[OKE] ONEOK: Can Record NGL Volumes Turn Back Into Profit in Q3 2026?

ONEOK is a US integrated midstream company that gathers and processes natural gas, gathers, fractionates and transports natural gas liquids (NGLs), runs interstate and intrastate natural gas pipelines, and operates refined products and crude oil pipelines[1]. The company has scheduled its earnings call for 2026-10-27 to report the third quarter 2026, the three months ended September 30, 2026[2]. In the latest disclosed period, the second quarter of 2026, adjusted EBITDA was $2.121 billion, up 7% year over year, net income was $967 million and diluted EPS was $1.53[3]; NGL raw feed throughput reached a record 1,630 thousand barrels per day, up 7%, and refined products volumes shipped rose 8%[4]. On August 3 the company raised full-year guidance for the second time, setting 2026 adjusted EBITDA at $8.2 billion to $8.5 billion and the diluted EPS midpoint at $5.68, while keeping capital expenditures at $2.7 billion to $3.2 billion[5]. Subtracting first-half adjusted EBITDA of $4.118 billion, that guidance implies about $2.04 billion to $2.19 billion per quarter in the second half, but it predates the late-August Apollo and Brazos transactions and excludes the minority-interest deduction, interest savings and Brazos consolidation[3]. Analyst consensus compiled by Drillr puts third-quarter EPS at $1.51 (10 analysts, range $1.12 to $1.66) and third-quarter EBITDA at $2.113 billion, inside the quarterly range implied by guidance[6]; full-year EPS consensus of $5.75 sits slightly above the company's guidance midpoint[7].

Three things matter most in this report. First, whether the NGL segment can turn volume growth back into profit: raw feed rose 7% in the second quarter, yet NGL segment EBITDA was only $659 million, $14 million below the prior-year quarter[8], and management attributed the gap to a rising share of lower-fee ethane[9]; if volumes keep climbing in the third quarter while profit fails to recover, lower rates look more structural. Second, how far natural gas pipeline spread income falls back: the segment's first-half EBITDA was $236 million higher than a year earlier, mostly from the Waha-to-Katy location spread[10], and management has said second-half earnings will decline as new Permian takeaway capacity enters service[11], so the size of the drop determines how much other segments must make up for full-year guidance. Third, the Apollo deal appears in the financial statements for the first time: the company received Apollo's $9 billion on September 10[12] and plans to use $5 billion of it to reduce debt[13], so third-quarter debt, interest expense and net income attributable to the minority interest are the first data points for testing the financing cost and the pace of deleveraging.

Company Background and Business Structure

ONEOK is a long-established midstream company that has grown quickly through a series of acquisitions. Founded in 1906 and headquartered in Tulsa, Oklahoma, it bought Magellan Midstream, a controlling interest in EnLink and Medallion over the past three years, with the latter two completed in October 2024[1], turning an NGL-focused midstream company into an integrated platform spanning natural gas, NGLs, refined products and crude oil. On September 10, 2026, the company completed a holding-company reorganization: the listed company now holds the operating entity ONEOK, L.L.C. (OpCo) through ONEOK Holdings, and Apollo paid $9 billion in cash for 900 million nonvoting Class B units in Holdings[12].

The company operates four segments, and NGLs are its largest profit source. In 2025 the four segments together generated $7.955 billion of adjusted EBITDA: NGLs $2.779 billion (about 35%), Refined Products and Crude $2.177 billion (about 27%), Natural Gas Gathering and Processing $2.138 billion (about 27%) and Natural Gas Pipelines $861 million (about 11%)[14]. The NGL segment collects mixed NGLs from its own and third-party processing plants and moves them by pipeline to Conway and Mont Belvieu for fractionation into purity products, with 1.2 million barrels per day of fractionation capacity; under most contracts it buys unfractionated NGLs at the plant tailgate and deducts contractual fees[15]. The gathering and processing segment gathers and processes natural gas for producers in the Rocky Mountain, Mid-Continent and Permian regions, with combined processing capacity of 7.2 Bcf per day across the three regions[16].

The other two segments are mainly transportation businesses. The Refined Products and Crude segment gathers, transports, stores and markets refined products and crude oil, and has been building a refined products pipeline expansion that connects to Denver International Airport[17]. The Natural Gas Pipelines segment earns most of its revenue from firm transportation, where customers reserve a fixed amount of capacity for a set period and pay whether or not they use it, and the company also uses its assets and contract portfolio to capture location price differences[18].

