Otis is half-toll-booth, half-industrial: 65% of revenue and 91% of segment profit comes from servicing a 2.5 million-unit installed base. At $77.08 versus a base IV of $125.18, the market is pricing China weakness as permanent.

Plain English

Otis sells elevators, but the real business is servicing them. Once an elevator goes into a building, it has to be inspected and maintained for 20-30 years, by law, almost everywhere. Otis has 2.5 million elevators under contract and 37,000 mechanics. Every month, building owners pay Otis a small fee. The new-elevator sale is mostly a way to win the long service contract that follows. About two-thirds of revenue and over nine-tenths of profit comes from this servicing work. It's like owning a toll bridge that has to be crossed monthly, around the world.

Thesis

Otis Worldwide is the global #1 elevator company and, more importantly, the global #1 elevator-servicer. The franchise has the rare structural feature Buffett looks for: a small, frequently-bought new-equipment sale that locks the customer into a 20-30 year stream of mandated, high-margin maintenance. Per the 2025 10-K, New Equipment is 35% of sales but only 9% of segment operating profit; Service is 65% of sales and 91% of segment profit. The maintenance portfolio is roughly 2.5 million units globally, served by 37,000 mechanics from 1,400 branches in 70+ countries. Once an elevator is installed, the building owner's switching cost is the operational risk of changing who is responsible for a life-safety device that is regulated, inspected, and rides on customer-specific service records.

The scorecard rates this an 80/100 composite (Profitability 19/25, Balance Sheet 18/25, Capital Allocation 20/25, Valuation 23/25). The headline metrics support the model: 10-year average ROIC of 154.6% (a function of the post-spin negative-equity capital structure — the right read is 'extremely capital-light service economics'), 5-year ROIIC of 35.0%, and a 10-year share-count change of -1.4% per year as Otis has steadily bought back stock since the 2020 spin from RTX.

The price-to-IV math: at $77.08 vs. base IV of $125.18, OTIS trades at 0.62x base IV. The reverse-DCF implied growth is 7.7%, well below the 14% base CAGR the scorer used (already clamped down from a 16.4% raw figure). IV-low is $69.24 — the stock is only ~11% above the bear-case fair value, giving downside protection. This is a moderate-conviction Buy: at $70 it becomes a high-conviction Buy; above $130 trim toward IV-high $162.50.

Moat

Otis is a textbook example of what Buffett calls a 'toll bridge' — and the toll is paid in 200+ countries every month under regulatory mandate. Five moat lenses:

1. Switching costs (HIGH) — This is the dominant moat. Once an elevator is installed in a building, it stays for 20-30 years and must, by law in essentially every developed jurisdiction, undergo periodic inspection and maintenance. Building owners face a strong status-quo bias because (a) the OEM holds the unit's service history, calibration data, and proprietary parts catalog, (b) any incident on a switched unit becomes a liability question, and (c) Otis ONE IoT data, calibration tools, and Gen360's electronic architecture reinforce the OEM-as-servicer logic. Per the 10-K, the maintenance portfolio is grown 'through conversion of newly installed units into maintenance contracts' — i.e., the New Equipment sale is in part a customer-acquisition expense for the Service annuity. Damodaran notes [2] that 'the most significant barrier to entry... is the cost to the end-user of switching from one product to a competitor.' Otis runs the Microsoft-Excel playbook of vertical integration into the user's daily operations.

2. Cost advantages (MEDIUM-HIGH from density) — 37,000 mechanics out of 1,400 branches is a route-density advantage that competitors cannot replicate locally without huge fixed-cost investments. Service margin scales with units-per-mechanic-per-day, which scales with route density, which scales with installed base — a positive feedback loop that has compounded for 100+ years. This is the same flexibility-from-fixed-cost advantage Damodaran [6] credits to Southwest. New entrants face a chicken-and-egg problem: they cannot win service routes without density, and cannot build density without service routes.

