ONEOK, Inc. (NYSE: OKE) — A Real NGL Franchise That Bought Its Growth, Now Priced for the Synergies to Land

An independent fundamental-research note. Report date: 2026-06-20. Price reference: $85.03 (close 2026-06-18). Currency: USD.

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice. The analysis that follows is deliberately position-free and carries no price target.

Verdict: HOLD / accumulate-on-weakness. A fairly-priced, well-run, low-beta yield-carry midstream with a genuine NGL franchise — but no margin of safety at $85, and a deal-driven growth story whose per-share payoff is still unproven. Constructive entry zone ~$70–78 (≈11–11.5x EV/EBITDA, ≈5.3–5.8% yield); above ~$90 the risk/reward inverts.

ONEOK is a better business than its ~8% ROIC suggests and a worse compounder than its 26-year dividend streak implies — and the tension between those two facts is the whole story. The crown jewel is real: the only integrated NGL system that physically links Bakken supply through the Mid-Continent to Gulf Coast fractionation and export, a genuine economies-of-scale-plus-customer-captivity moat that throws off ~90% fee-based, utility-stable cash flow. But between May-2023 and January-2025 management spent ~$24B (Magellan, EnLink, Medallion) to bolt diversification and Permian crude onto that franchise — and the realized scorecard through 2025 is sobering: share count up ~41% (447M→630M), net debt up from $13B to $33B (~4.5x EBITDA vs. a ~3.5x target), ROIC compressed from 11% to ~8% (now roughly at WACC), and diluted EPS dead flat at ~$5.4 even as adjusted EBITDA nearly doubled. The market is being asked to pay ~12x EV/EBITDA / ~15x earnings today for a post-mid-2027 free-cash-flow inflection that hasn’t arrived. That’s not a bad bet, but it is a priced bet.

I land on HOLD because the valuation is honest — own-history valuation percentiles put OKE at the 43rd percentile composite (P/E 39th, P/B 28th), i.e. squarely mid-range, neither the screaming bargain nor the bubble — and the ~4.9% dividend is well-covered (1.2–1.3x DCF) and almost certain to keep growing. The factor read confirms the framing: this is a low-beta (0.62), high-dividend-yield energy carry name (top single-stock comp is Targa; the rest of the comp set is energy ETFs), up ~19% over six months and off only ~3.5% last quarter — not a momentum trade, not deep value, and not a falling knife. You get paid to wait, and the FCF inflection plus a still-~88%-unfunded $2B buyback are genuine forward catalysts. But you are buying scale that hasn’t yet translated to per-share value, at a price that already credits the synergies. Conviction: medium. Flips bullish if 2026–27 delivers the guided ~$8.25B+ EBITDA with leverage through 3.5x and EPS finally breaking above ~$6 (proof the deals compound per-share). Flips bearish on a Permian/Bakken volume rollover or NGL-price break that exposes the ~10% commodity tail and stalls deleveraging with debt at 4.5x. Tag: “Bought scale at full price; the per-share payoff is still on the come.”

📈 Stock Price Action — Five-Year Event Map

Over the trailing five years OKE round-tripped from a post-COVID recovery base of ~$49 (Aug-2021) to an all-time high of ~$117 (22-Nov-2024), then gave back roughly a third before stabilizing — it now trades at ~$85 (close 2026-06-18), about 21.6% below its 5-year high, inside a 52-week range of ~$62.7 (Nov-2025) to ~$95.2 (19-May-2026). The stock has been a low-beta (~0.62), ~4.9%-yield energy carry vehicle, not a high-momentum name: the dominant moves track oil/energy-sector beta and the dividend-yield factor, not idiosyncratic re-rating.

Cycle narrative. (1) OKE rode the 2021–22 energy bull as post-COVID demand and the Russia-Ukraine oil spike lifted the whole midstream complex. (2) It then sold off into mid-2022 on recession fears and aggressive Fed hikes — a high-yield, rate-sensitive name de-rating with the macro. (3) The 2022–2023 recovery was reinforced by the ~$18.8B Magellan acquisition (closed Sep-2023), which added a refined-products/crude segment and a growth narrative. (4) The 2024 leg to the ~$117 ATH paired the EnLink/Medallion Permian expansion with a broad risk-on bid for energy and utility-like yield. (5) From the Nov-2024 peak the stock lost ~43% into Oct-2025 as the market repriced integration risk, elevated leverage (~4.5x net debt/EBITDA vs. ~3.5x target), EPS held flat by acquisition dilution, and softer oil. (6) The Oct-2025→May-2026 recovery — a ~+19% raw six-month total-return move — was driven by raised 2026 guidance and a renewed dividend-yield/oil-beta bid. (7) The most recent leg is a shallow ~11% give-back off the May high as oil and energy-sector factor returns rolled over; the 18-Jun-2026 shelf prospectus carried no disclosed terms. Price moves are facts from the 5-year daily price series; attributed causes are interpretation.

1. Executive Summary

ONEOK is one of the largest integrated North-American midstream operators — a ~$57B-equity / ~$90B-enterprise-value company running a roughly 60,000-mile network that gathers, processes, fractionates, transports, stores, and exports natural gas, natural gas liquids (NGLs), refined products, and crude oil across four segments. It is structured as a C-corporation (1099, not a K-1), distinguishing it from MLP peers such as Energy Transfer, Enterprise Products, and MPLX, which broadens its investor base.

The investment question is not whether OKE owns good assets — it does. The question is whether the 2023–2025 acquisition program created per-share value or merely scale. In ~16 months management acquired Magellan Midstream (~$18.8B, Sep-2023), the controlling and then remaining interests in EnLink Midstream (2024–Jan-2025), and Medallion Midstream (~$2.6B Permian crude, Oct-2024). Asset base went from ~$24B to ~$67B; adjusted EBITDA roughly doubled to ~$7.3B (FY2025) with 2026 guided to ~$8.25B. But the deals were funded with ~41% share dilution and a tripling of net debt to ~$33B (4.5x EBITDA vs. a 3.5x target), and diluted EPS was flat — $5.48 (2023) → $5.42 (2025) — while ROIC fell from 11.0% to ~8.2%, roughly OKE’s cost of capital.

