Kinder Morgan Inc KMI

Thesis

Kinder Morgan (KMI) is the largest natural-gas transmission system in North America: roughly 40% of U.S. gas moves on its pipes, plus terminals, CO2/EOR, and the Elba LNG facility. The economics are toll-bridge: ~64% of EBITDA from take-or-pay capacity contracts, ~26% from fee-for-service, weighted-average contract life ~7 years, blue-chip utility/LDC/producer counterparties.

The scorecard tells the right story when you read it carefully. Headline ROIC is just 3.31% (10-yr avg), and EV/FCF of 36.8x looks expensive — both reflect a heavy build-out era plus reported D&A that overstates true economic depreciation on long-lived steel-in-the-ground. What matters more: owner earnings TTM of $3.34B, FCF-conversion 10.5x net income (the GAAP P/E of 27.97 is meaningless against a real cash yield ~9% on EV), share count flat over a decade (-0.07%), and management's own reverse-DCF implies just 3.75% growth required to justify today's $32.53 — laughable against a $9-10B contracted backlog and 8-10% project ROIs.

IV base $72.80 vs. price $32.53 = px/IV 0.45. Even the conservative IV low ($35.64) sits above today's price. The bear case (rate compression, regulatory hostility, gas demand peak) doesn't get you below $25-28 without imagining away the contracted backlog. The bull case (LNG export tripling by 2030, AI-driven gas-fired power additions, FERC-blessed rate-base growth) gets you to $80+. The asymmetry is the trade. Buy under $35, trim above $73.

Moat

Kinder Morgan's moat is a textbook cost-advantage / regulatory-intangible hybrid — the same shape Buffett describes when he writes about BHE pipelines and BNSF rail [1][2][6]. Five-moat audit:

1. Cost advantages — WIDE. A natural-gas pipeline is the cheapest, safest, lowest-emission way to move methane in volume. Trucks, rail, and LNG-by-truck are not substitutes at scale. Buffett's framing on BNSF — "a ton of freight 500 miles on a single gallon" [1][2] — applies even more strongly here: a pipeline moves gas at a fraction of a cent per MMBtu-mile with near-zero variable cost once built. KMI's 79,000-mile network was assembled over 80+ years from Tennessee Gas, El Paso, and Copano roots. Replacement cost is the moat. A new interstate gas pipe in the U.S. today costs $5-8M per mile, requires 5-8 years of FERC permitting, hundreds of landowner easements, and faces near-certain litigation. Williams' Constitution Pipeline died after a decade of permitting. Mountain Valley took 9 years and $7B for 303 miles. Nobody is replicating KMI's footprint with $10B and 5 years — the Munger stress test fails on its face.

2. Regulatory intangibles — WIDE for interstate, NARROW for intrastate. ~70% of KMI's gas-segment EBITDA flows through FERC-regulated interstate pipes. FERC sets rates on a cost-of-service basis with allowed returns on equity (typically 10-12%) and grandfathered negotiated rates. Once a pipe is in rate base, the regulatory bargain Buffett describes [4][6] applies: "society will forever need huge investments in… energy. It is in the self-interest of governments to treat capital providers in a manner that will ensure the continued flow of funds." KMI lives inside that bargain. The intangible is the certificate of public convenience and necessity (CPCN) — a license that competitors can't get and incumbents can't lose absent gross misconduct.

3. Switching costs — NARROW but real. A power plant, LDC, or LNG facility is physically connected to a specific pipeline interconnect. Switching means building a lateral (months and millions) or accepting basis-blowout risk. Counterparties sign 10-20 year take-or-pay contracts because the alternative is unreliable supply. This is why ~64% of EBITDA is capacity-reservation revenue independent of throughput — the customer pays whether they use the pipe or not. Switching costs are narrower than, say, an enterprise SaaS lock-in (a customer can re-contract at expiry), but they create the contract stickiness that makes cash flows utility-like.

4. Network effects — NARROW. More interconnects = more optionality for shippers, which makes the system more valuable. KMI's network connects every major producing basin (Permian, Haynesville, Marcellus, Eagle Ford) to every major demand center (Gulf LNG, Florida power, Northeast LDCs). New basins or new LNG plants prefer to connect to the largest grid. This is a real but secondary effect.

5. Pricing power — NARROW. FERC caps the upside (allowed ROE), but the floor is solid: rates index to cost of service, including capex. In a high-capex era (now), tariffs rise. KMI cannot price like a luxury good, but it cannot be undercut either — the regulator protects the incumbent's economics in exchange for service obligations [4][6].