Revenue mainly reflects pass-through commodity sales and does not translate directly into profit. Of $38.579 billion in combined segment revenue in 2025, liquids commodity sales were $30.408 billion and residue natural gas sales were $3.372 billion, while exchange services and gathering and processing fees were only $1.473 billion and transportation and storage revenue was $3.160 billion[14]. One customer accounted for about 12% of consolidated revenue that year[1]. The company says about 90% of its 2025 earnings were fee-based[19] and expects that share to stay around 90% in 2026[20].

Financial History and Current Position

Acquisitions drove a sharp rise in ONEOK's earnings over three years. Adjusted EBITDA rose from $5.243 billion in 2023 to $6.784 billion in 2024 and $8.020 billion in 2025, while net income including noncontrolling interests was $2.659 billion, $3.112 billion and $3.462 billion[21]. Interest expense rose from $866 million to $1.783 billion over the same period, showing that the expansion was funded mainly with debt[21].

In 2025 operating cash flow covered capital spending, but not capital spending and dividends together. Operating cash flow was $5.599 billion, capital expenditures were $3.152 billion, dividends were $2.583 billion, contributions to unconsolidated affiliates were $622 million, and the company issued $2.989 billion of new long-term debt[22]. Total debt at year-end was $32.816 billion[23], the credit-agreement ratio of debt to adjusted EBITDA was 4.3 times, and long-term debt was rated Baa2 by Moody's and BBB by S&P and Fitch[24].

Earnings kept growing in the second quarter of 2026, but the sources of growth were uneven. Second-quarter adjusted EBITDA was $2.121 billion (prior year $1.981 billion), net income was $967 million, net income attributable to ONEOK was $966 million and diluted EPS was $1.53 (prior year $1.34); first-half adjusted EBITDA was $4.118 billion and diluted EPS was $2.75, including a $60 million impairment of a joint-venture investment in the first quarter[3]. By segment, second-quarter EBITDA was $659 million for NGLs, $627 million for Refined Products and Crude, $546 million for gathering and processing and $297 million for Natural Gas Pipelines; pipelines posted the largest year-over-year gain, while NGLs declined[8].

First-half cash flow and the balance sheet still reflect the period before the transactions. First-half operating cash flow was $2.987 billion, capital expenditures were $1.477 billion, contributions to affiliates were $353 million and dividends were $1.348 billion; the quarterly dividend was $1.07 per share, or $4.28 annualized, up 4% year over year[25]. At June 30, cash was $161 million and total debt was $33.022 billion, made up of $750 million of current maturities, $1.499 billion of short-term borrowings and $30.773 billion of long-term debt[23], so the effect of debt reduction will first appear in the third-quarter report.

Operating Model

ONEOK's revenue moves with commodity prices, but its profit depends mainly on volumes and fees. Revenue equals index-priced commodity sales plus gathering and processing fees, NGL exchange service fees and transportation and storage fees; because commodity purchase costs are largely passed through, the level of revenue says little about the level of profit[14]. In a typical NGL exchange services arrangement, the company buys unfractionated NGLs at the plant tailgate, deducts contractual transportation and fractionation fees and resells the product, and some contracts carry minimum volume commitments[15].

Adjusted EBITDA is the sum of each segment's revenue minus cost of sales and fuel, minus operating costs, plus EBITDA from unconsolidated affiliates. In 2025 operating costs across the four segments totaled $2.956 billion and affiliates contributed $516 million[14]; moving from adjusted EBITDA to net income, the company deducted $1.514 billion of depreciation and amortization, $1.783 billion of interest and $1.028 billion of income taxes that year[21]. In the second quarter of 2026 the corresponding deductions were $434 million of interest, $387 million of depreciation and amortization and $299 million of income taxes[3]; from the third quarter onward, net income attributable to ONEOK must also subtract Apollo's noncontrolling interest, calculated at about 1.75% of its remaining capital per quarter[26].