3. Intangibles / brand (MEDIUM) — 'Otis' is one of the oldest brand names in industrial America (1853). For building owners specifying equipment in tall residential and commercial towers, the brand is a procurement-risk reducer. But this is weaker than Coca-Cola brand power [1]: building owners care about reliability and safety records, not emotional affinity. The brand matters at spec-time more than at consumer-purchase-time.

4. Network effects (LOW) — Limited; the IoT layer (Otis ONE) creates some weak network learnings (more units = better predictive maintenance models) but the moat is not network-shaped.

5. Pricing power (MEDIUM in Service, LOW in New Equipment) — Service contracts have annual escalators and Otis can re-price modernization upgrades cycle-over-cycle; price increases are typically absorbed because they are small line items in a building's operating budget. New Equipment is highly competitive — particularly in China, where local competitors (KONE has retreated, Schindler is pressured, Otis itself reorganized China operations in January 2025 and disposed of part of Otis Electric to its JV partner in October 2025).

Competitor stress test ($10B + 5 years) — Could a deep-pocketed entrant break this? No. $10B over 5 years is not enough to build a 37,000-mechanic field force in 70 countries, win regulatory approvals, accumulate 100+ years of installation records, or convert the 2.5M-unit installed base. The only credible attackers are existing peers — KONE, Schindler, TK Elevator (private), Mitsubishi, Hitachi. The industry is a stable global oligopoly. KONE/Schindler have similar Service-led economics; competition is by region, not winner-take-all.

Erosion risks — (a) Independent service providers (ISPs) chip at older units, particularly out-of-warranty Gen2's; Otis's response is Gen360 with proprietary electronics that are harder for ISPs to service, plus IoT data lock-in. (b) China new-build collapse permanently reduces the conversion funnel — already happening, reflected in the China reorganization. (c) Modernization is a hidden-but-large profit pool that depends on building owners choosing to upgrade rather than replace; if owners opt for tear-out/replace tenders, competitive bid dynamics return. The Buffett 2007 letter [4] note about See's applies: 'Long-term competitive advantage in a stable industry is what we seek' — Otis qualifies.

Moat verdict: WIDE.

Management

Otis was spun out of United Technologies (now RTX) on April 3, 2020. The post-spin track record is now ~6 years long — enough to evaluate a capital-allocation pattern but short enough that the founding-CEO instincts still dominate.

Reinvestment — Otis has reinvested in product (Gen3, Gen360, Otis ONE IoT), in field-service productivity (UpLift transformation program, $150M total cost, substantially complete by Dec 2025), and in service-portfolio M&A (small bolt-on acquisitions of independent service routes). The 5-year ROIIC of 35.0% indicates that incremental capital is earning very high returns — a hallmark of a compounder where the reinvestment runway exists. The reverse-DCF implied growth of 7.7% versus the scorer's 14% base CAGR suggests the market is not paying for the ROIIC. ROIC of 154.6% (10-year average) is a capital-structure artifact post-spin (negative book equity) but directionally confirms a capital-light annuity.

Acquisitions — Otis has avoided 'transformative' deals. The bolt-on service-route acquisitions are small, accretive, and operationally sensible (the new acquired routes plug into existing mechanic density). The October 2025 disposal of part of Otis Electric to its Otis China JV partner is a value-recognition event for an underperforming China asset rather than capital destruction. So far, no Quaker-Snapple-style mistake [1].

Debt — This is the area to watch. Net Debt / EBITDA is 3.38x — elevated for an industrial. Interest coverage of 9.52x is comfortable, but the leverage was deliberately set high at the spin (United Technologies optimized the spinco capital structure). Management has been deleveraging gradually rather than aggressively. The August 2025 amendment of the 5-year revolver and ongoing maturities (Notes 0.318% due 2026, 2.875% due 2027, etc.) put refinancing into a higher-rate environment than the original issuance — a real but manageable headwind.

Buybacks — As of March 31, 2026 the Board authorization had $900M remaining of a $2B program. Q1 2026: 4.5M shares for $400M (avg ~$89). Q1 2025: 2.6M shares for $253M (avg ~$97). Net 10-year share count change is -1.4% per year — modest but consistent. The critical question is the avg P/IV at which buybacks were executed: with current IV-base $125, recent buybacks at ~$89-97 were done at 0.7-0.8x base IV — value-creating. Buying back stock at 62% of IV (today's price) would be a layup; at 110% of IV it would be the See's-style mistake [4] of putting capital where it doesn't earn.

