United Airlines Holdings Inc UAL

Thesis

United Airlines Holdings is one of three legacy U.S. network carriers, operating roughly a thousand mainline aircraft across United and United Express through hubs in Newark, Chicago, Denver, Houston, Los Angeles, San Francisco, and Washington Dulles. It sells passenger seats, cargo lift, and a co-branded MileagePlus loyalty program. The business compounds capital only when industry-wide capacity discipline holds, fuel cooperates, labor contracts stay open, and the macro cycle turns up. Three of those four are outside the company's control.

The scorecard tells the story plainly: ten-year average ROIC of 1.8% (below any reasonable cost of capital), trailing five-year FCF conversion of 0.0%, no interest coverage figure available, and net-debt-to-EBITDA of -1.66 (a function of EBITDA volatility through 2020-2021, not balance sheet strength). Maintenance capex is flagged by the scorer as having a >50% spread, meaning we do not actually know what owner earnings are. The scorer prints owner earnings TTM of $4.48B and a base IV of $447.84 against a price of $92.52 — a px/IV ratio of 0.2066 and a composite of 69. On screen, this is a 5x.

It is also exactly the trap Buffett describes in the 1994 and 1996 letters [1][2][3]. A long history of profitable operation and a senior security did not save USAir from a $2.4B four-year loss and a 75% writedown. Cheap commodity businesses with fragile cost structures are not compounders — they are option premium on the cycle. The price/IV math is real, but only if a normal-cycle owner-earnings stream is durable. With ROIC at 1.8% and zero FCF conversion, that premise is unsupported. Pass on quality grounds.

Moat

Pricing power: Effectively none. Domestic and trans-Atlantic capacity is sold on metasearch sites where the marginal traveler picks on schedule and price. United has some yield premium versus ULCCs on premium cabins and in fortress hubs (Newark, Houston, Denver, San Francisco), but those rents are rebated to labor at every contract renewal. Buffett: "in a business selling a commodity-type product, it's impossible to be a lot smarter than your dumbest competitor." [3 / failures excerpts] That dumb competitor in U.S. aviation has historically been a carrier just out of Chapter 11 with a fresh-start cost structure [1, latticework].

Switching costs: Modest, asymmetrically distributed. MileagePlus and Premier elite status create real stickiness for the top decile of business flyers, and the co-branded credit card with JPMorgan Chase generates high-margin, off-cycle revenue. This is the one genuinely attractive asset inside UAL — loyalty programs are estimated by sell-side analysts to be worth a meaningful fraction of enterprise value. But it is a moat around a slice of the customer base, not the airline. The other 80% of seats are sold to price-takers.

Network effects: Hub-and-spoke is a real but narrow advantage. Slot-constrained airports (Newark, Washington National, London Heathrow) and bilateral international authority create defensible nodes. New entrants cannot replicate Newark. But network effects in airlines do not compound — they merely set a floor. Once Delta and American match the network, surplus accrues to passengers and pilots, not shareholders.

Intangibles: Brand is weak. Travelers buy schedule and price, then complain about the experience. Regulatory intangibles (slots, route authorities, FAA certifications) are real but shared across the legacy three. They are entry barriers, not pricing power.

Cost advantages: This is where the thesis dies. Legacy carriers run with unionized pilots, flight attendants, mechanics, and ground crews on pattern-bargaining contracts. Every successful negotiation at one carrier resets the floor at the others — exactly the "moving target" Buffett described in 1994 [1, latticework]. Fuel is a pure pass-through commodity. Aircraft are bought from a duopoly (Boeing/Airbus) that captures most of the surplus from any technological improvement [4, Damodaran on Boeing pricing power]. Maintenance, airport fees, and ATC are largely fixed. The carriers with structural cost advantages are Southwest (single fleet, point-to-point) and the ULCCs (Spirit, Frontier, Allegiant) — not the legacies.

Competitor stress test ($10B + 5 years): A well-funded entrant cannot replicate Newark slots in five years. But that is not the relevant test. The relevant test is whether $10B of capacity added by Delta, American, JetBlue, or a recapitalized Spirit on UAL's profitable routes would compress unit revenue. The answer from forty years of deregulated history is yes, every cycle, without exception. Buffett's 1992 line still holds: airlines behave "as if they are members of a competitive tontine, which they wish to bring to its conclusion as rapidly as possible" [5, moat excerpts].

Erosion risk: High and continuous. Pilot scope clauses, FAA pilot supply, and post-2026 labor contracts at Delta and American will reset wage benchmarks. Sustainable aviation fuel mandates are a cost wedge, not a revenue opportunity. NDC distribution shifts may compress the 2-3 points of margin currently captured from corporate travel agencies.