The moat is real but partial. The genuine franchise is the integrated NGL system — the only one physically linking Bakken supply through the Mid-Continent to Gulf Coast fractionation and export — a scale-plus-customer-captivity advantage that delivers ~90% fee-based, utility-stable cash flow. But only the ~8,300-mile interstate-pipeline segment (~11% of segment EBITDA) is true FERC-regulated franchise; the rest (gathering & processing, NGL, refined products & crude) is contestable, capital-intensive, volume- and commodity-levered logistics. That mix is why returns are utility-grade (~8%) rather than franchise-grade (15%+).

Valuation is mid-range, not cheap. EV/EBITDA ~12.1x, P/E ~15.2x, ~4.9% dividend yield, and own-history valuation percentiles at the 43rd composite — neither bargain nor bubble. The market is paying today for a synergy-and-deleveraging story that inflects to meaningful free cash flow only after the elevated growth capex rolls off around mid-2027. The 26-year dividend-growth streak (quarterly now $1.07, +4% YoY) is well-covered (~1.2–1.3x distributable-cash-flow coverage) and the executive-comp plan is anchored on ROIC and relative TSR — both genuine positives that partially offset the empire-building risk. The balance of evidence: a high-quality, defensive, fairly-valued midstream whose per-share value creation from a full-price deal spree remains to be proven.

No recommendation and no price target appear below this line (the single exception is Claude’s Take above).

2. Business Overview

ONEOK, Inc. (Oklahoma-incorporated, HQ Tulsa; CEO Pierce H. Norton II; CFO Walter Hulse; COO Randy Lentz; Chief Commercial Officer Sheridan Swords; FY-end December 31) is an integrated midstream energy company. It does not produce hydrocarbons; it provides the toll-road infrastructure between wellhead and end-market — gathering raw gas, processing it to remove NGLs, fractionating NGLs into purity products (ethane, propane, butane, natural gasoline), transporting and storing all of it, and increasingly marketing and exporting it. The economic engine is fees for moving volume, not commodity price speculation — management states consolidated earnings were “approximately 90% fee-based in 2025” (FY2025 10-K, Executive Summary; verified verbatim in the filing).

Four reportable segments (FY2025 10-K, Note R — segment adjusted EBITDA):

*FY2023 NGL includes a ~$778M Medford insurance gain; normalized ~$2,267M. **FY2023 R&C is a partial year (Magellan closed Sep-2023). (Source: OKE 10-K FY2025.)

Revenue character. Reported revenue (FY2025 ~$33.6B) is a misleading top line — it grosses up large commodity purchase-and-resale flows (cost of sales ~$24.9B), so revenue growth mostly reflects commodity price/marketing volume, not economic value. The honest lens is segment adjusted EBITDA and fee margin, which is why this memo leads with EBITDA, not revenue. The “fee-based” label also carries a nuance: it includes fee-with-percent-of-proceeds (POP) contracts in G&P, where OKE takes title to the gas, sells the commodity, and keeps a fee — leaving residual commodity and, more importantly, volumetric exposure. The cleanest demand-charge revenue sits in Natural Gas Pipelines (firm capacity reservations paid regardless of throughput); G&P is the most cyclical (tracks drilling and well declines); NGL exchange services include minimum-volume-commitment (MVC) contracts that floor revenue.

Customers and footprint. Counterparties are major and independent E&P producers (fee payers), petrochemical companies, refiners, utilities, LDCs, power generators, and exporters. Key basins: the Williston/Bakken (>3M dedicated acres — OKE’s crown G&P and NGL-takeaway position), Powder River, the Mid-Continent (Anadarko/SCOOP/STACK + Barnett, >1M acres), the Permian (Midland + Delaware, growing via Medallion/Bighorn), the Conway (Kansas) NGL hub, and the Gulf Coast (Mont Belvieu fractionation + LPG export).

Verdict: a genuine, large-scale, fee-based toll-road business with one premier franchise asset (the integrated NGL system) wrapped in a larger, more cyclical commodity-logistics book. Recurring/contracted revenue dominates, but ~10% commodity exposure and meaningful volumetric sensitivity in G&P mean this is defensive, not bond-like.

3. Industry Dynamics

US midstream is best understood as two businesses bolted together (a framing consistent with public midstream industry analysis).

Tier 1 — Interstate FERC pipelines (regulated quasi-utility). Interstate gas pipelines (Natural Gas Act) and interstate NGL/refined-products pipelines (Interstate Commerce Act tariffs) earn cost-of-service allowed returns on certificated routes, with near-zero market-share volatility and demand-charge revenue largely insensitive to volume. Structurally excellent — stable, defensive, high barriers — but returns are statutorily capped at “just and reasonable,” not excess. This tier is why well-run midstream caps out around 8–13% ROIC rather than compounding at 20%.

Tier 2 — Gathering & processing and NGL gathering (competitive). G&P plants and gathering systems are explicitly exempt from FERC jurisdiction (the NGA gathering/processing carve-out). They compete basin-by-basin for acreage dedications, carry volume and commodity cyclicality, and face lower barriers to entry. Structurally average.

OKE’s mix is the key structural fact. Only the ~8,300-mile Pipelines segment (~11% of segment EBITDA) sits in Tier 1; the other ~89% (G&P, NGL, Refined Products & Crude) is Tier 2 commodity logistics. OKE is materially more weighted toward the average/cyclical tier than WMB (whose Transco interstate system is a Tier-1 route monopoly). This is the single most important industry distinction between the two and helps explain OKE’s lower regulated-moat protection and its commodity/volume beta.

