail: CenterPoint had leased ~$800M of oversized 32MW+ mobile generators (post-Winter-Storm-Uri) that were useless for a distribution outage — the CEO conceded they were meant for a winter load-shed scenario — yet the PUCT had allowed CNP to recover the lease and earn a return on it. Management made ~42 remediation commitments (later reported completed ahead of schedule), voluntarily agreed to stop charging Houston customers for the large generators (relocating them to San Antonio), and issued public apologies; no C-suite firings followed. Assessment: Beryl damaged CenterPoint’s regulatory and political standing badly, but — paradoxically — accelerated the growth thesis by forcing a large, state-blessed grid-hardening program. The wound is largely healed but the scar (heightened scrutiny, an affordability ceiling) is permanent.

2. The Greater Houston Resiliency Initiative (GHRI) and System Resiliency Plan (SRP). In Beryl’s wake, CNP launched GHRI (Phases 1 & 2: >10,600 storm-resilient poles, >3,400 miles of vegetation cleared, ~370 self-healing automation devices, ~200 miles undergrounded) and filed a 2026–2028 SRP at $5.75B. The PUCT, after a $3.2B settlement with Houston-area cities, approved a trimmed ~$2.7B (August 2025). This is a double-edged development: it added multi-billion-dollar rate-based investment (bullish for the plan) but the ~50%+ haircut from the original ask demonstrated the regulator’s willingness to push back.

3. The record capital plan. Management raised the plan repeatedly through 2025 — ~$48.5B → ~$53B (through 2030) — culminating in the record $65.5B ten-year plan (2026–2035) unveiled with Q4-2025 results, plus >$10B of identified incremental opportunity. This is the single biggest positive change to the thesis.

4. Load-growth acceleration. The “firmly committed” Houston industrial load was raised from 7.5 GW to 12.2 GW in Q1-2026, with the ~50% peak-demand increase pulled forward ~2 years — the driver behind the plan and the multiple.

5. Portfolio simplification (Louisiana/Mississippi gas sold 2025; Ohio gas exiting 2026; see the Capital Allocation section).

6. Financing de-risking. ~70% of 2026 financing completed early; a $650M convertible (Feb-2026) reduced floating-rate exposure; the corporate-AMT change removed ~$150M/yr of cash taxes.

Headwinds: the Baa2 negative Moody’s outlook and ~12.5% FFO/debt (below the ~13–14% threshold); a rising interest bill; perpetual dilution; the lingering Beryl affordability/political ceiling; and the risk that the self-forecasted load pipeline converts more slowly than management projects.

Verdict: on balance the last two years strengthened the investment thesis — the load boom and the record capital plan are transformational positives that dwarf the Beryl damage, which the market has already largely forgiven (the stock is at an all-time high). But they also raised the stakes: a bigger plan on a tighter balance sheet with less political headroom.

9. Risk Analysis

The dominant risk is unambiguous: at an all-time-high, richest-ever multiple, valuation (Risks 1–3) is the primary exposure — a de-rate, a credit event, or a rate-driven bond-proxy compression — not a break in the underlying business, which is durable and growing. Catastrophic loss risk is low (regulated monopoly, essential service); the realistic downside is a multi-year period of flat-to-negative total return as the multiple normalizes even while earnings rise.

10. Valuation Discussion (Embedded Expectations)

Where it trades. At $44.61 (2026-07-02, an all-time high): market cap ~$29.3B; net debt ~$22.9B; EV ~$52B. On non-GAAP EPS: ~23.5× current-year ($1.90 FY2026E) and ~21.5× FY27E (~$2.05–2.10); on GAAP TTM, ~27.7×. EV/EBITDA ~13.0× TTM / ~14× forward; P/B 2.6×; dividend yield ~2.06%. On the stock’s own ten-year history (AZI percentiles): P/E 94th, P/B 91st, P/S 99.9th, composite 95th — the richest it has ever been on nearly every metric, and EV/EBITDA at a decade high.

Peer comparison (as of early July 2026; forward P/E = NTM consensus, yfinance).