The four segments run on different drivers and lags. NGL segment profit is roughly raw feed volume times a blended per-barrel fee, minus transportation, fractionation and operating costs; volume changes reach profit in the same quarter, while the ethane share and contract renewal rates decide how much is kept per barrel[27]. The gathering and processing segment carries both volume and price risk, because fee-with-POP contracts let the company keep part of the commodity sales proceeds, and the company plans to hedge about 75% of its monthly equity volumes over time[16]. Firm transportation revenue in Natural Gas Pipelines depends on contracted capacity and reservation rates and changes slowly, while marketing income moves with location spreads in the same quarter[18]; refined products pipeline profit depends on volumes and tariff rates, and crude marketing income moves with price differentials[28].

The key cash flow variables are fee-based earnings, cash taxes and the new minority distributions. Management said US tax legislation will provide about $2.6 billion of cumulative cash tax benefits, deferring meaningful cash income tax payments until 2031[29]. Starting in September 2026, the Class B interest receives 15% of OpCo's quarterly cash flow from operations, which the company can raise to 20%, and distributions above the 7.0% capped return reduce Apollo's capital balance[26]; management expects capital spending to fall to $2 billion to $2.5 billion a year once the current major projects are completed in 2026 and 2027[30].

Industry and Competitive Position

ONEOK's main advantage is a complete NGL chain that runs from the wellhead to the export dock. Its assets extend from gathering and processing in the Rocky Mountain, Mid-Continent and Permian regions to fractionation at Mont Belvieu, and it is building a Texas City liquefied petroleum gas export terminal with 400 thousand barrels per day of design capacity, scheduled for completion in early 2028[31]. Its refined products network covers the central US, and the Denver expansion entered service on August 1, 2026, adding 35 thousand barrels per day of capacity[29].

The company faces competition from several types of rivals, and asset location and cost of capital sit at the center of it. Competitors include other midstream companies, integrated oil companies and producers that own gathering, processing, fractionation and pipeline assets, and the competitive factors include service quality, producer drilling activity, contract fees, proximity of assets to supply areas and markets, petrochemical utilization and feedstock demand, and cost of capital[32]. After the Brazos acquisition, the company's Midland Basin processing capacity will more than double to about 2.3 Bcf per day, and those assets carry about 600,000 dedicated acres under contracts with a weighted average remaining term of more than 12 years[33].

The available material has clear limits for comparing competitive position. The company does not disclose NGL fees by product[4] or the actual Waha-to-Katy spread, so per-barrel EBITDA and pipeline marketing margin in this analysis are proxies calculated from segment data[10]. The material also contains no peer segment margin data, so competitive position can only be judged from asset coverage and contract structure, not from a direct comparison of peer margins.

Core Debates

ONEOK moved record NGL volumes in the second quarter, yet NGL segment earnings fell below the prior year. In the third quarter, will higher volumes turn back into higher earnings, or will a heavier ethane mix and lower fee rates keep absorbing the growth?

This debate decides the quality of growth at ONEOK's largest profit source. The NGL segment contributed 35% of four-segment EBITDA in 2025[14], and the company's long-term growth case rests on filling its existing pipelines and fractionation capacity; if per-barrel profit keeps falling, the same volume growth produces much less EBITDA and weakens the returns on the Medford fractionator rebuild and the Texas City export terminal[31].

Volumes and profit clearly diverged in the second quarter. Raw feed was 1,630 thousand barrels per day, up 7%, including a 15% increase in the Gulf Coast/Permian region[5], yet NGL segment EBITDA was $659 million, $14 million below the prior-year quarter[4]. The 10-Q bridge shows that within exchange services, volumes added $28 million, but transportation and fractionation costs rose $12 million and captured price differentials fell $11 million, while operating costs rose another $18 million; in the first half, lower fee rates and narrower product differentials together cost $71 million[34].

There are two different explanations for this divergence. Management says the cause is more low-fee ethane recovery in the Bakken and Mid-Continent, because long-term contracts charge less for ethane than for propane and heavier products (C3+), and it said NGL volumes in July and August were stronger than expected[9]. The other explanation is that lower renewal rates in the Mid-Continent and Permian are structural: lower average Mid-Continent fee rates already cost exchange services $44 million in 2025[27], and management also acknowledged that some expiring Mid-Continent contracts may roll over at lower current market rates[9].

