Pipelines / midstream (new) Constructive — favor gas ▲

Salzman (Barron's, Dec 2): a record North-American pipeline build-out — ~$53B of growth capex this year (past the $49B 2019 peak), driven by Gulf-Coast LNG exports and natural-gas pipelines feeding power plants for new data centers (Kinder Morgan projects US gas demand +28 Bcf/d, >25% above 2024, by 2030; Energy Transfer is building gas lines to three Oracle data centers, two in Texas). Forward adjusted-earnings growth is being marked up to 6–8% (from 4–6%). Yet the stocks lag badly — the Tortoise North American Pipeline Fund is +4% vs the S&P's +16% — on weak oil (−15% YTD) and the memory of the 2015 bust, when even Kinder Morgan cut its dividend. Westwood's Parag Sanghani: bigger cash-flow cushions than 2015 de-risk the dividends, so favor natural-gas pipelines (Williams +10%, Kinder Morgan) over oil/liquids transporters (Energy Transfer −15%, Oneok); Tortoise's Rob Thummel pins the drag on oil. "A great time in general to be allocated to the space." Lundberg (Baytex CEO, In the Money Jul 14): the Canadian leg has genuinely turned — two pipeline proposals across all layers of government in one week ("bonafide pipeline being talked about, start dates that are concrete") and the carbon-tax cap pushed from a $170/ton-by-2030 peak out to 2040 with a lower cap, are finally defining the "rules of engagement" — an operator says previously-shelved projects (small-scale SAGD) are "absolutely on the table now." Egress today is "reasonably comfortable," but the open question for the growth plans is where the excess takeaway capacity comes from; foreign-investor conversations about Canadian energy have flipped constructive. 2026-AUG-06/10 — CNBC Halftime: Terranova's JOET energy sleeve is midstream-heavy (EPD, KMI, OKE, TRGP, WMB, LNG); Baruch runs the same shape (XOM, KMI, LNG) — "the demand in the United States is going to be needed for power and AI." Amoroso: with US production bumped up and LNG exports growing, "the pipelines, parts of the energy ecosystem can continue to do well" even if the Strait reopens — "you can't power data centers without energy… for all of that you need the molecule." 2026-AUG-11 — Hay (Haymaker Daily): the Permian takeaway build-out sized — "America's vibrant mid-stream energy industry has been aggressively constructing added takeaway capacity out of the Permian… multiple new pipelines under construction" totalling ~15 Bcf/d, about 12½% of aggregate U.S. marketed gas output. Until they open, Permian associated gas often trades negative and is flared; the completions transfer value to producers with significant Permian gas (EOG named) — the rare catalyst with a construction schedule rather than a re-rating. 2026-AUG-18 — Adam Waterous (In the Money w/ Amber Kanwar): the public-vs-private cost math on the Canadian West Coast line, read as a policy tell. Private benchmark = South Bow's Prairie Connector: ~$15B for 550,000 bbl/d (expandable to a million), a ~$9/bbl toll, an implied ~12% return. A West Coast barrel earns ~$2 more, supporting ~20% higher tolls, so a private builder could afford ~$18B — against an announced $36–43B. So "the federal government effectively is covering the extra 25 billion," plus $10B for the Vancouver port, at roughly a 5% rate of return vs a private 12%. Cause: industry's reform letter (scrap C-69, C-48, the industrial carbon tax, six-month approvals) was declined, so "by definition it'll be the public sector" — Trans Mountain as crown-corp proponent with Pembina "assisting" but non-committal ("we could walk away at any time"). His frame: "circling back 50 years" to Petro-Canada. Halftime panel (Aug 21): Harrington's explicit rotation within energy — out of the oil-price majors, into ET / EPD / KMI / MPLX: "up 25–30% on the year but they don't have the exposure… these guys produce significant cash here and now today, and that's valuable." (CNBC Halftime panel, 2026-aug-21.) 2026-SEP-08 (David Hay) — a mark-to-market on the midstream call, and a trim argued on tax rather than on the business. MPLX, flagged in Feb-2024 on a “Magnificent Double Nine” (a bit over 9× earnings and a 9% distribution yield) plus an upside range expansion through an eight-year ceiling, has returned 93.7% with distributions against 61% for the S&P 500 and 95.7% for the Magnificent Seven — “particularly adjusting for its high cash flow/low risk attributes” — and still sits below its 2015 peak of $78 though EPS/unit “more than tripled.” Forward: “the easy answer is: a lot less,” so “trimming a bit… might now make sense.” But the trim rests on tax (MLP-in-an-IRA taxable income; recapture of the 60–70% of distributions that were sheltered; the stepped-up basis at death), not on the asset: “it continues to yield over 7% and it has raised the payout by 9% a year… better than the 7% we projected.” Arjun Murti (2026-SEP-12): the best gas exposure has been midstream/downstream — LNG and pipelines — rather than shale producers; select midstream companies (with oil services) are extending into distributed power generation. Parag Sanghani (Westwood, 2026-Sep-17): gas pipelines (Williams) are the transportation layer of the AI power build-out and — with fuel producers — the slower-growth but highest-income part of the chain; tens of thousands of miles of aging US power lines also need replacing.