The read. CenterPoint carries the highest forward multiple in the group after Entergy, and the lowest dividend yield of all — while its ~7–9% growth is matched or beaten by ETR, SRE, and PPL at cheaper multiples and, in AEP/ATO/PPL/SRE’s cases, with stronger holdco credit. The premium rests almost entirely on the Houston load narrative and a defensive-momentum bid, not on a demonstrably superior return or credit profile. Where PPL offers comparable ~10% rate-base growth at ~17× and a 3.1% yield, CNP asks ~21–23× and pays 2.06% — you are paying a full turn or two and giving up a point of yield for the quality of the growth story, which is real but not free.

Embedded-expectations / reverse read. Decompose the ~10%/yr expected total return: ~8% EPS growth + ~2% yield — if the multiple holds. The multiple is the entire swing factor. A Gordon-style justified P/E (payout ~52% / (cost of equity ~7.75% − growth ~6.5%)) is hyper-sensitive and prices ~40× — i.e., the current ~21–23× is not demanding on a pure DDM if you believe ~6.5% perpetual growth at a ~7.75% cost of equity. But that is precisely the fragility: the valuation is underwriting durable ~8% growth and no de-rating and a low bond-proxy discount rate, simultaneously. Change any one — a 50bp higher discount rate, a slip to 6–7% growth, or a routine reversion toward the peer-average ~18× — and the math turns. A move from ~21.5× to ~18× FY27E over, say, three years is ~−5–6%/yr of multiple drag, roughly cancelling the EPS growth and producing a flat-to-negative total return even as earnings rise.

Scenarios (illustrative, FY27E non-GAAP EPS ~$2.05–2.10):

• Bear (~$35–38): multiple reverts toward the peer-average ~17–18× as rates stay higher-for-longer and/or a credit downgrade forces slower spend/more dilution; load converts slower than forecast. EPS still grows, but the de-rate dominates → 3-year flat-to-negative total return.
• Base (~$42–47): ~8% EPS growth delivered, multiple compresses modestly toward ~19–20× (a rich but defensible premium for the best growth in the group); ~2% yield → ~7–9%/yr total return, most of it earnings not multiple.
• Bull (~$52–58): load materializes faster, the plan is raised beyond $65.5B, EPS growth runs at the high end (9%+), and the ATH multiple (~22×+) holds on continued AI-power-demand enthusiasm and rate cuts → low-teens total return.

No price target. No buy/sell. The embedded expectation is clear: CenterPoint is priced for its plan to work in full, with the richest multiple in its history doing a great deal of the work. The business quality supports a premium; the size of today’s premium leaves little margin for the many things (rates, credit, regulation, load timing) that could go modestly wrong.

11. Variant Perception

Consensus view. CenterPoint is a premium regulated-growth utility with the best organic load story in the sector (Houston data centers + AI + industrial), a record $65.5B plan, a long 7–9% EPS runway to 2035, and constructive regulation — a “must-own” way to play US electrification and AI power demand with utility-grade risk. The Street rewards it with a top-of-sector multiple and the lowest yield in the group.

The strongest bull case. The Houston load boom is real, diversified and early — 12.2 GW committed is a floor, not a ceiling, with a transmission study due 2H-2026 likely to add projects, >$10B of identified incremental capex, and an Indiana large-load option on top. The affordability flywheel (load lowers per-customer bills → political headroom → more investment) is self-reinforcing and rare. Management has consistently delivered mid-to-high-end EPS growth and rebased off actuals. If the plan is raised toward ~$75B+ and EPS compounds at 9%, the multiple can hold and the stock compounds at low-double-digits — the ATO/SRE outcome of a great franchise staying expensive because it keeps delivering.

The strongest bear case. You are paying the richest multiple in the company’s history (99.9th-pctile P/S) and the lowest yield in the group for a Baa2/negative-outlook holdco at 92% debt-to-cap, running the largest capex plan in its history on FFO/debt that’s below the downgrade threshold — two years after its flagship grid failed catastrophically in a hurricane. The 12.2 GW is mostly the company’s own forecast (only 3.2 GW ERCOT-approved), and critically the load doesn’t directly drive rate base in ERCOT. Any of a dozen ordinary developments — a rate backup, a Moody’s downgrade, a slower load conversion, a Texas ROE trim, a second storm — compresses a multiple that is priced for none of them, and the de-rate alone can produce years of flat-to-negative return while EPS quietly rises. This is a bond-proxy at full extension dressed as a growth stock.