Dividends — Otis pays a modest dividend with steady increases since the spin. Distribution policy is balanced rather than dogmatic.

Communication — Investor day disclosures are thorough (segment splits, maintenance portfolio size, modernization opportunity sizing). Bullet points avoid the Buffett-flagged sin [1] of 'extraordinarily optimistic' assumptions. The German tax litigation and Belgium tax loss are disclosed clearly with reserves taken (the corpse is filing the death certificate, in the right way).

Open issues — (a) Why has deleveraging been slow? (b) Is the buyback authorization being expanded fast enough at today's price? (c) The China reorganization required restructuring charges in two consecutive years — was the original underwriting of China growth too aggressive? (d) UpLift charged $150M and is supposed to be one-time — watch for 'continuous improvement' creep.

Capital allocator: B+. Not a Berkshire-grade allocator yet, but disciplined, value-aware, and not yet caught making a deal-of-ego mistake. The buyback discipline at sub-IV prices is the most important forward signal.

Capital allocator: B.

Industry

The global elevator industry is a stable, century-old oligopoly of six players (Otis, KONE, Schindler, TK Elevator, Mitsubishi, Hitachi) competing primarily by geography and Service-portfolio density. Porter's Five Forces:

1. Threat of new entrants (LOW) — Building a global elevator service network requires (a) regulatory approvals in 70+ jurisdictions, (b) tens of thousands of trained mechanics, (c) 100+ years of installed-base records, (d) parts-supply chains for a multi-decade product life, and (e) brand acceptance from real-estate developers and building owners. The capital required is in the tens of billions and the time required is decades. There has been no successful new global entrant in living memory. Chinese local competitors have grown in China specifically but have not exported globally.

2. Bargaining power of buyers (MEDIUM in New Equipment, LOW in Service) — On New Equipment, large developers run competitive bids; the Big-Six bid each other down in China particularly. On Service, switching costs are high (life-safety equipment, OEM data, parts), customer fragmentation is enormous (millions of building owners worldwide, no single customer material), and the Service line item is small relative to a building's total operating cost. The Service segment's 91%-of-profit weight tells you which side of the business actually has pricing power.

3. Bargaining power of suppliers (LOW-MEDIUM) — Steel, copper, electronics. Otis's 240-day supplier finance program (disclosed in the 10-K) and global procurement scale (further centralized via UpLift) provide sourcing leverage. Component suppliers are not concentrated, and elevators are a low-tech-as-a-percentage product.

4. Threat of substitutes (LOW) — Stairs do not work in tall buildings. Inclined moving walks substitute marginally for short-distance horizontal moves. Building height itself is the only real substitute, and the urbanization trend is for taller, denser, more elevators. Modernization is sometimes a substitute for full replacement, but Otis captures both ends.

5. Competitive rivalry (MEDIUM) — High in New Equipment in China; moderate in mature markets where service relationships and density determine wins; low at the regulatory/safety reputation level. The Big-Six all have decent ROICs and stable share — a sign that rivalry is not destroying industry value pools.

Value pool location and trajectory — The high-value pool is in (a) Service (recurring, high-margin, 91% of segment profit), (b) Modernization (cyclical-but-large; the global installed base of older units is a multi-decade upgrade tailwind in developed markets), and (c) New Equipment in select developed markets (lower margin but the customer-acquisition channel into Service). The low-value pool is China New Equipment, which is structurally pressured. Otis's portfolio mix is shifting — by intention and by China weakness — toward the higher-value pools, which should be margin-accretive over time.

Long-term durability — Urbanization, aging building stock in Europe and the U.S., regulatory mandates for safety upgrades (post-Grenfell-style code revisions), and IoT-enabled service expansion all support a 10-20 year tailwind. The key risk is China — both as a New Equipment market in continued retrenchment and as a competitive launching pad for Chinese OEMs to globalize.