Demand drivers — four real secular tailwinds. (i) Permian associated-gas growth from oil drilling (mgmt cites Gulf Coast volumes +30% YoY in Q1-2026); (ii) NGL/LPG export growth (the Texas City ~400 MBbl/d LPG export terminal JV with MPLX, ~$1.0B); (iii) LNG feedgas (management guides US LNG export capacity to more than double over the decade; the Eiger Express ~3.7 Bcf/d Permian-to-Katy project); and (iv) data-center power demand for natural gas, explicitly cited as a Natural Gas Pipelines growth driver. US dry-gas production exceeds ~105 Bcf/d and is growing. These tailwinds are genuine and largely contracted, not speculative.

Capital cycle (Marathon lens) — the central caution. The entire sector is racing capital at the same demand pull simultaneously: new Permian-to-Gulf gas pipes (Matterhorn, Eiger, Blackcomb, Hugh Brinson), new NGL fractionation, and new LPG export docks are being built by OKE, WMB, ET, EPD, TRGP, KMI and MPLX at once. The permitting/right-of-way moat protects existing corridors but does not exempt this wave of new capacity from the classic capital-cycle pattern — high returns attract capital, lumpy supply arrives with a lag, returns compress. OKE’s growth capex is elevated (~$3.1B FY2025; capex well above D&A) through ~mid-2027 before an FCF inflection — textbook late-boom positioning, mitigated only by the demand-pull being real and largely contracted.

Verdict: a structurally mixed-to-favorable industry. Defensive and growing in aggregate, with multiple secular demand tailwinds — but OKE sits more toward the average/cyclical tier than the regulated/excellent tier, and the whole sector is in the capital-intensive, supply-adding phase of the cycle. That argues for owning the franchise assets but being disciplined on price and skeptical of straight-line EBITDA extrapolation.

4. Competitive Position

The moat, named (Greenwald taxonomy). OKE has a partial moat: economies of scale on an irreplaceable physical network + customer captivity (switching costs), plus a thin FERC-franchise sliver. The genuine crown jewel is the integrated NGL system — the only one that physically links Bakken NGL supply (where OKE holds the dominant gathering position) through the Mid-Continent (Conway) to the Gulf Coast (Mont Belvieu fractionation + export). Once a producer’s wells and a processing plant are physically tied into OKE’s NGL gathering header, switching is prohibitively costly (a competitor would have to lay new pipe to the same molecules), and the gather→fractionate→transport→export integration captures margin at each step. That integration, plus ~1.2 MMBbl/d of fractionation at 94% utilization, is slow and expensive to replicate. The interstate-pipe sliver (Northern Border, Roadrunner, Guardian, Viking) adds genuine FERC route-monopoly economics, but it is small relative to the whole.

Pressure-test — franchise or commodity logistics? The honest answer is partial, and utility-grade rather than high-return-compounding. The tell is the financial outcome: a moat that does not show up in returns is not much of a moat. OKE’s consolidated ROIC is ~8.2% (FY2025), down from ~11% in 2023, with return-on-capital ~12.7% — below the 15%+ threshold for a true Greenwald franchise. The moat manifests in cash-flow stability and ~90% fee-based revenue far more than in return level. The share-stability test passes on the NGL backbone (physics + integration lock-in) but is weak on G&P, where acreage dedications are contested at the margin (Targa, ET, KMI compete for the same Permian/Mid-Continent volumes) and earnings follow the drill bit.

Versus peers (public financial data, 2026-06):

The read: OKE is mid-pack on multiple and at the low end on ROIC. Crucially, TRGP and EPD earn 11–13% ROIC on comparable NGL/G&P assets — so OKE’s ~8% is not a structural ceiling for the asset class. It partly reflects (i) the ~$11B of goodwill and intangibles from the Magellan/EnLink/Medallion spree inflating the invested-capital base, and (ii) a thinner regulated core than WMB. OKE’s differentiator is the irreplaceable Bakken NGL takeaway franchise (its single best, most-captive asset) and wellhead-to-export-dock integration; its weakness is that the surrounding book is ordinary, capital-hungry, commodity-levered logistics now carrying acquisition goodwill.

Verdict: a real but partial scale-plus-captivity moat anchored on the integrated NGL system, generating utility-grade (~8%) returns — not a high-ROIC compounder. The integrated Bakken→Mid-Continent→Gulf NGL chain is genuinely defensible and hard to replicate; the G&P and commodity-marketing book around it is contestable. OKE owns one of the better NGL franchises in North America but earns franchise-grade stability, not franchise-grade returns — and recent EBITDA growth was bought substantially with acquisitions that lifted scale while diluting ROIC.

5. Growth History and Forward Opportunities

Historical growth — scale up, per-share flat. Adjusted EBITDA roughly doubled from ~$3.4B (2022) to ~$7.3B (2025), a ~29% CAGR — but almost entirely acquired, not organic. The cleanest evidence is per-share: diluted EPS went $3.84 (2022) → $5.48 (2023) → $5.17 (2024) → $5.42 (2025), i.e. flat across the three deal years as the ~41% share-count increase (447M→630M) offset nearly all the EBITDA growth. Dividend per share grew steadily ($3.74→$4.13) but at a far slower pace than EBITDA. This is the defining tension: OKE grew the enterprise aggressively and the per-share claim barely at all over 2023–2025.

Organic growth is genuine but slower-burn. Beneath the M&A, base business volumes are growing: Q1-2026 Gulf Coast Permian volumes +30% YoY; processed gas, NGL raw feed throughput, and fractionation utilization (94%) are trending up. Management’s organic growth-capex program targets NGL pipeline expansions (Elk Creek, West Texas NGL), the Eiger Express gas pipeline (~3.7 Bcf/d Permian→Katy), the Texas City LPG export JV with MPLX (~400 MBbl/d, ~$1.0B), and Bakken/Permian processing additions.

Forward opportunities (largely optionality, not yet in numbers).