The 3–5 assumptions that matter most:

1. Multiple durability — does ~21–23× / 2% yield persist, or revert toward the ~18× peer band? (The single biggest swing factor.)
2. Load conversion — does the 12.2 GW “committed” (mostly forecast) actually energize on schedule, and does it translate into rate base via the coming transmission need?
3. Credit — does CNP hold Baa2 and keep FFO/debt above the threshold while funding $65.5B, or does it downgrade / dilute more?
4. Rates — does the 10-year yield fall (supporting the bond-proxy bid) or stay higher-for-longer (compressing it and raising interest expense)?
5. Texas regulation/politics — does the post-Beryl constructive relationship hold, or does an affordability backlash claw back returns?

Factor-positioning read (what the tape is pricing). The FactorsToday model reads CenterPoint as a textbook defensive: Utilities-sector loading +0.90, positive LowVolatility (+0.28), negative Growth (−0.42) and Quality (−0.29), and slightly negative Value (−0.03, i.e. not cheap) — with no momentum loading despite the strong tape, and the stock sitting within ~1% of its 12-month relative-strength peak (rs_12m +24.9%), beta 0.18, 1-year Sharpe 1.54, max drawdown just ~7%. This is a crowded, low-volatility, momentum-adjacent defensive bid at its relative-strength peak — the profile of a name the market has bid to perfection on the rate-cut + AI-power theme, not a value name with room to re-rate up. It is evidence that consensus is fully on this side of the boat, which is exactly where the multiple-compression risk (Risk 1) lives — not a price call, but a caution that the easy money in the re-rating has already been made.

Verdict: The variant perception is not that the business is bad — it plainly is not — but that the market is paying a growth-stock price and accepting a bond-proxy yield for a utility whose best-in-class growth is real but partly self-forecasted, whose balance sheet is tighter than the narrative admits, and whose multiple has no cushion.* The edge, if there is one, is in waiting for the price, not in the story.

12. Fact vs. Interpretation Table

13. Open Questions

1. How firm is the 12.2 GW? What share is contracted/energized versus queue positions and MOUs? Only 3.2 GW is ERCOT-approved — the conversion rate and timing are the key bull-case validation and are externally unverified.
2. What does the 2H-2026 transmission study add to the plan? The load-to-rate-base translation runs through this study; the incremental capex it defines is the mechanism by which the load story becomes an earnings story.
3. Does CNP hold Baa2 / keep FFO/debt above threshold through the $65.5B build, or is a downgrade / larger equity slug coming? What is the precise Moody’s/S&P downgrade trigger and current cushion?
4. Rate base by jurisdiction is not disclosed in the 10-K segment note (only total assets by segment) — the Texas vs Indiana vs Minnesota split, and the earned-vs-allowed ROE by jurisdiction, would sharpen the regulatory-lag read.
5. Any insider open-market buying? The Form 4 record shows routine grants/vesting; a discretionary code-P purchase into any weakness would be a genuine conviction signal — none evident so far.
6. Second-storm exposure — how much has GHRI/SRP hardening actually reduced the probability and severity of another Beryl-scale outage, and what is the securitization capacity for the next event?

14. What Must Be True

For the bull case to win (own it here and be rewarded):

• Houston load energizes on schedule (the 12.2 GW converts, the transmission study adds rate base), and the plan is raised toward ~$75B+.
• EPS compounds at the high end (9%+) of the range through the late 2020s, and CNP holds Baa2 without a dilution shock.
• The multiple holds near ~21–23× — i.e., the AI-power/rate-cut defensive bid persists and rates drift lower.
• Falsification test: if, over the next 4–6 quarters, ERCOT-approved load fails to climb well beyond 3.2 GW, the transmission study adds little to rate base, or Moody’s downgrades / the stock de-rates below ~18× FY27E while EPS still grows — the bull thesis (that the premium is earned and durable) is broken.