The decline in per-barrel profit can already be measured. Dividing segment EBITDA by raw feed volume gives about $4.44 per barrel in the second quarter, below about $4.84 a year earlier, about $5.25 in the first quarter of 2026 and about $5.06 in the second half of 2025[8]. NGL segment EBITDA was $1.471 billion in the second half of 2025, or about $736 million per quarter[14]; at second-quarter volumes, every $0.10 drop in per-barrel profit costs about $15 million per quarter.

The transmission runs from upstream production and plant connections to raw feed volume, which is multiplied by the blended per-barrel fee and reduced by transportation, fractionation and operating costs to produce NGL segment EBITDA, which then flows into consolidated EBITDA and net income. The third-quarter report should show whether NGL segment EBITDA returns above $736 million, whether raw feed exceeds 1,630 thousand barrels per day, whether calculated per-barrel EBITDA gets back above $4.80, and which way the fee and differential items move in the exchange services bridge; whether phase one of Medford, adding 100 thousand barrels per day of fractionation capacity, still enters service in the fourth quarter is also worth watching[29]. If volumes keep growing in the third quarter while NGL segment EBITDA stays below $659 million, or if management describes the lower fees as an ongoing contract reset, the view that volume growth is being offset by falling per-barrel profit would hold, and NGL growth would pass through to earnings weakly for a long time.

ONEOK's natural gas pipeline earnings jumped in the first half, mostly from the Waha-to-Katy price spread created by tight Permian takeaway. As new pipelines come online and that spread narrows, where will third-quarter earnings settle?

Pipeline spread income matters for how full-year guidance gets delivered in the second half. The segment's first-half EBITDA was $636 million, $236 million more than a year earlier[10], roughly matching the size of the guidance increases; the company's 2026 adjusted EBITDA guidance midpoint is now $250 million above its original February guidance[30]. If this profit disappears along with the spread, other segments must fill the gap in the second half; if firm transportation revenue has truly moved up, the pipeline business has a higher base as well.

The second-quarter increase came from both marketing income and firm transportation revenue. Second-quarter pipeline EBITDA was $297 million, up $109 million year over year, and the 10-Q attributes the gain to higher optimization and marketing and higher firm transportation revenue[10]. Calculated from the segment table, residue gas sales and other revenue minus cost of sales and fuel rose from $37 million a year earlier to $123 million, after about $145 million in the first quarter; transportation revenue rose from $104 million to $123 million, and contracted transportation capacity was 7,735 thousand dekatherms per day, or 92% subscribed[4].

Management has already signaled that second-half earnings will fall. On the second-quarter call, the company said pipeline segment earnings will decline in the second half as new Permian takeaway capacity enters service and differentials narrow, consistent with full-year guidance assumptions[11]; on the first-quarter call, management had already said Waha spreads were expected to normalize in the second and third quarters[35]. The other possibility is that the spread narrows more slowly than expected and marketing income stays high in the third quarter.

The transmission runs from Permian takeaway constraints to the Waha-to-Katy location spread, which produces marketing margin as the segment buys low and sells high; contracted transportation capacity times reservation rates produces firm transportation revenue; the two together, minus operating costs and plus affiliate EBITDA, make up segment EBITDA. Pipeline segment EBITDA was $461 million in the second half of 2025, or about $230 million per quarter[14], which serves as a reference level for after the spread falls back.

The third-quarter report needs to show four separate signals: whether pipeline segment EBITDA stays above $230 million, whether the marketing margin proxy drops back below $60 million, whether transportation revenue holds at or above $123 million, and how management describes the Waha-to-Katy spread. Separate movements in the marketing margin proxy and transportation revenue can distinguish a temporarily wide spread from a higher base of firm revenue; if segment EBITDA falls below $230 million, the first-half increase was essentially temporary spread income.

ONEOK's refined products pipelines grew volumes 8% in the second quarter, and crude marketing added extra profit. Once crude marketing gains and refinery-outage effects fade in the third quarter, can fee-based refined products shipping hold up this segment's earnings?

This segment's performance decides whether the company's claim that about 90% of earnings are fee-based holds up in the second half. Refined Products and Crude contributed 27% of segment EBITDA in 2025[14], mostly from the refined products network that came with Magellan, and it is one of ONEOK's most stable fee-based businesses[20].