• NGL/LPG export: the Texas City terminal extends OKE’s export reach and pulls more NGL through the integrated system — the highest-quality forward driver because it monetizes the existing franchise.
• LNG feedgas: Eiger Express and Gulf Coast gas takeaway tie OKE to the doubling of US LNG export capacity.
• Data-center / power-generation demand: management has floated power-gen and data-center-adjacent JV ventures and cites data-center demand as a Natural Gas Pipelines growth driver. Interpretation: this is the same narrative every midstream is now telling; treat as optionality pending actual capital commitments and contracts.
• The FCF inflection: the cleanest forward catalyst is mechanical — larger growth-capex projects complete around mid-2027, after which free cash flow “really kicks in” (CFO, Q1-2026 call), enabling faster deleveraging and a more meaningful buyback.

2026 guidance (raised, Q1-2026 call): adjusted EBITDA midpoint ~$8.25B (up from prior), net income ~$3.5B, diluted EPS ~$5.53 — the first guided year EPS meaningfully exceeds the 2023 level, the early signal that per-share growth may finally be resuming as integration matures.

Verdict: mixed-quality growth. The enterprise growth was real but expensively acquired and per-share-dilutive through 2025; the organic growth is genuine, franchise-monetizing, and higher-quality but slower. The thesis hinges on the 2026 guided EPS step-up and the post-2027 FCF inflection converting scale into per-share value. Promising, but unproven.

6. Financial Quality

Income statement (annual, $M):

Revenue swings with commodity/marketing pass-through and is not a quality signal; the FY2025 EBITDA-margin drop to 21.8% reflects the lower-margin Magellan refined-products marketing flows grossing up the denominator, not a deterioration in economics. The real story is EBITDA growth (acquired) and flat EPS (dilution).

Cash flow (annual, $M):

Operating cash flow is high-quality and converts well above net income (OCF/NI ~1.65x, reflecting heavy non-cash D&A and deferred tax). Quality-of-earnings note: the gap between OCF and net income is favorable here — D&A (~$1.5B) and deferred taxes (~$0.96B in 2025) are real non-cash adds, not aggressive accruals. The pinch point is free cash flow after growth capex: 2025 OCF $5.6B − ~$3.1B growth capex ≈ $2.5B, against $2.58B of dividends — i.e. dividend coverage on a post-growth-capex FCF basis was roughly 1.0x in the peak-capex year. Management’s distributable-cash-flow coverage (which adds back growth-capex as discretionary) is healthier at ~1.2–1.3x. Both can be true; the point is that until growth capex rolls off (~mid-2027), there is little organic FCF left over after the dividend to delever quickly — deleveraging has relied partly on EBITDA growth raising the denominator.

Balance sheet (annual, $M):

The balance sheet is the chief concern. Net debt tripled from $13.4B to $32.7B in three years; net-debt/EBITDA is ~4.5x against a stated ~3.5x target. Interest expense is now ~$1.78B/yr and an August-2025 refinancing carried coupons up to ~6.25% (2055 notes) — higher-cost debt replacing maturities. Goodwill + intangibles total ~$11B; tangible book value is ~$18.6/share, so the stock trades ~4.6x TBV (typical for an asset-heavy network but a reminder that much of the equity is acquisition premium). Liquidity is adequate (revolver, $1.0B ATM untapped, investment-grade ratings) but the leverage leaves limited cushion for a volume/commodity downturn.

Returns. ROIC ~8.2% (2025), down from 11.0% (2023); return-on-capital ~12.7%. ROE optics (>170%) are meaningless — they reflect a thin GAAP equity base relative to debt, not high profitability. The honest return metric is ROIC near WACC.

Verdict: high-quality cash generation, stretched balance sheet, returns near cost of capital. Economics do not yet improve with scale — the opposite happened (ROIC fell as the company grew by acquisition). The bull case requires that to reverse as synergies mature and capex rolls off; the evidence through 2025 does not yet show it.

7. Capital Allocation

The M&A scorecard — ~$24B in ~16 months:

The Magellan multiple is the crux. Street pegged the headline at ~11.5–13x forward EBITDA — full for a low-growth refined-products MLP. Management’s counter was that a ~$1.5B tax-basis step-up (from the C-corp conversion) plus a guided $200M+ synergy ramp pulled the effective multiple toward ~8x. That is the bull-case math and remains partly interpretation, not fully realized fact. EnLink and Medallion at ~8–11x were more defensible and Permian-strategic (crude pull-through into the Gulf Coast system).

Smart allocation or empire-building? The case against is documented above: ~41% dilution, net debt $13B→$33B, ROIC 11%→8% (now ~WACC), and flat EPS even as EBITDA doubled — the textbook signature of capital deployed at the cusp of value-neutrality. The case for: the moves are strategically coherent — diversification into crude/refined products lowers NGL-price beta, and Permian crude gathering feeds volume pull-through into OKE’s NGL/Gulf-Coast network (not just bolt-on EBITDA). The honest grade is C+ / “bought scale at a full price; per-share value creation unproven” — empire-building risk materially mitigated by genuine strategic logic and a credible path to the FCF inflection.

Dividend — the crown of the capital story. 26+ consecutive years of dividend growth, one of the longest streaks in midstream. The quarterly was raised to $1.07 (+4% YoY), a ~$4.28 annualized run-rate; yield ~4.9% at $85. GAAP payout ~73%, DCF coverage ~1.2–1.3x. The dividend was protected and grown straight through the M&A spree — real discipline on the distribution even as leverage rose.

Buyback — authorized but barely used. The Board authorized a $2.0B repurchase program (Jan-2024), but only ~$234M (~12%) had been bought through 12/31/2025 — deleveraging and growth capex took priority. Meaningful buyback is an option, not an active lever, and likely stays muted until leverage reaches ~3.5x (post-2027). A fresh shelf prospectus was filed 18-Jun-2026 (issuance capacity, not a current dilution event); the $1.0B ATM was untapped as of Feb-2026.