Moat verdict: NARROW (and that is generous; the honest answer for the airline ex-loyalty is NONE).

Management & Capital Allocation

Management at UAL under Scott Kirby is widely regarded as among the more strategic operators in the industry — disciplined on capacity, focused on premium cabin densification, and willing to cut unprofitable flying. The United Next fleet plan, the Polaris business cabin investment, and the Newark slot recapture are real operational wins. None of that changes the structural math of the industry, but it materially separates UAL from a passive operator. We grade the operator, then re-grade against industry constraints.

Reinvest: The fleet renewal program (Boeing 737 MAX, 787, Airbus A321neo, A321XLR) is enormous — a multi-decade, multi-tens-of-billions commitment. The scorer's 5-year ROIIC of 2.7 looks attractive on the surface but is contaminated by base-period distortion (2020-2021 had near-zero invested-capital denominators in some segments). On a normalized basis, returns on fleet capex track industry ROIC, which is the 1.8% ten-year figure on the scorecard. Reinvest grade: average. The capital is going to work, but at returns that do not exceed the cost of capital across a cycle.

Acquire: UAL has not done a transformational acquisition under the current team. Industry consolidation in the U.S. is functionally complete. Watch for any move on a distressed ULCC at the bottom of the next cycle — that would be the one acquisition that historically creates value (slot and gate accretion at distressed prices). No grade penalty here.

Debt: The capital structure remains heavy. Net-debt-to-EBITDA at -1.66 reads as cash-rich on the headline, but that ratio is dominated by lease-accounting and EBITDA volatility; on a gross-debt basis UAL carries tens of billions. Interest coverage is null on the scorecard, which is itself a tell. The company refinanced through 2020-2021 at favorable rates secured by the MileagePlus program — a creative move at the time, and one that mortgaged the loyalty asset. Grade on debt: cautious B-minus.

Buybacks: Resumed in 2024-2025 after the post-COVID share-issuance trough. The 10-year share count change of +0.1% on the scorecard tells you UAL is not a serial diluter — but the relevant test is average price paid versus IV. We do not have the average buyback price disclosed in this brief. Given that the stock has traded in a $30-$110 range over five years and current price is $92.52 versus base IV $447.84, opportunistic buybacks at any price below $200 would create value if the IV holds. The IV does not hold (see inversion). Therefore buybacks are likely accretive in the short run and value-neutral over a cycle. Grade: B.

Dividends: None. Appropriate for the industry. No grade penalty.

Communication quality: Investor-day disclosure under Kirby and Andrew Nocella is above industry average. RASM/CASM-ex bridges, fleet plans, and segment economics are detailed. The risk factor language in 10-K filings is honest about labor, fuel, and macro exposure. No accounting tells, no working-capital games we can identify in this brief.

Net of all five: this is a B-grade operator running a C-grade business. Buffett's 1996 line is the right calibration: "USAir's performance has recently been helped significantly by an industry tailwind that may be cyclical in nature" [3, moat]. Substitute UAL 2024-2026 for USAir 1996 and re-read.

Capital allocator: B

Industry Structure

Threat of new entrants: Moderate. Slot-constrained airports and FAA Part 121 certification create real friction. But the post-COVID era has shown that ULCCs (Frontier, Spirit, Avelo, Breeze) can launch with point-to-point models on undefended secondary routes, and Chapter 11 routinely recapitalizes weak carriers with fresh-start cost structures — Buffett's "bankruptcy court for airlines has become a health spa" [1, latticework]. New entrant threat to the network legacy is low; new entrant threat to industry pricing is structural.

Bargaining power of suppliers: HIGH and rising. Three suppliers matter: Boeing/Airbus (aircraft duopoly), the unionized labor force, and the oil market. Boeing and Airbus both have multi-year backlogs and are passing through inflation; Damodaran's Boeing case study [4, moat] is a clean illustration of how aircraft OEMs capture most of the surplus from any innovation. Pilot, flight attendant, and mechanic unions extract pattern-bargaining outcomes — every successful negotiation resets the floor for the next one [1, latticework]. Fuel is a commodity input with no negotiating leverage and no hedge that survives a cycle.

Bargaining power of buyers: HIGH. Online travel agencies and metasearch (Google Flights, Expedia, Kayak) have made price comparison frictionless. Corporate travel buyers consolidate spend and demand discounts. The marginal leisure traveler picks on price. Only the elite-status frequent flyer is captive, and that is maybe 10-15% of revenue.

Threat of substitutes: Moderate. For sub-300-mile trips, driving and (in select corridors) Amtrak compete. Video conferencing has structurally reduced some business travel post-2020 — Zoom has not killed the road warrior, but it has bent the curve. High-speed rail in Europe is a meaningful substitute on intra-European routes. The substitution threat is gradual but cumulative.