Second-quarter growth included both a fee-based part and a marketing and timing part. Segment EBITDA was $627 million, up $70 million year over year: transportation and storage added $79 million, crude marketing added $48 million and liquids blending subtracted $8 million, while operating costs rose $48 million[28]. Refined products volumes shipped were 1,629 thousand barrels per day, up 8%, while crude oil volumes shipped fell from 1,782 to 1,766 thousand barrels per day[4]; the 10-Q says part of the volume growth came from marketing-related volumes and the timing of refinery outages, and that lower crude volumes reflected exiting lower-margin short-haul movements[28].

Views differ on the nature of the second-quarter growth. One view holds that growth came mainly from refined products demand and higher rates: the 10-Q attributes the transportation and storage increase to higher refined products volumes and rates[28], and the Denver expansion adds 35 thousand barrels per day of system capacity[17]. The other view holds that growth relied mainly on refinery outages and the timing of crude marketing gains, which will fade in the third quarter.

The reference base for this segment is fairly clear. Segment EBITDA was $1.149 billion in the second half of 2025, or about $575 million per quarter, and $492 million in the first quarter of 2026[8]; second-quarter transportation revenue was $452 million, compared with $425 million a year earlier[28]. The transmission runs from refined products demand and refinery utilization to pipeline volumes, which multiplied by tariff rates produce transportation and storage revenue, while crude price differentials drive crude marketing margin; after operating costs such as labor, outside services and property taxes, the result is segment EBITDA.

The third-quarter report should show whether segment EBITDA stays at or above $575 million, whether refined products volumes hold above 1,600 thousand barrels per day, whether transportation revenue stays at or above $452 million, and whether year-over-year operating cost growth narrows. If transportation revenue and volumes hold while marketing weakens, growth is mainly fee-based; if segment EBITDA falls below $575 million, second-quarter growth depended more on timing gains.

ONEOK brought in a $9 billion minority stake from Apollo, using it to retire about $5 billion of debt and buy Brazos Midland. The third-quarter report is the first to show the deal: did debt really fall by about $5 billion, and is Apollo's quarterly share of earnings in line with what the company described?

This transaction changes how ONEOK's profit is divided. Debt reduction lowers interest expense, but Apollo's noncontrolling interest is charged at about 1.75% of its remaining capital each quarter, and the Class B interest also receives 15% of OpCo's quarterly cash flow[26]; the company says the combination accelerates its deleveraging to 3.25 times debt to EBITDA[13], and the third- and fourth-quarter reports are the first data for testing these claims.

The deal terms make both the cost cap and the use of proceeds explicit. On August 30, the company announced it would acquire Brazos Midstream's Midland Basin natural gas gathering and processing assets for $4.425 billion in cash, funded by a $9 billion nonvoting minority equity investment from funds managed by Apollo, with $5 billion used to repay existing debt[13]. The terms cap Apollo's return at a 7.0% IRR for the first nine years, with value above the cap accruing to common shareholders[26]; the acquisition is expected to close in the fourth quarter, so third-quarter earnings exclude Brazos[33].

Closing and debt reduction moved ahead in September. The transaction closed on September 10, with Apollo receiving 900 million Class B units for $9 billion, and all credit rating agencies consider the transaction credit-enhancing[12]; tender offer results announced on September 15 showed about $2.46 billion in principal amount of notes accepted for a total price of no more than $2 billion[36], and the company also plans to repay its $1.2 billion term loan and exercise make-whole calls on certain notes[33]. For comparison, total debt was $33.022 billion at June 30[23], and the August 3 full-year guidance predates the deal and excludes the minority-interest deduction, interest savings and gains or losses on debt extinguishment[5].

Before the deal, the noncontrolling interest had almost no effect on profit. Net income attributable to noncontrolling interests was about $1 million in the second quarter and about $3 million in the first half, second-quarter interest expense was $434 million and diluted EPS was $1.53[3]. Under the method the company disclosed, the third quarter includes only about 21 days after closing, so earnings attributable to Apollo should be on the order of $36 million, compared with about $157.5 million for a full quarter[26].

There are two readings of the deal's net effect. The company's position is that it is cheaper than issuing common stock and allows faster deleveraging while leaving room for buybacks and dividends[13]; the other reading is that the minority-interest deduction and Class B cash distributions will clearly offset interest savings in a full quarter, and that the low third-quarter deduction reflects only 21 days, so the fourth-quarter comparison of the minority-interest deduction with interest expense will better separate the two.