Executive comp — better-aligned than the spree suggests (DEF 14A, 2026). The annual incentive plan is keyed on EPS + ROIC plus operational metrics; long-term performance units on EPS, ROIC, and relative TSR vs. a midstream peer set. Tellingly, the 2022–2025 performance units paid out at only ~50% of target — the plan docked executives for exactly the period of dilution and flat EPS, a credible alignment signal. Anchoring LTI to ROIC and relative TSR (not size or absolute EBITDA) is the structure that disincentivizes value-destructive empire-building.

Verdict: a protected 26-year dividend and a well-designed, ROIC/TSR-anchored comp plan are genuine positives — but the headline scorecard (41% dilution, 3x’d net debt, flat EPS, ROIC to WACC, ~88%-unfunded buyback) marks 2023–25 as strategically coherent yet expensively executed, with per-share returns still on the come.

8. Changes and Headwinds — Last Two Years

The transformation. OKE went from a focused NGL/gas-pipeline operator (~$24B assets, 2022) to a four-segment diversified midstream giant (~$67B assets, 2025) by absorbing Magellan (and its MLP→C-corp conversion), EnLink (now 100%-owned), and Medallion in ~16 months. The addition of the Refined Products & Crude segment, Permian crude gathering, and substantially more leverage makes today’s OKE a different, more complex, more crude-exposed company than the 2022 version.

Forward catalysts (2025–2026 narrative pivot). Management has shifted emphasis to demand tailwinds: data-center/power electricity demand, US LNG export capacity guided to more than double over the decade, and Permian-to-Gulf-Coast volumes +30% YoY. Power-generation and data-center-adjacent JV ventures have been floated. Interpretation: real secular tailwinds, but largely optionality not yet in numbers, and the data-center framing is now ubiquitous across midstream — treat as hypothesis pending firm capital commitments.

Headwinds. (1) Leverage at ~4.5x net-debt/EBITDA (~1x above target) with ~$1.78B/yr interest expense and refinancing at higher coupons; (2) integration execution across three large deals at once; (3) commodity/volume sensitivity — a Permian/Bakken drilling slowdown or NGL-price break hits gathered volumes and the ~10% commodity tail; (4) per-share dilution overhang until the FCF inflection; (5) interest-rate/refinancing sensitivity on the floating and near-maturity portions.

Insider & 8-K read. The trailing-24-month 8-K timeline is routine: buyback + dividend authorization (Jan-2024); EnLink-GIP / Medallion agreements (Aug–Sep 2024); EnLink full-acquisition completion (Jan-2025); a $3.0B note offering (Aug-2025 refi); quarterly earnings/dividends; a fresh shelf (Jun-2026). No litigation or guidance-cut bombshells. The Form 4 corpus is overwhelmingly routine (RSU/director grants, option-exercise tax withholding); the lone discretionary open-market purchase in 24 months is director Brian L. Derksen buying 2,500 shares at ~$66 (3-Nov-2025) — a small but genuine conviction buy ~22% below today’s price. Insiders are not signaling distress; one director put real money in near the lows — a mild positive, not a strong tell.

Verdict: the changes meaningfully increase scale and optionality but also complexity, leverage, and execution risk — net, they raise both the potential reward and the downside, and the thesis now rests on integration and deleveraging delivering as promised.

9. Risk Analysis

Catastrophic-loss assessment: the probability of permanent total loss is very low — OKE is investment-grade, hard-asset-backed, and ~90% fee-based, with a 26-year dividend record through multiple cycles (including the −80% 2020 drawdown, from which it fully recovered). The realistic downside is a multi-year de-rating + dividend-growth stall if leverage and volumes disappoint, not impairment of the franchise itself.

10. Valuation Discussion (Embedded Expectations)

OKE trades at ~12.1x EV/EBITDA (TTM), ~15.2x P/E, ~2.6x EV/Sales, and a ~4.9% dividend yield, with EV ~$90.5B on ~$57B equity and ~$33B net debt (aggregated financial data, 2026-06). On own-history valuation percentiles, the stock sits at the 43rd composite percentile (P/E 39th, P/B 28th, P/S 63rd) — mid-range versus its own multi-year history, neither cheap nor expensive. Against peers, the ~12.1x multiple is mid-pack (cheaper than TRGP ~14x and WMB ~13.5x; in line with EPD ~12.3x; richer than ET) — appropriate for a company with a lower ROIC and higher leverage than the premium names.

What the current price embeds (reverse-DCF logic). At ~12x EV/EBITDA on ~$8.25B 2026 guided EBITDA, the market is underwriting: (1) the guided EBITDA growth is delivered and continues mid-single-digit-plus beyond 2026; (2) leverage descends to ~3.5x without a dilutive equity raise, on the back of EBITDA growth and the post-2027 FCF inflection; (3) synergies and Permian/NGL/LNG/export volumes ramp roughly as guided; and (4) the ~4.9% dividend keeps growing at a low-to-mid-single-digit pace. In short, the price credits a successful integration and clean deleveraging — the bull path — at a fair, not cheap, multiple.

Scenario sketch (illustrative, not a target):

• Bear: a Permian/Bakken volume rollover or NGL-price break stalls EBITDA near ~$7.5–8B, deleveraging stalls at ~4.3–4.5x, dividend growth slows to ~2%; the multiple compresses toward ~10–10.5x and the equity de-rates (echo of the Nov-2024→Oct-2025 −43% move). The yield + the franchise provide a floor; the downside is a multi-year stall, not impairment.
• Base: 2026 EBITDA ~$8.25B as guided, EPS steps to ~$5.5, leverage grinds toward ~4.0x by 2027 then ~3.5x as capex rolls off; the multiple holds ~11.5–12.5x; total return ≈ ~4.9% yield + low-to-mid-single-digit growth.
• Bull: synergies fully land, EBITDA pushes toward ~$9B+, the post-2027 FCF inflection funds a real buyback and faster deleveraging, EPS breaks above ~$6, and the market re-rates OKE toward the EPD/TRGP quality tier (~13x+). This is the scenario the bulls pay for.