Competitor stress test ($10B + 5 years): Williams (WMB), Energy Transfer (ET), Enterprise (EPD), and TC Energy already have $10B+ each. None has dislodged KMI from its core corridors in 20 years. The relevant competitor is not another pipeline — it's electrification of heating, renewables displacing gas-fired power, or a regulator that turns hostile. Those are macro/political risks, not competitive ones (handled in inversion).

Erosion risk: The biggest erosion vector is demand (gas-fired generation share, end-state heating mix), not competition. Once a pipe is permitted and built, it is essentially a 50-year asset with no economic substitute on its corridor. The 10-yr ROIC of 3.31% understates the moat's quality because it's depressed by (a) a decade of rate-base build-out where new capex sits in CWIP earning nothing, and (b) GAAP D&A on steel pipe with 70-100 year economic lives — Buffett's repeated point that EBITDA-style metrics flatter capital-light businesses and hurt capital-heavy regulated ones is exactly inverted here.

Moat verdict: WIDE.

Management & Capital Allocation

Rich Kinder founded KMI in 1997 out of Enron's pipeline carcass and still owns 10% of the company ($7B). He took $1/year in salary for two decades. CEO Kim Dang (since 2024, prior CFO/COO since 2002) is a Kinder protégée. The board is unusually owner-aligned for a midstream — insider ownership ~13%, the highest among large-cap midstreams.

Capital allocation, the five choices:

1. Reinvest: Project backlog is $9-10B as of Q1 2026, up from $3B two years ago. Average expected unlevered project IRR ~16-20% per management; allowed FERC returns 10-12%. Build mix is heavily weighted to South System Expansion 4 (Southeast power demand), Permian-to-Katy expansions (LNG feed), and Trident (Gulf Coast LNG corridor). This is the highest-quality reinvestment vintage since 2014. The 3.31% 10-yr ROIC is a backward-looking artifact of the 2015-2020 dividend-cut/deleveraging era when growth capex was suppressed; the forward incremental return on the current $9-10B is where the value creation lives, and it is structurally above cost of capital.

2. Acquire: KMI overpaid at the top of the cycle for Copano (2013) and El Paso (2012) — these acquisitions plus 2015 Kinder Morgan Canada drove the leverage spike that forced the 2015 75% dividend cut. Management has been disciplined since: 2024 STX Midstream ($1.8B) and 2025 Outrigger Energy ($640M) were tuck-ins at single-digit EBITDA multiples on contracted assets. Grade improvement clear. They have not chased Permian-takeaway M&A at peak multiples.

3. Debt: Net debt/EBITDA 4.79x (per scorecard) — stable in the 4.5-5.0x band management targets, investment-grade BBB-stable from S&P/Moody's. This is high in absolute terms but appropriate for contracted utility-like cash flows; Buffett explicitly endorses this structure for BHE pipelines [3][6]: "huge investment in very long-lived, regulated assets, with these partially funded by large amounts of long-term debt." Interest coverage isn't shown in the scorecard (null), but FFO/interest is ~5-6x — adequate, not luxurious. The 2015 lesson — leverage was 5.6x and the dividend was 7%-yielding — has been internalized.

4. Buybacks: Authorized $3B program; bought back ~$700M in 2024 and ~$200M in 2025 at avg ~$18-21/share. Share count change over 10 years: -0.07% (essentially flat). KMI is not a buyback machine — it issues equity to fund growth and bought back only when shares were below $20. P/IV at those repurchase prices was ~0.25-0.30, so the buybacks are accretive. They have not repurchased aggressively at $32+, which is appropriate (P/IV 0.45 is still cheap but the equity is also useful funding for the backlog). Discipline grade: B+.

5. Dividends: Q1 2026 dividend $0.2925, raised to $0.2975 in April 2026 — annualized $1.19, ~3.7% yield at $32.53. Payout ~55% of distributable cash flow, leaving meaningful retained cash for the backlog. The 2015 cut from $2.04 to $0.50 was the company's character moment; since then, eight straight years of small raises with no covenant stress. Management explicitly prioritizes the balance sheet over dividend growth — Buffett's framing of MidAmerican retaining all earnings to fund renewables [1][4] is the philosophical cousin.

Communication quality: Quarterly investor presentations are dense, project-by-project, with explicit ROI and EBITDA-build math. Rich Kinder's annual letter is written in plain English with no IR-speak. Disclosures on contract life, basin exposure, and project economics are best-in-class for the sector. They guide DCF and EBITDA, not adjusted EPS gymnastics.