Rivalry among existing competitors: EXTREME and structural. Capacity is added in lumps (you cannot buy half a 787), capacity cannot be easily removed (aircraft must fly to cover lease costs), and marginal cost per seat once the plane is scheduled is near zero — the textbook setup for price wars. Buffett 1992: airlines behave "as if they are members of a competitive tontine" [5, moat]. Forty years of deregulated history shows that any period of industry profitability invites capacity addition that destroys it. The current 2023-2026 "capacity discipline" narrative will end the way every previous one has.

Value pool location and trajectory: The value pool in U.S. aviation has migrated decisively away from passenger flying and toward (a) credit-card co-brand economics (Chase, Amex, Citi capturing card-issuer rents and sharing with airlines), (b) cargo (transient, currently softening), and (c) loyalty-program redemptions and currency float. The flying business itself is a low-return commodity. UAL captures part of the loyalty rent through MileagePlus, which is the one asset inside the enterprise that a Buffett-Munger investor would actually want to own.

Industry Verdict: Poor

Inversion (Bear Case)

I am short United Airlines Holdings at $92.52 with a twelve-to-thirty-six month horizon. Here is the bear case, played straight.

1. The single event that kills this. A new pilot contract benchmark, set either at Delta in 2026-2027 or via an FAA-mandated rest-rule change, raises mainline pilot CASM by 12-15%. Pattern bargaining drags UAL's contract higher on the next opening. CASM-ex rises 4-5% in a single year while a normalizing macro cuts unit revenue by 6-8%. Operating margin compresses from current high-single-digits to near zero. The market re-rates the stock from a 9x cyclical-peak earnings multiple to a 5x trough-earnings multiple on a halved earnings base. That is a 65-70% drawdown from $92.52 — not a tail outcome, the central case in any prior airline cycle.

2. Why the moat is narrower than bulls think. Bulls point to Newark, MileagePlus, premium cabins, and "capacity discipline." Newark is a rent that is rebated to pilots at every contract reset — UAL is the collection agent for ALPA, not the residual claimant. MileagePlus is genuinely valuable, but it is encumbered by debt collateralization and its economics depend on the JPMorgan Chase co-brand renewal cycle, which the bank holds the cards on. Premium cabin yield is a function of corporate travel budgets, which are a derivative of the macro and have already been structurally reduced by remote work. "Capacity discipline" is a euphemism for the current point in the cycle; every cycle has had its capacity-discipline narrative, and every one has ended in a fare war. Net-net: the moat is a thin loyalty-program rent surrounded by a commodity flying business. Buffett's 1996 verdict on USAir applies directly: protected by "a long history of profitable operations" while "costs were a holdover from the days when regulation protected profits" [2, moat].

3. Why management is worse than it appears. Kirby is competent and visible — that is the problem. The market is paying for "good management of an airline," which is exactly the trap Buffett identified at USAir under Colodny, Schofield, and Wolf [1, 2, 6 moat]. All three were credible, intelligent operators, and they presided over a $2.4B aggregate loss and a 75% writedown of Berkshire's preferred. The 5-year ROIIC of 2.7 on the scorecard is misleading — it captures a recovery off a destroyed 2020 base, not the marginal return on the next $20B of fleet capex, which will earn industry ROIC of ~2-4%. The fleet plan is committed; reversing it is not realistic. The MileagePlus collateralization mortgaged the one asset worth owning. These are operator-grade decisions, not capital-allocator-grade decisions.

4. What bulls are extrapolating that won't hold. (a) Premium-cabin yield growth — driven by post-COVID revenge travel that is mean-reverting. (b) Capacity discipline among legacies — true today, structurally untrue across cycles. (c) ULCC weakness as permanent — Spirit/Frontier are recapitalizing and will return to growth at the bottom of the cycle. (d) Fuel as a tailwind — a regime-dependent assumption; a single supply shock or sustainable-fuel mandate flips it. (e) MileagePlus as a SaaS-like compounder — it is a payments-rebate share agreement with a bank, repriced every contract cycle, not a network-effect monopoly. (f) The IV-to-price ratio of 0.21 as a margin of safety — the IV is built on TTM owner earnings of $4.48B, and the scorer itself flags maintenance capex with a >50% spread. The base IV of $447.84 is a calculation, not a forecast.