What the market may be getting right vs. wrong. Right: the cash-flow stability, the franchise quality of the NGL system, and the deleveraging trajectory given EBITDA growth. Potentially wrong (either direction): whether the per-share value creation finally arrives — the bears can point to three years of flat EPS and ROIC at WACC; the bulls to the 2026 guided EPS step-up and the mechanical FCF inflection. The valuation does not obviously misprice the stock in either direction — which is precisely why this is a “fairly valued, prove-it” situation. No price target.

11. Variant Perception

Consensus view. OKE is a high-quality, diversified, ~90%-fee-based midstream compounder with a 26-year dividend streak, a clear deleveraging path, and multiple secular demand tailwinds (LNG, NGL exports, data-center gas) — a defensive yield-plus-growth core holding. Sell-side is broadly constructive.

Strongest bull case. The acquisition spree built irreplaceable scale and a uniquely integrated wellhead-to-export franchise at a cyclically reasonable effective price; synergies and organic volumes will lift EBITDA toward $9B+, the post-2027 FCF inflection will simultaneously fund deleveraging and buybacks, EPS will finally break out, ROIC will recover toward the peer 11–13%, and the market re-rates the stock toward the premium-quality tier. You are paid ~4.9% to wait for a mechanical inflection.

Strongest bear case. OKE bought EBITDA, not per-share value: three years of flat EPS, ROIC compressed to WACC, ~41% dilution, and net debt tripled to 4.5x. The “fee-based” label masks real volumetric and ~10% commodity exposure; a Permian/Bakken slowdown or NGL break would stall both EBITDA and deleveraging with little FCF cushion after the dividend. The whole sector is adding capacity into the same demand, setting up capital-cycle return compression. At ~12x EV/EBITDA you are paying a full price for an unproven integration with a stretched balance sheet.

The 3–5 assumptions that matter most: (1) integration synergies are realized and the 2026 EPS step-up holds; (2) leverage descends to ~3.5x without dilutive equity; (3) Permian/Bakken volumes and NGL/export demand ramp roughly as guided; (4) the post-2027 FCF inflection actually arrives on schedule; (5) NGL/commodity prices and oil-price beta don’t break the ~10% commodity tail. Falsification: a volume/commodity rollover that stalls EBITDA and deleveraging (bear-confirming), or EPS breaking decisively above ~$6 with leverage through 3.5x (bull-confirming).

Factor-positioning input (quantitative factor model). OKE is, empirically, a low-beta (0.62), high-dividend-yield energy carry name — its loadings are dominated by Sector:Energy (+0.91), OilPrice (+0.89), and DividendYield (+0.78), with Momentum near zero, Value absent, and Growth negative. The top single-stock comp is Targa (cosine 0.969); the rest of the comp set is energy/MLP ETFs. The risk-adjusted record is solid-but-unspectacular (3yr +18%/yr, Sharpe 0.62). The recent shallow pullback (−3.5% last quarter after +19% over six months) is fully explained by softening energy/oil factor returns, not anything stock-specific. The regime is mixed — the dividend-yield factor is being paid (z +1.44), but low-volatility (OKE’s profile) is out of favor (z −1.23). Implication for consensus: the tape is not signaling that consensus is badly offsides in either direction. This is a fairly-priced carry name, which supports the “prove-it HOLD” framing rather than a contrarian high-conviction call. Statistical estimates, not a price call.

12. Fact vs. Interpretation Table

13. Open Questions

1. Realized vs. promised Magellan synergies — how much of the $200M+ run-rate target has actually hit EBITDA, and what is the verified effective acquisition multiple today?
2. FCF inflection timing — is mid-2027 firm, or will incremental data-center/LNG/export projects keep growth capex elevated and push it out?
3. Leverage path — can OKE reach 3.5x by 2027–28 without any equity issuance, under a flat-to-down commodity scenario?
4. Commodity tail — what is the precise EBITDA sensitivity to NGL frac spreads and gas/oil prices within the “~10% commodity-exposed” bucket?
5. Buyback intent — will the post-inflection FCF actually be returned via buyback, or absorbed by further M&A? (Management’s bias has been to acquire.)
6. Data-center/power JVs — are these real, contracted, capital-committed projects or narrative optionality?
7. G&P volume durability — how exposed is the largest EBITDA segment to a Permian/Bakken drilling slowdown over the next two years?

14. What Must Be True

Bull case — what must be true: (a) 2026 adjusted EBITDA lands at/above the ~$8.25B guide and EPS steps decisively above the ~$5.4 plateau toward $6+; (b) leverage descends to ~3.5x by ~2027–28 with no dilutive equity; © synergies and Permian/NGL/LNG/export volumes ramp roughly as guided; (d) the post-2027 FCF inflection funds both deleveraging and a real buyback. Falsification test: if by year-end 2027 EPS is still stuck near ~$5.5 and leverage remains above ~4.0x, the bull thesis (that scale converts to per-share value) is broken — the deals will have bought EBITDA and nothing else.

Bear case — what must be true: (a) a Permian/Bakken volume rollover or NGL-price break stalls EBITDA and exposes the commodity/volumetric tail; (b) deleveraging stalls with little FCF cushion after the dividend; © sector-wide capacity additions compress incremental returns; (d) the stock de-rates toward ~10x as the growth-and-deleveraging narrative disappoints. Falsification test: if OKE delivers the guided EBITDA, takes leverage through 3.5x, and grows EPS above ~$6 while sustaining ~5% dividend growth, the bear thesis (over-levered, value-neutral empire-building) is refuted.