The third-quarter report should show how much total debt at September 30 fell from June 30, whether net income attributable to noncontrolling interests is on the order of $36 million, whether interest expense comes in below the second quarter's $434 million, whether one-time items such as gains or losses on debt extinguishment appear, and whether full-year guidance incorporates the transaction. If debt falls by less than $4 billion, deleveraging is behind plan; if the minority-interest deduction is clearly higher than the disclosed method implies, the deal's accretion to earnings per share needs to be recalculated.

Risks and Falsifiers

Slower upstream activity and lower commodity prices would pressure both volumes and prices in the gathering and processing and NGL segments. Both segments carry volume risk from drilling and completion activity and changes in production[19], and fee-with-POP contracts in gathering and processing also leave the company with part of the commodity sales proceeds, with plans to hedge about 75% of monthly equity volumes over time[16]. Lower realized NGL and natural gas prices already reduced gathering and processing EBITDA by $53 million in the first half; the segment earned $2.138 billion of EBITDA in 2025 and $1.013 billion in the first half of 2026[37]. If third-quarter natural gas processed is no lower than the second quarter's 5,707 million cubic feet per day and segment EBITDA is flat or higher year over year, this risk has not materialized[4].

Faster capital spending and the integration of new assets could delay cash flow. Management expects capital spending to accelerate in the second half, pushing the full year toward the upper end of the $2.7 billion to $3.2 billion range[30], while the Brazos acquisition will be consolidated in the fourth quarter and the Medford, Bighorn and Texas City projects move ahead at the same time[31]. First-half operating cash flow of $2.987 billion left only about $162 million after $1.477 billion of capital expenditures and $1.348 billion of dividends, and the $353 million of affiliate contributions turned that into a shortfall[25]; if the third-quarter report keeps capital guidance unchanged, Brazos closes as planned and major project schedules hold, this risk has not appeared.

A rising ethane share and contract rate resets would reduce how much volume growth turns into profit. Long-term contracts charge less for ethane than for C3+, and some expiring Mid-Continent contracts may renew at lower market rates[9]; NGL segment EBITDA was $2.779 billion in 2025, the second quarter implies about $4.44 per barrel, and each $0.10 per barrel decline at 1,630 thousand barrels per day costs about $15 million per quarter[34]. If third-quarter NGL segment EBITDA is no lower than $736 million and calculated per-barrel profit returns above $4.80, this risk is falsified.

A fading spread could make pipeline segment earnings fall back quickly. First-half pipeline segment EBITDA was $236 million higher than a year earlier, mostly from the Waha-to-Katy spread, and the marketing margin proxy was $123 million in the second quarter[10], while new Permian takeaway pipelines will narrow the spread once in service[11]. If third-quarter pipeline segment EBITDA exceeds $230 million and transportation revenue is no lower than $123 million, this risk is falsified.

Falling marketing and timing gains combined with rising costs would depress Refined Products and Crude earnings. Crude marketing added $48 million in the second quarter, volumes were helped in part by refinery outages, and operating costs rose $48 million[28]; if the first two fade while costs stay, segment profit would drop below its second-half 2025 level. If third-quarter segment EBITDA is no lower than the second quarter's $627 million and transportation revenue is no lower than $452 million, this risk is falsified.

Apollo's noncontrolling interest could weigh on earnings attributable to ONEOK for a long time. Apollo's interest is charged at about 1.75% of its remaining capital each quarter and shares in 15% of the cash flow from OpCo's existing businesses[26]; on $9 billion, a full quarter's deduction is about $157.5 million, and if overall cash flow cannot quickly reduce that capital balance, the cost would depress net income attributable to ONEOK and diluted EPS for a long time[12]. If, from the fourth quarter onward, interest savings plus the Brazos contribution exceed the minority-interest deduction and the Class B capital balance declines every quarter, this risk is falsified.