For the bear case to win (wait for a lower price):

• The multiple reverts toward the ~17–18× peer band (rates higher-for-longer, or a credit event, or simple mean-reversion from a 99.9th-pctile starting point), producing a flat-to-negative multi-year total return even as EPS rises.
• A Texas regulatory/political setback (ROE trim, resiliency clawback, affordability backlash) or a second major storm re-opens the Beryl wound.
• Falsification test: if CNP delivers 8–9% EPS growth and the multiple holds at ~21×+ and the yield stays sub-2.3% for the next two years (i.e., the premium proves durable), the bear thesis (that the valuation is a trap) is broken — the franchise will have earned its price.

The synthesis: the business will almost certainly keep growing earnings ~8%; the entire debate is whether today’s all-time-high, richest-ever multiple is a fair price for that growth or a ceiling. What must be true for a buyer at $44.61 is that the multiple holds — an assumption the history of 95th-percentile valuations does not flatter.

15. Source Appendix

See the accompanying Source Appendix (CNP_source_appendix.md) for the full, categorized list of primary and secondary sources with URLs and access dates. Principal sources relied on:

• CenterPoint Energy FY2025 Form 10-K (filed 2026-02-19; cnp-20251231.htm) — segments, customers, franchises, generation, regulatory matters, ratings, debt, ZENS, forward-looking statements.
• CenterPoint Q1-2026 earnings call transcript & release (2026-04-22/23) — 12.2 GW committed load, $65.5B plan, FY2026 guidance $1.89–$1.91, 7–9% growth to 2035, FFO/debt, financing.
• CenterPoint Q4-2025 results / investor day materials (Feb 2026; Oct-2025 investor update) — $65.5B ten-year plan, ~11% rate-base growth, load forecast.
• PUCT / Texas regulatory filings & press — 2024 Houston Electric rate case settlement (9.65% ROE); System Resiliency Plan approval (~$2.7B, Aug-2025); Beryl investigation.
• M&A / portfolio — Enable/Energy Transfer (2021); Louisiana & Mississippi gas sale to Bernhard Capital/Delta Utilities ($1.2B, closed 2025); Ohio gas sale; Energy Systems Group sale.
• Quantitative data — AZI price CSV & valuation percentiles; ROIC.ai fundamentals/EV/ratios (reconciled to 10-K); FactorsToday factor model; yfinance peer snapshot.
• DEF 14A (2026-03-04) — compensation and incentive design; insider Form 4 activity.

Facts are cited to primary sources; interpretations are labeled as such. Management commentary is treated as hypothesis and validated against filings, financials, and external evidence.

APPENDIX A — Standard Diligence Questionnaire

CenterPoint Energy, Inc. (NYSE: CNP) — as of 2026-07-03

Supplemental diligence questionnaire. Fact / Interpretation / Assumption labeled where it matters. Where a question does not map to a regulated utility, the correct sector analog is given.

General

What thoughtful questions have other investors asked about this company?

• Is the 12.2 GW of “firmly committed” Houston load real, and does it actually drive rate base? (The critical question — only 3.2 GW is ERCOT-approved, and in ERCOT the load drives demand charges + indirect transmission, not direct capex.) (Interpretation)
• Can a Baa2/negative-outlook holdco at 92% debt-to-cap fund a record $65.5B plan without a downgrade or a dilution shock? (The balance sheet is the binding constraint.) (Fact-based)
• Is the post-Beryl Texas regulatory relationship durably repaired, or one storm away from another clawback? (Interpretation)
• Is the richest-ever multiple (95th-pctile composite, lowest yield in the group) a fair price for the best growth in the sector, or a ceiling? (Interpretation)

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Neither in the traditional sense — regulated-utility earnings are administratively set (rate base × allowed ROE), not cyclical. They are on a secular uptrend driven by rate-base growth, not a cyclical peak. (Fact/Interpretation) The one cyclical element is interest expense (rising with rates), which is a headwind.