15. Source Appendix

See the separate Source Appendix (Appendix B in the combined report) for the full citation list. Primary sources: ONEOK 10-K FY2025 (filed 2026-02-24, CIK 0001039684), 10-Qs and 8-Ks (2021–2026), DEF 14A (2026), Form 4 filings; Q1-2026 earnings call transcript (2026-04-29); aggregated financial data (accessed 2026-06-20); daily price history and own-history valuation percentiles (2026-06-18); a quantitative factor model (2026-06-18/19); public Williams Companies (WMB) disclosures (peer/industry framing). All figures reconciled to the FY2025 10-K where applicable; third-party aggregated data are cross-checks, not primary.

APPENDIX A — Standard Diligence Questionnaire

Supplemental to the research memo. Report date 2026-06-20. Fact/Interpretation/Assumption labeled where it matters.

General

What thoughtful questions have other investors asked about this company? The dominant institutional debate is whether the ~$24B Magellan/EnLink/Medallion acquisition program created per-share value or just scale — with skeptics pointing to flat EPS (~$5.4 across 2023–2025), ROIC compressed to ~WACC (~8%), and net debt tripled to ~4.5x EBITDA. Second-order questions: (i) when the post-mid-2027 free-cash-flow inflection actually arrives and whether it funds buybacks vs. more M&A; (ii) the realized vs. promised Magellan synergies and effective multiple; (iii) the true commodity/volumetric sensitivity beneath the “~90% fee-based” label; and (iv) whether the data-center/LNG/NGL-export tailwinds are contracted or narrative.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: mid-cycle, skewed slightly high on commodity/volume but depressed on a per-share basis. Absolute EBITDA (~$7.3B, guided ~$8.25B for 2026) is at a record, lifted by acquisitions; per-share earnings are below potential because integration capex and dilution are still being absorbed. Driven by external environment or internal actions? Both — acquired EBITDA is internal action; the ~10% commodity tail and G&P volumes are external (Permian/Bakken drilling, NGL prices, oil-price beta ~0.96). How stable are revenues? Reported revenue swings with commodity pass-through and is not a stability signal; the relevant metric — fee-based segment EBITDA — is highly stable (~90% fee-based, MVCs, firm pipeline capacity). Outlook for products/services? Structurally growing: US gas/NGL production >105 Bcf/d, LNG export capacity guided to more than double over the decade, NGL/LPG exports rising, data-center power demand emerging. How big, growing or shrinking, domestic or international? A large and growing domestic infrastructure market with international demand pull (LNG/NGL exports); OKE’s assets are US-based with Gulf Coast export reach.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Interpretation: more competitive in the gathering/processing tier (basin-by-basin acreage competition, sector-wide capacity additions into the same demand — a Marathon capital-cycle caution), but stable in the regulated interstate-pipeline tier. How profitable is the business (ROIC, ROE)? ROIC ~8.2% (FY2025), down from 11.0% (2023) — utility-grade, near WACC; reported ROE (>170%) is a meaningless artifact of thin GAAP equity. How profitable is the industry — competitors, barriers? Mid — interstate pipes earn FERC-capped returns; G&P is contestable. Premium peers (TRGP ~13.3%, EPD ~10.8% ROIC) earn more on comparable assets, so OKE’s ~8% is partly self-inflicted (acquisition goodwill, leverage). Barriers: high for existing corridors (right-of-way, permitting, integration), lower for new G&P entrants. Can the business be easily understood? Yes — a fee-for-volume toll-road on hydrocarbons, four segments. Undermined by foreign low-cost labor? No — fixed domestic infrastructure. Do brands matter? No; what matters is physical network integration and switching costs (once connected, producers can’t cheaply switch). Nature of competition? Acreage dedications, contract terms, basin connectivity, and integration reach. Customers’ switching costs? High on the integrated NGL backbone (physics + dedicated infrastructure); lower at the gathering margin.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The irreplaceable right-of-way/network position and integration value are worth more than book; conversely ~$11B of goodwill/intangibles is acquisition premium. Off-balance-sheet liabilities? Standard for the sector — operating leases, JV obligations, environmental/asset-retirement obligations, pipeline-integrity commitments; nothing flagged as outsized. How conservative is the accounting? Interpretation: reasonably conservative — OCF converts well above net income (~1.65x) on real non-cash D&A/deferred tax, not aggressive accruals; the main critique is the gross-revenue presentation that flatters scale optics. How CapEx-hungry is the business? Very — growth capex ~$3.1B in 2025 (above D&A); this is the central near-term constraint, easing only after ~mid-2027.

Capital Allocation & Management

How much FCF, and how is it used? OCF ~$5.6B (2025); after ~$3.1B growth capex, ~$2.5B FCF roughly matched the ~$2.58B dividend — i.e. post-growth-capex FCF coverage ~1.0x in the peak-capex year (DCF-basis coverage ~1.2–1.3x). Priorities: dividend (protected, 26-yr growth), then deleveraging, then a token buyback. Philosophy? Investment-grade balance sheet, growing dividend, disciplined growth capex, opportunistic M&A — with a recent bias toward acquisition. Significant acquisitions recently? Yes — the defining feature: Magellan (~$18.8B, 2023), EnLink (2024–25), Medallion (~$2.6B, 2024). Buying back shares? Authorized $2.0B (Jan-2024); only ~$234M (~12%) used through 2025 — minimal. Issuing shares to insiders? Routine RSU/PSU grants; ~41% total share growth was almost entirely acquisition stock, not insider largesse. Compensation policy? Anchored on EPS, ROIC, and relative TSR; 2022–25 PSUs paid only ~50% of target — a credible, well-aligned plan that docked execs for the dilution period. Motivations of management? Interpretation: growth-oriented and acquisitive, but constrained by a ROIC/TSR-linked comp structure and a protected-dividend culture; one director made a genuine open-market buy near the 2025 lows.