Concerns: (a) 2015 was a real failure — they over-levered into a commodity downturn and cut the distribution 75%. (b) C-corp governance is fine but historical MLP-roll-up complexity left some scar tissue with retail investors. (c) Compensation tied partly to DCF/share, which can incentivize debt-funded growth — but the leverage target acts as a governor.

Capital allocator: B+. Rich Kinder's skin in the game and the post-2015 discipline outweigh the historical M&A overpays. Not Buffett-tier (that's an A), but well above midstream median.

Industry Structure

Porter's Five Forces on U.S. midstream natural-gas transmission:

1. Threat of new entrants — VERY LOW. A new interstate gas pipeline requires a FERC Certificate of Public Convenience and Necessity, state-level permits in every crossed jurisdiction, federal NEPA review, hundreds-to-thousands of landowner easements, and survival of inevitable Sierra Club / state AG litigation. Mountain Valley took 9 years and $7B for 303 miles; Constitution Pipeline died after a decade. The Inflation Reduction Act and bipartisan permitting reform proposals have not materially shortened timelines. Capital alone cannot enter this market — you also need political license, which incumbents accumulate over decades. Buffett's regulatory bargain framing [4][6] applies fully: incumbents are protected in exchange for service obligations. New entrant threat is essentially zero on existing corridors.

2. Bargaining power of suppliers — LOW. Suppliers to KMI are steel-pipe manufacturers, compressor-station EPCs, and labor. All are commoditized or sourced from competitive pools. Inflation passed through to customers via FERC rate cases (with a lag). No critical input has scarcity pricing power.

3. Bargaining power of buyers — MODERATE. Buyers are LNG exporters (Cheniere, Venture Global, NextDecade), gas-fired power generators (utilities, IPPs), local distribution companies (Atmos, Sempra LDCs), and producers needing takeaway. Large LNG facilities can in theory negotiate hard on new contracts, but they need physical interconnects to specific basins (Permian, Haynesville) and KMI's pipes are often the only or cheapest route. Existing 10-20 year take-or-pay contracts neutralize buyer power for the contract life. New incremental capacity is competitive — Williams, Energy Transfer, Enterprise all bid for LNG feed contracts — but the market is growing fast enough that all credible incumbents win share.

4. Threat of substitutes — MODERATE and rising long-term. Substitutes for natural gas in end-use: (a) electrification of heating (heat pumps), shrinking residential/commercial gas LDC volumes 1-2% per year in mature regions; (b) renewables + battery displacing gas-fired generation — but this is partially offset by gas-fired generation rising as baseload to firm intermittent renewables; (c) for LNG export, the substitute is Qatari, Australian, or Russian LNG — U.S. is structurally cost-advantaged on Henry Hub feedstock. Direct substitute for the pipeline itself is virtually nil; the substitute risk is for gas as a fuel, on a 20-30 year horizon.

5. Rivalry among existing competitors — LOW to MODERATE. The U.S. midstream gas-transmission market is an oligopoly: KMI, Williams (WMB), Energy Transfer (ET), Enterprise (EPD), TC Energy (TRP). Each has corridor strongholds — Williams in Transco/Northeast-to-Southeast, KMI in Texas/Southeast/Rockies, Energy Transfer in Permian/Mid-Con. Direct overbuild is rare because FERC won't approve duplicate capacity that fragments rate-base economics. Pricing on contract renewal is competitive but rational; capacity additions are demand-pulled, not speculative. The 2014-2020 overbuild period cured itself via project cancellations and bankruptcies.

Value pool location and trajectory: The midstream value pool is migrating toward the Gulf Coast (LNG) and Southeast (data-center / AI power demand) and away from the Northeast (declining LDC volumes, anti-pipe politics in NY/NJ). KMI's footprint is favorably positioned: Texas Intrastate, Tennessee Gas to the Southeast, and EPNG/Sierrita to Mexico are all in the right basins. The economics are also shifting from low-single-digit growth + 7% yield (yieldco era 2015-2022) to mid-single-digit EBITDA growth + 4% yield + buybacks (utility-growth era 2024-onward). LNG export capacity is on track to roughly double by 2028 and triple by 2030 vs. 2024 — directly drives KMI feed-gas demand. AI/data-center load growth is adding ~25-50 GW of behind-the-meter and grid gas-fired demand by 2030 per multiple grid-operator IRPs.