What to Watch Next

• NGL volume versus per-barrel profit: second-quarter NGL segment EBITDA was $659 million on 1,630 thousand barrels per day, or about $4.44 per barrel. EBITDA back above $736 million with per-barrel profit above $4.80 would confirm a recovery; higher volumes with EBITDA still below $659 million would falsify it. Also watch the fee and differential items in the bridge and whether Medford phase one enters service in the fourth quarter.
• Pipeline spread normalization: second-quarter segment EBITDA, the marketing margin proxy and transportation revenue were $297 million, $123 million and $123 million. EBITDA above $230 million with transportation revenue of at least $123 million would point to a higher base; EBITDA below $230 million would show the gain was a temporary spread.
• Refined products fees versus crude marketing: second-quarter segment EBITDA was $627 million, refined products volumes were 1,629 thousand barrels per day and transportation revenue was $452 million. EBITDA of at least $575 million with volumes above 1,600 thousand barrels per day would confirm fee-based strength; EBITDA below $575 million would show reliance on timing gains.
• Apollo minority stake and debt reduction: total debt was $33.022 billion at June 30, and second-quarter minority-interest earnings and interest expense were about $1 million and $434 million. A debt reduction of about $5 billion with a deduction of about $36 million would confirm the company's description; a reduction below $4 billion or a clearly higher deduction would falsify it. Also watch whether full-year guidance incorporates the deal.

Conclusion

ONEOK's profit is driven by volumes and fees, with about 90% coming from fee-based businesses[20], and second-quarter 2026 adjusted EBITDA was $2.121 billion, up 7% year over year[3]. The structure of that growth is changing, however: the largest segment, NGLs, set a volume record while its profit fell, a sizable part of first-half extra profit came from a pipeline spread that is now narrowing, and the Apollo deal traded a minority-interest deduction of about 1.75% per quarter for roughly $5 billion of debt reduction[26]. The central unresolved relationship is whether volume growth can still turn into higher earnings attributable to ONEOK after the spread fades and the minority deduction is taken.

Only one independent outside commentary published after the second-quarter results can be cited, and it addresses the Apollo deal. Writing for Insider Monkey on Yahoo Finance, Jeff Lewis argued that Apollo's 7.0% capped return, while below what ONEOK describes as its cost of public equity, works out to about $630 million a year on the initial $9 billion balance, and that actual cash distributions may be higher; because the Class B interest shares in all of ONEOK's existing operations rather than only the acquired assets, the capped return is attractive only if cash generation across the broader portfolio rapidly reduces the minority capital balance[38]. This view does not directly conflict with the company's claim that the structure is cheaper than issuing common stock, but it reminds readers that avoiding a common stock issue is not free, and it bears on the fourth core debate; beyond it, outside coverage after the second-quarter report consisted mainly of deal news and sell-side rating changes, with no citable independent analysis of the NGL per-barrel and pipeline spread debates.

If later reports show raw feed still growing with calculated per-barrel EBITDA back above $4.80, pipeline segment EBITDA holding above $230 million, debt about $5 billion lower than at the end of June and a minority-interest deduction consistent with the disclosed method, the view that volume growth can still support earnings growth would be materially strengthened. Conversely, if NGL segment EBITDA stays below $659 million, pipeline earnings fall below $230 million, Refined Products and Crude drops below $575 million and interest savings fail to offset the minority deduction, the view that growth depends more on temporary spreads and marketing gains would prevail.

Sources

[1] OKE 10-K filed 2026-02-24 · four reportable segments and customer concentration · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[2] Drillr earnings calendar (updated 2026-10-16) · OKE 2026-10-27 call · 2026-10-16 · Drillr earnings calendar

[3] OKE 8-K filed 2026-08-03 · 2Q26 financial highlights and EBITDA reconciliation · 2026-08-03 · 8-K · https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001039684&type=8-K&dateb=&owner=include&count=40

[4] OKE 8-K filed 2026-08-03 · operating statistics 2Q26 · 2026-08-03 · 8-K · https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001039684&type=8-K&dateb=&owner=include&count=40

[5] OKE 8-K filed 2026-08-03 · 2Q26 results and raised 2026 guidance · 2026-08-03 · 8-K · https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001039684&type=8-K&dateb=&owner=include&count=40

[6] Drillr analyst_financial_estimates (updated 2026-10-16) · OKE quarter ending 2026-09-30 · 2026-10-16 · Drillr analyst_financial_estimates · https://gateway.drillr.ai/mcp/private

[7] Drillr analyst_financial_estimates (updated 2026-10-16) · OKE FY2026 · 2026-10-16 · Drillr analyst_financial_estimates · https://gateway.drillr.ai/mcp/private