Valuation & Market Data

ADR, MLP, or K-1 issuer? Neither — OKE is a C-corporation issuing a 1099 (a structural advantage vs. MLP peers ET/EPD/MPLX that issue K-1s and complicate tax-exempt/foreign ownership). Dividend policy? ~$1.07/quarter ($4.28 annualized run-rate), ~4.9% yield, 26+ years of growth, ~73% GAAP payout / ~1.2–1.3x DCF coverage. How profitable? Utility-grade — ~8% ROIC, ~22% EBITDA margin (depressed optically by Magellan marketing pass-through), high and stable cash generation. Net income diverging from cash from operations? Yes, favorably — OCF runs ~1.65x net income on non-cash D&A/deferred tax; no red-flag divergence.

Risks & Downside

What would cause the stock to decline? A Permian/Bakken volume rollover or NGL/commodity-price break stalling EBITDA and deleveraging; rising rates pressuring a high-yield, rate-sensitive equity; integration/synergy disappointment; or a dilutive equity raise. The Nov-2024→Oct-2025 −43% drawdown is the template. Risk of catastrophic loss? Low — investment-grade, hard-asset, ~90% fee-based, fully recovered from the −80% 2020 drawdown; a major pipeline incident is a low-probability/high-impact tail (insured). Chance of total loss? Very low — total loss would require extreme, sustained distress inconsistent with the asset base and contract profile.

Recent News & Events

Has the business environment changed recently? Yes structurally over two years (the Magellan/EnLink/Medallion transformation); the 2026 narrative pivot to data-center/LNG/NGL-export demand is the current emphasis. Significant acquisitions? Covered above. Change in accounting policies? None material flagged. Recent changes — new markets, facilities, management? New Refined Products & Crude segment and Permian crude footprint; raised 2026 guidance (adj. EBITDA ~$8.25B); fresh shelf prospectus (Jun-2026, no terms); stable senior management (Norton/Hulse/Lentz/Swords).

APPENDIX B — Source Appendix

Report date 2026-06-20. Primary sources prioritized; third-party aggregated data labeled as cross-checks. All material figures reconciled to SEC filings where applicable.

Primary — SEC filings (CIK 0001039684)

• Form 10-K, FY2025 — filed 2026-02-24 (oke-20251231.htm). Item 1 Business (segment descriptions, ~90% fee-based statement, basin footprint, mileage/capacity), Item 1A Risk Factors, Item 7 MD&A, Note R Segments (segment adjusted EBITDA), debt footnotes, dividend history.
• Form 10-K, FY2024 / FY2023 / FY2022 — filed 2025-02-25 / 2024-02-27 / 2023-02-28. Multi-year segment, balance-sheet, and capex history.
• Form 10-Q — quarterly filings 2022–2026.
• Form 8-K — material events: $2.0B buyback + dividend authorization (Jan-2024); EnLink-GIP / Medallion purchase agreements (Aug–Sep 2024); EnLink full-acquisition completion (Jan-2025); $3.0B senior notes offering (Aug-2025); quarterly earnings/dividend declarations…
• DEF 14A (proxy) — 2026; executive compensation structure (EPS / ROIC / relative-TSR metrics), 2022–2025 performance-unit payout at ~50% of target, clawback policy, board composition…
• Form 4 (insider transactions) — trailing 24 months; overwhelmingly routine grants/withholding; lone discretionary open-market purchase: director Brian L. Derksen, 2,500 shares at ~$66, 2025-11-03.
• S-3ASR / POSASR shelf prospectus — filed 2026-06-18 (issuance capacity; no terms).

Primary — Earnings call transcript

• Q1-2026 earnings call — 2026-04-29 (company earnings call). Raised 2026 guidance (adj. EBITDA midpoint ~$8.25B, net income ~$3.5B, diluted EPS ~$5.53); Gulf Coast Permian volumes +30% YoY; capital program completing ~mid-2027 with FCF inflection thereafter; LNG export capacity guided to more than double over the decade; data-center power-demand framing; capital-allocation priorities (dividend, debt repayment, other shareholder returns). Management: Pierce Norton (CEO), Walter Hulse (CFO), Randy Lentz (COO), Sheridan Swords (CCO).
• Prior earnings calls (Q4-2025 through Q2-2024) available publicly for trend context.

Quantitative cross-checks (third-party aggregated — not primary)

• Aggregated financial data providers (accessed 2026-06-20): income statement, balance sheet, cash flow (FY2020–2025); profitability ratios (ROIC, ROE, margins); per-share data; enterprise value (EV ~$90.5B, EV/EBITDA ~12.1x TTM); valuation multiples; company profile; peer ROIC/EV-EBITDA (EPD, TRGP, WMB, KMI, ET). Reconciled to the 10-K.
• Daily price history (2026-06-18): 5-year OHLCV, dividends, EMAs (21/50/200), beta ~0.62, alpha. Basis for the Five-Year Event Map.
• Own-history valuation percentiles (2026-06-18): own-history percentiles — P/E 15.16 (39th), P/B 2.40 (28th), P/S 1.52 (63rd), composite 43rd.
• News-flow review (2026-06-20): recent-events triage (sector-level; no thesis-changing items).
• Quantitative factor model (2026-06-18/19): stock-loadings (Sector:Energy +0.91, OilPrice +0.89, DividendYield +0.78; Momentum ~0, Value absent), leaderboard (y3 +18.0%/yr Sharpe 0.62; y5 +15.2%; m6 +42.7% ann; m3 −13.5% ann; lifetime maxDD −80%), stock-info (beta 0.62), related-stocks (top comp Targa, cosine 0.969; rest energy/MLP ETFs), factor-returns regime (DividendYield z +1.44; LowVol z −1.23).

Notes on method

• Revenue figures are gross of commodity pass-through and are not used as a quality metric; segment adjusted EBITDA and fee margin are the economic lens.
• Acquisition multiples (Magellan ~11.5–13x headline / ~8x management-effective; EnLink/Medallion ~8–11x) blend filing disclosure and street consensus; the effective Magellan multiple remains partly interpretive pending realized synergies.
• All Fact/Interpretation/Assumption distinctions are carried through the memo body and the Fact-vs-Interpretation table.