Industry Verdict: Good. Not Excellent because (a) terminal-value risk from electrification is real on a 30-year horizon, (b) regulatory bargain depends on political stability that is not guaranteed, and (c) ROEs are capped by FERC. But the structural moat, growing value pool, and oligopolistic discipline place this firmly above Average.

Inversion (Bear Case)

I am now playing the short-seller. I will not hedge.

1. The single event that kills this. A serious methane-leak incident on a major KMI line — a Pacific Gas-style San Bruno or a Colonial Pipeline-class cyber/operational failure — combined with a Democratic trifecta in 2028 produces a regulatory regime change. FERC issues a moratorium on new gas-pipeline certificates and tightens existing-pipeline integrity requirements. EPA finalizes methane fee rules at the high end. State AGs in NY, NJ, MA, IL, CA file coordinated suits to deny pipeline reauthorizations. KMI's terminal-value assumption — that pipes have 50-year economic lives because they get re-certified — breaks. Suddenly the duration of the cash flow is 10-15 years, not 50. The DCF compresses brutally. This is not science fiction: PG&E went through it in 2010-2018, the Atlantic Coast Pipeline died in 2020, and the Biden EPA already proposed methane rules KMI lobbied against.

2. Why the moat is narrower than bulls think. The bulls invoke "79,000 miles, irreplaceable" — but moats erode from the customer side, not the competitor side. KMI's customers are utilities and power generators whose own customers are state PUCs and municipal voters. When New York bans new gas hookups (already done), Massachusetts mandates building electrification (in progress), and California shifts heating subsidies to heat pumps, the LDCs feeding KMI's pipes shrink their gas books. Throughput holds for 10 years on take-or-pay contracts, then falls off a cliff at recontracting. The LNG offset is real but concentrated in 6-8 terminal customers — a single failed terminal (Rio Grande LNG, Driftwood) wipes out a billion in contract value. The "irreplaceable" framing confuses can't be rebuilt with can't be made obsolete. Whale-oil tankers were also irreplaceable.

3. Why management is worse than it appears. Rich Kinder is 81 years old. Kim Dang is competent but hasn't been tested in a downturn as CEO. The 2015 dividend cut wasn't a one-time failure — it was the consequence of a debt-funded acquisition spree (Copano, El Paso, KMC) at peak multiples that management defended for two years before capitulating. The same cultural pattern — "build the empire now, justify it later" — is detectable in the current $9-10B backlog: management is capex-led during a period of secular demand uncertainty. The 16-20% project IRRs cited in investor decks assume LNG plants ship full and gas-fired plants run baseload through 2050 — both contestable. Net debt/EBITDA at 4.79x is not conservative for a business with 10-15 year recontracting risk; it's late-cycle leverage. Bondholders get paid before equity, and they are the ones who set the terms of the next refinancing wave (~$10B of debt maturing 2026-2028 into a higher-rate environment).

4. What bulls are extrapolating that won't hold. Bulls extrapolate that LNG export capacity triples by 2030 and AI power demand adds 50+ GW of gas-fired generation. Both are forecasts, not contracts. (a) LNG: Global LNG supply is in massive overbuild — Qatar adding 64 mtpa, U.S. adding ~80 mtpa, Australia/Africa modest growth. By 2028 the market is structurally oversupplied; Henry Hub-to-JKM spreads compress; marginal U.S. terminals get curtailed; KMI's new feed contracts get re-priced lower or canceled (Tellurian/Driftwood already showed this). (b) AI power: Hyperscalers are simultaneously contracting nuclear (Three Mile Island, SMR deals), geothermal, and behind-the-meter solar+storage. Gas peakers fill the gap, but baseload data-center gas demand is contested. Bulls also extrapolate that FERC stays accommodative; the agency is split 2-2 with one open seat, and the next administration's appointees set the regime for a decade.

5. Valuation trap (multiple compression / regime change). The current EV/FCF of 36.8x and P/E of 27.97 already embed a recovery narrative. If terminal growth resets from 4% to 1.5% (electrification + LNG oversupply), the IV recompiles to ~$22. If WACC rises 100 bps on credit-spread widening into a recession, IV drops another ~15%. If the dividend is cut a second time to defend the balance sheet during a refinancing wave, the equity loses its income-investor base — déjà vu of 2015 when KMI fell from $44 to $11. The bull P/IV of 0.45 is not a margin of safety; it is the market correctly discounting tail risks the bulls dismiss as paranoid.

If I am right, the stock could be worth $18 within 3 years.