Driven by external environment or internal actions? Predominantly internal/structural — the capex-and-rate-base flywheel — amplified by an external tailwind (Houston load growth from AI/data centers/industrial) that is genuinely exogenous and favorable. (Interpretation)

How stable are revenues? Extremely — monopoly delivery service to a captive, growing customer base under multi-decade franchises, with commodity costs largely passed through. FY revenue $8.35B (2021) → $9.36B (2025). (Fact)

Outlook for products/services? Structurally growing: electricity delivery in the fastest-growing US large-load corridor; management guides ~11% rate-base and 7–9% EPS growth through 2035. (Fact, as guided)

How big will this market be — growing, shrinking, domestic/international? 100% domestic (Texas Gulf Coast core + Indiana/Minnesota). The Houston electric market is growing fast (peak demand forecast +~50% by ~2029, nearly doubling by mid-2030s); the gas business is flat-to-shrinking and being deliberately pruned. (Fact/management forecast)

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Not applicable in the usual sense — it is a legal monopoly with absolute barriers to entry. Competitive intensity is ~zero for the wires; the “competition” is with the regulator over allowed returns and with the political system over affordability. (Fact)

How profitable is the business (ROIC, ROE)? ROE ~9.6% (at the allowed return); ROIC ~5.4% (structurally below WACC, as for all regulated utilities — the equity return clears the cost of equity only via leverage). ROA ~2.3%. (Fact, computed)

How profitable is the industry — competitors, barriers to entry? Barriers are absolute (franchise/CCN monopoly); returns are stable but capped by regulators at ~9.4–9.8% allowed ROE. A wide-but-shallow moat: durable, not super-profitable. (Fact/Interpretation)

Can the business be easily understood? Yes — rate base × allowed return, plus a load-growth story. Among the simplest business models in the market. (Fact)

Undermined by foreign low-cost labor? No — a physical, local, regulated grid; not tradable. (Fact)

Do brands matter? No — customers cannot choose their wires provider. Reputation matters politically (Beryl showed a damaged reputation invites regulatory clawback), but there is no consumer brand premium. (Interpretation)

Nature of competition? With the regulator (rate cases) and the political system (affordability), not with rival firms. (Interpretation)

Customers’ switching costs? Infinite — there is no alternative T&D provider. (Fact)

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The core asset — the Houston franchise/monopoly and its growth optionality — is not a balance-sheet item; value accrues through the regulated rate base. Large regulatory assets (recoverable storm/deferred costs) are on-balance-sheet. (Fact/Interpretation)

Off-balance-sheet liabilities? The ZENS (indexed exchangeable notes) create GAAP mark-to-market volatility; pension obligations (~$0.5B) are modest. Operating leases and purchase obligations are disclosed. No unusual off-balance-sheet exposure flagged. (Fact)

How conservative is the accounting? Reasonably conservative and regulator-governed; the GAAP-vs-non-GAAP gap is genuine non-operating noise (ZENS, divestitures), not aggressive add-backs. CFO > net income by ~D&A — clean. (Interpretation)

How CapEx-hungry is the business? Extremely — ~$5.4B capex (2025), $6.8B budgeted (2026), $65.5B over ten years, against ~$2.5B CFO and ~$1.05B net income. Chronically FCF-negative by design; the gap is funded with debt and equity. (Fact)

Capital Allocation & Management

How much FCF does the business generate, and how is it used? Equity FCF is negative — capex vastly exceeds CFO. The correct utility framing: CFO (~$2.5B) funds a fraction of the capex; the rest is external capital; the dividend (~$0.6B) is paid on top and also externally funded at the margin. Capital allocation is the capex plan. (Fact)

Philosophy? Grow rate base at the allowed return; fund via internal cash + debt + equity within credit constraints; return ~52–55% of non-GAAP EPS as a growing dividend; simplify the portfolio toward the highest-growth regulated core. (Fact/Interpretation)

Significant acquisitions recently? No acquisitions — the direction is divestiture (Louisiana/Mississippi gas $1.2B to Bernhard Capital/Delta Utilities, 2025; Ohio gas exiting 2026; Energy Systems Group 2023; Enable midstream 2021). Sold slower-growth assets at rich multiples (LDC ~32× earnings), redeployed into Texas electric. (Fact)

Buying back shares? No — a growth utility issues shares. Count rose 593M → 656M (2021–2025), ~+11%, via ATM/forward programs (new ~$1.0B ATM May-2026). (Fact)

Issuing large amounts of new shares to insiders? No unusual insider issuance; standard equity-comp grants (PSUs/RSUs). Broad equity issuance is to the market to fund capex, not to insiders. (Fact)