[8] OKE 10-Q filed 2026-08-04 · segment information 2Q26 and 1H26 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[9] OKE Q2 2026 earnings call 2026-08-04 · NGL margin mix and stated risks · 2026-08-04 · earnings-call · https://gateway.drillr.ai/mcp/private

[10] OKE 10-Q filed 2026-08-04 · Natural Gas Pipelines results 2Q26 and 1H26 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[11] OKE Q2 2026 earnings call 2026-08-04 · segment performance · 2026-08-04 · earnings-call · https://gateway.drillr.ai/mcp/private

[12] OKE 8-K12B filed 2026-09-10 · Apollo investment closing and reorganization · 2026-09-10 · 8-K · https://www.sec.gov/Archives/edgar/data/0001039684/000119312526387972/d81803dex991.htm

[13] OKE 8-K filed 2026-08-31 · Brazos Midland acquisition and Apollo minority investment · 2026-08-31 · 8-K · https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001039684&type=8-K&dateb=&owner=include&count=40

[14] OKE 10-K filed 2026-02-24 · segment financial information 2025 · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[15] OKE 10-K filed 2026-02-24 · NGL segment assets and sources of earnings · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[16] OKE 10-K filed 2026-02-24 · G&P contracts, capacity and commodity hedging · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[17] OKE 10-K filed 2026-02-24 · Refined Products and Crude 2025 and Denver expansion · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[18] OKE 10-K filed 2026-02-24 · natural gas pipelines contracts and marketing · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[19] OKE 10-K filed 2026-02-24 · fee-based earnings and volumetric risk · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[20] OKE 10-Q filed 2026-08-04 · business update and fee-based mix · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[21] OKE 10-K filed 2026-02-24 · net income to adjusted EBITDA 2023-2025 · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[22] OKE 10-K filed 2026-02-24 · consolidated cash flows 2023-2025 · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[23] OKE 10-Q filed 2026-08-04 · balance sheet, debt and liquidity at 30 June 2026 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[24] OKE 10-K filed 2026-02-24 · credit ratings and covenant leverage · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[25] OKE 10-Q filed 2026-08-04 · cash flows and dividends 1H26 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[26] OKE 8-K filed 2026-08-31 · minority equity investment terms and NCI accounting · 2026-08-31 · 8-K · https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001039684&type=8-K&dateb=&owner=include&count=40

[27] OKE 10-K filed 2026-02-24 · NGL segment 2025 results and ethane economics · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[28] OKE 10-Q filed 2026-08-04 · Refined Products and Crude results 2Q26 and 1H26 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[29] OKE Q2 2026 earnings call 2026-08-04 · project updates and cash taxes · 2026-08-04 · earnings-call · https://gateway.drillr.ai/mcp/private

[30] OKE Q2 2026 earnings call 2026-08-04 · guidance and capital spending · 2026-08-04 · earnings-call · https://gateway.drillr.ai/mcp/private

[31] OKE 10-Q filed 2026-08-04 · capital projects · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[32] OKE 10-K filed 2026-02-24 · competition · 2026-02-24 · 10-K · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000006/

[33] OKE 8-K filed 2026-08-31 · Brazos Midland assets, multiple and timing · 2026-08-31 · 8-K · https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001039684&type=8-K&dateb=&owner=include&count=40

[34] OKE 10-Q filed 2026-08-04 · NGL segment results 2Q26 and 1H26 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[35] OKE Q1 2026 earnings call 2026-04-29 · guidance and seasonality · 2026-04-29 · earnings-call · https://gateway.drillr.ai/mcp/private

[36] OKE 8-K filed 2026-09-15 · tender offer early results and pricing · 2026-09-15 · 8-K · https://www.sec.gov/Archives/edgar/data/0001039684/000119312526392003/d162704dex991.htm

[37] OKE 10-Q filed 2026-08-04 · G&P results 2Q26 and 1H26 · 2026-08-04 · 10-Q · https://www.sec.gov/Archives/edgar/data/1039684/000103968426000029/

[38] Yahoo Finance (Insider Monkey) 2026-09-04 · ONEOK (OKE) is Funding a $4.4B Acquisition With a $9B Minority Investment. Is the 7% Capped Return Attractive? · 2026-09-04 · Yahoo Finance(转载 Insider Monkey) · https://finance.yahoo.com/markets/stocks/articles/oneok-oke-funding-4-4b-150349706.html