Lollapalooza Bias Check

Biases active in me as the analyst right now:

Anchoring — strongly active. I am anchored on the scorecard's IV base of $72.80. That number is the output of a model with a base CAGR clamped at 14.0% (per scorer notes) and "maintenance capex uncertain (>50% spread)." The clamp itself signals the model wanted a higher growth number than the analyst was comfortable with — meaning IV could be over- or under-stated. A different reasonable maintenance-capex assumption could move IV by 30%. I should hold the IV range as a range, not a point.

Authority — moderately active. Buffett owns gas pipelines (BHE) and writes admiringly about the regulated-utility model [1][3][4][6]. Munger praised Kinder Morgan-style toll-bridge businesses. I am borrowing conviction from their endorsement of a different asset (BHE is regulated retail electric utilities, not interstate gas transmission) and applying it to KMI. The structural similarities are real but not identical — BHE has captive ratepayers; KMI has contracted shippers with renewal risk.

Confirmation — strongly active. I notice I am building the bull case more easily than the bear case. The inversion section forced me to confront this; without that discipline, I would have under-weighted the recontracting risk and over-weighted the LNG/AI tailwind. Re-reading my own bear case, the demand-side arguments are not absurd — they are mainstream views among ESG-aligned analysts. I should not dismiss them as buyer-strike noise.

Contrarian/social-proof inversion — active. Energy infrastructure is unloved by ESG capital. There is a mild social-proof reward in the value-investor community for owning what indexers can't, which can substitute for genuine analytical edge. "Hated stock = good idea" is a fallacy when the hatred is fundamental, not merely fashionable. I need to verify the hatred is the second kind, not the first.

Recency — moderately active. The Trump 2024 win and pro-energy policy stance is fresh; I may be over-weighting near-term regulatory tailwinds and under-weighting that the political cycle reverses. KMI is a 20-year holding, not a 4-year one — the 4-year political tailwind is noise inside a two-cycle horizon.

Deprival super-reaction — mildly active. Knowing the stock is at $32.53 and the IV base is $72.80, I feel the pull of "if I don't act I'll miss the rerating." That feeling is the deprival reflex — it is exactly the wrong input to a sizing decision. Position sizing should come from conviction × downside, not from FOMO.

Not active or low: Commitment (no prior position), incentive (no compensation tied to outcome), envy.

Net read: my biases tilt me bullish. The base case ($45-55 fair-value range, recommendation Buy) is robust to debiasing. A more aggressive Strong Buy call would not be — it would be the biases speaking.

10-Year Outlook

Same fundamental business model in 10 years? Yes, with one material caveat. KMI in 2036 is still moving molecules through steel pipe under FERC-regulated tariffs. The business shape — capacity reservation contracts on long-lived regulated assets — is identical to BHE's pipelines, identical to KMI today, identical to KMI 20 years ago. The caveat: the molecule mix may shift. Hydrogen blending (5-20%) is technically feasible on a portion of the network; CO2-for-sequestration (KMI already operates the largest CO2 pipeline system in the U.S.) is a growing rate base. Renewable natural gas (RNG) blending is small but growing. The pipeline as infrastructure outlives any specific molecule.

Customer base larger? Probably yes in dollar terms, possibly flat in volume terms. LNG export capacity roughly triples by 2030 and adds another tranche by 2035 — that is the largest customer-cohort growth driver. AI/data-center gas-fired generation adds a new high-quality cohort (long-duration contracts, hyperscaler counterparties). LDC volumes shrink in the Northeast / California, hold flat in Texas / Southeast. Net: customer base grows in revenue and in credit quality.

Profit per customer higher? Yes. Inflation flow-through plus rate-base growth at 5-7% per year compounds the per-mcf tariff. Take-or-pay contracts re-price higher on renewal in inflationary regimes.

Moat wider? Slightly. Permitting difficulty has gotten worse over the past decade, not better — every year that passes makes KMI's existing footprint harder to replicate. Mountain Valley Pipeline confirmed this: $7B and 9 years for 303 miles. The intangible certificate value rises as new entry becomes harder.

Single biggest threat? Regulatory regime change driven by a major incident (methane leak, explosion, cyber failure) plus political reversal. This is the same threat Buffett flags for BHE [1][4][6] — "we put a large amount of trust in future regulation" — and the same threat that materialized at PG&E in 2018-2020. KMI's safety record is industry-median, not best-in-class; a single multi-fatality incident would reset the political math for a generation.

Confidence: The business is comprehensible, the moat shape is durable, and the demand-side risks are bounded by 10-15 year contracts. Forecasting precise growth rates is hard but the direction is clear. I am willing to sign up for 10 years.

CONFIDENCE: medium