5. Valuation trap. The most dangerous form of mispricing is a cyclical at peak earnings on a low multiple. UAL today: 8.4x TTM P/E, $92.52 price, 0.21x px/base-IV. The bull narrative is "only 8x earnings, IV says $250-$676." The bear translation: 8x peak earnings is 16x trough earnings. If through-cycle owner earnings are $1.5B (one-third of TTM, consistent with 1.8% 10y ROIC and historical industry behavior), the through-cycle IV at a 12x multiple is $18B equity value, against a current market cap of roughly $30B. That is roughly $55 per share — a 40% downside before any tail outcome. Add credit stress in a recession (the company carries tens of billions of gross debt, and EBITDA volatility is the reason the net-debt/EBITDA print is -1.66 today and was infinitely negative four years ago) and the equity option value compresses further. Every prior airline cycle has produced at least one Chapter 11 among the legacies; the bear case does not require it, but the option is live.

If I am right, the stock could be worth $40-$55 within 24-36 months.

Lollapalooza Bias Check

Anchoring. The 0.21x px/IV ratio and the $447.84 base IV are anchors. They are derived from a TTM owner-earnings figure of $4.48B that the scorer itself flags as having a maintenance-capex range of >50%. I have to consciously refuse to let the screened cheapness pull me toward a positive thesis on a business that fails the fundamental quality test. The composite score of 69 is similarly an anchor — it sits in the recommendable range for many tickers, but the components (valuation 20, capital allocation 19) are doing all the work, while profitability at 13 is the actual signal.

Authority bias, in reverse. Buffett wrote letter after letter about USAir as his archetypal mistake, and Munger spoke directly about airlines as a too-hard category. There is a real risk of overcorrecting — refusing to look at any airline at any price because two authorities once said the industry is hard. The discipline is to apply the four Munger circle-of-competence tests fresh and let UAL fail or pass on its own merits. UAL fails them on its own merits — but the analyst should arrive there from the tests, not from deference.

Recency bias. The 2023-2025 stretch has been unusually profitable for legacies, with capacity-disciplined behavior, recovered international demand, and a strong premium-cabin yield environment. Modeling the next ten years off the last three is the classic cyclical analyst error. I am consciously discounting the recent margin print.

Confirmation bias. Once I noticed the canon excerpts are 70% USAir failure quotes, every fact I read in the 10-K reinforced the bear case. The honest test: is there any fact in the filing that would change my mind? A clean break in pilot pattern bargaining, a regulatory change that limited new-entrant slot access, or a co-brand renewal at materially better economics would matter. I did not see those. But I also did not exhaustively read the filing for them, and the absence of evidence is not evidence of absence.

Social proof. The sell-side bull narrative (United Next, premium-cabin densification, MileagePlus monetization) is internally coherent and broadly held. Refusing to participate when consensus is bullish requires accepting the possibility of being wrong for two to four more years. That is the right trade-off for a Buffett-Munger framework, but it is uncomfortable, and the discomfort is itself a bias to track.

Incentive bias. None operating on me directly in this analysis, but worth noting: management compensation is largely tied to TSR and adjusted operating margin, which incentivizes capacity discipline in the short run and aircraft order commitments that lock in growth in the long run. The order book is, in effect, a future capacity-discipline-breaker.

10-Year Outlook

Same fundamental business model in 2036? Yes. UAL will still be a hub-and-spoke network carrier flying narrowbodies and widebodies on a mix of domestic, trans-Atlantic, and trans-Pacific routes. Sustainable aviation fuel mix will be higher and somewhere between 5-15% of fuel cost. Pilot supply will have re-tightened or re-loosened depending on the trailing demographic curve. Hub structure unchanged.

Customer base larger? Modestly. Long-run U.S. air travel growth tracks GDP plus 1-2% with cyclical noise. Available-seat-miles will be higher in 2036; revenue per passenger in real terms will be roughly flat to modestly down (the long-run trend since 1978).

Profit per customer higher? Highly uncertain — and probably no, in real terms, on a through-cycle basis. The structural cost pressures (labor, SAF mandates, OEM pricing power) outweigh the structural revenue tailwinds (premium cabin, loyalty monetization). The industry has not produced sustained ROIC above cost of capital in any decade-long window since deregulation.

Moat wider? No. The loyalty program may be slightly wider; the flying business will not be. Likely net narrower as pattern-bargaining catches up to current-cycle profitability.

Single biggest threat? A pilot contract reset at Delta or American that pattern-drags UAL's labor cost up 10-15% inside a softening macro. Second-biggest: a recapitalized ULCC adding 200-300 aircraft of capacity into UAL's profitable secondary markets at the bottom of the cycle.

The ten-year picture is recognizable but not better. ROIC is unlikely to be structurally higher than the 1.8% ten-year average on the scorecard. Owner earnings will be higher in nominal terms and lower in real terms relative to today's elevated print.

CONFIDENCE: low