Compensation policy of directors/management? CEO Wells 2025 total comp ~$12.1M (nearly doubled, drew Beryl-related criticism). STI heavily weighted to non-GAAP Adjusted EPS (paid 159% for 2025); LTI/PSU ~35% relative TSR / ~35% cumulative Adjusted EPS / ~30% carbon. Pure rate-base-growth alignment. (Fact)

Motivations of management? Aligned to the grow-the-rate-base-per-share flywheel and relative TSR. This also incentivizes the aggressive capex/issuance model and can pay out well even in a reputationally poor year (a governance-optics negative). (Interpretation)

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — a standard US C-corp common stock (NYSE: CNP), 1099 dividends. (Fact)

Dividend policy? ~$0.92/yr annualized ($0.23/quarter, raised April-2026), ~52–55% payout of non-GAAP EPS, grown ~6–8%/yr in line with earnings. The 2020 cut (−48%, Enable/Vectren crisis) is the scar that shapes the conservative policy. Yield ~2.06% — the lowest in the large-cap peer group, a valuation artifact. (Fact)

How profitable is the business? ROE ~9.6% (at allowed); net margin ~11%; EBITDA margin ~39%. Stable, capped, mid-single-digit-to-~10% equity returns. (Fact)

Is net income diverging from cash from operations? No problematic divergence — CFO ($2.49B) exceeds net income ($1.05B) by roughly D&A. Clean. (The free-cash-flow shortfall is a capex phenomenon, not an earnings-quality issue.) (Fact)

Risks & Downside

What factors would cause the stock to decline? (1) Multiple compression from a 95th-percentile, all-time-high starting point (the dominant risk); (2) a credit downgrade / FFO-to-debt breach forcing slower spend or more dilution; (3) higher-for-longer rates compressing the bond-proxy complex and raising interest expense; (4) a Texas regulatory/political setback (ROE cut, resiliency clawback); (5) a second major storm; (6) slower-than-forecast load conversion. (Interpretation)

Risk of a catastrophic loss? Low. A regulated monopoly providing essential service does not go to zero absent extraordinary mismanagement or a catastrophic uninsured event. The realistic downside is a multi-year flat-to-negative total return as the multiple normalizes, not a permanent capital impairment. (Interpretation)

Chance of a total loss? Negligible under any plausible scenario. (Interpretation)

Recent News & Events

Has the business environment changed recently? Yes, materially and mostly favorably: the Houston load boom accelerated (committed load 7.5 → 12.2 GW in one quarter), the capital plan was raised to a record $65.5B, and financing was de-risked (~70% of 2026 done; AMT cash-tax relief). The one adverse structural change is a tighter credit backdrop (Baa2 negative, FFO/debt ~12.5%). Hurricane Beryl (July-2024) remains the defining recent shock — damaging but ultimately thesis-accelerating. (Fact/Interpretation)

Significant acquisitions? None (divestiture-oriented; see above). (Fact)

Change in accounting policies? None material flagged. (Fact)

Recent changes — new markets, facilities, management? CEO transition (Wells since Jan-2024); exit from Louisiana/Mississippi/Ohio gas; post-Beryl GHRI grid-hardening; the record capital plan and 2035 growth-guidance extension. (Fact)

APPENDIX B — Source Appendix

CenterPoint Energy, Inc. (NYSE: CNP) — Research Sources (as of 2026-07-03)

Primary sources prioritized over secondary. Facts cited to primary sources; management commentary treated as hypothesis and validated against filings, financials, and external evidence. Access date 2026-07-03 unless noted.

1. Company SEC Filings (primary)

2. Earnings Call Transcripts (primary/near-primary)

3. Regulatory & Industry Sources (secondary/primary)

4. M&A / Portfolio (secondary/primary)

5. Quantitative Data Sources (third-party; reconciled to filings)

6. Peer Comparison (public data)

Note on data reconciliation: ROIC.ai’s per-share book value and return-on-common-equity figures for CNP are internally inconsistent (implying an erroneous equity denominator) and were discarded in favor of the FY2025 10-K balance-sheet equity of $11,153M (book value ~$17.00/share). All material figures reconcile to the FY2025 10-K or the Q1-2026 earnings materials.