Industry Verdict: Good. Not 'Excellent' because of China overhang and the New Equipment leg's commoditized economics, but the Service-dominated industry structure is closer to See's [4] than to autos.

Inversion

I am now playing short-seller. Forget the bull case.

1. The single event that kills this — A multi-jurisdictional safety-product-liability event tied to a specific Otis unit family — say, a Gen2 controller defect or a Gen360 electronic failure mode causing a series of entrapments or injuries — could simultaneously trigger (a) regulatory recall costs, (b) brand erosion in Service (the brand-as-safety-shorthand effect cuts both ways), (c) class-action litigation in the U.S. and Europe, and (d) competitive defection of installed-base service contracts. Elevators are life-safety equipment; the asymmetric loss function that protects Otis's moat in good times becomes a guillotine in a bad one. The German tax litigation and Belgium tax loss already disclose how slowly these contingent obligations resolve in Otis's jurisdictions.

2. Why the moat is narrower than bulls think — Bulls assume the 2.5M-unit installed base translates into permanent Service economics. But: (a) Independent service providers (ISPs) win 5-15% of post-warranty units in most mature markets — the bull narrative under-states this churn. (b) Modernization is competitively bid — when an old unit comes up for full upgrade, it is essentially a new RFP and Otis competes against KONE/Schindler/TK on price. (c) Otis ONE and Gen360 IoT lock-in is real but not unique — every competitor has a similar offering, and IoT data interoperability requirements are creeping into European regulation. (d) The 'OEM-as-servicer' logic depends on building-owner inertia, which younger, data-native facility managers are eroding. Switching costs that look like switching costs in 2026 may look like a marketing argument by 2036.

3. Why management is worse than it appears — (a) Net Debt / EBITDA of 3.38x is high for a cyclical-with-Service-overlay; the leverage was inherited from the spin but management has not aggressively de-levered, suggesting either complacency or a buyback bias that is more shareholder-pleasing than balance-sheet-prudent. (b) UpLift was sold as a one-time transformation; expect 'continuous improvement' charges to keep recurring under different names. (c) The China reorganization required restructuring charges in 2024, 2025, and now Q1 2026 — three rounds of 'one-time' costs is a pattern, not an event. (d) The October 2025 disposal of part of Otis Electric to the JV partner is being framed as portfolio-shaping; another reading is that Otis was forced into a value-conceding sale because its China position was deteriorating faster than expected.

4. What bulls are extrapolating that won't hold — Bulls extrapolate (a) 14% base CAGR (the scorer already clamped this from 16.4%), (b) Service margin expansion from UpLift productivity, (c) Modernization tailwind from aging European stock, and (d) maintained buyback pace. But: New Equipment in China is in structural decline, and that 35% of revenue weight gets smaller — sounds bullish for mix-shift, but it also means the conversion funnel into future Service shrinks. By 2035, the maintenance portfolio could plateau or shrink in China, the world's largest installed base. Modernization tailwinds in Europe are real but get competitively bid away. UpLift productivity gets matched by competitors over 2-3 years. The 'recurring' Service revenue is not literally annuity — contracts have 1-5 year terms, and ISP penetration is real.

5. Valuation trap (multiple compression / regime change) — TTM P/E of 20.04x is reasonable; 10-year average P/E is 31.16x. Bulls anchor to the 10-year average and call the stock 'cheap.' But that 10-year average reflects (a) the post-spin honeymoon, (b) ZIRP-era quality multiples, and (c) pre-China-collapse growth narratives. The true through-cycle P/E for an industrial with cyclical New Equipment exposure and modest Service growth might be 15-18x. If you re-rate Otis to 16x on TTM owner earnings of $1.17B, you get an EV roughly $19B versus a market cap closer to $30B. The IV-low of $69.24 in the scorecard captures part of this risk; a true bear case multiple-compression scenario gets you to the high $50s. Combined with leveraged balance sheet and a refinancing wall in the higher-rate environment, multiple compression and earnings disappointment can compound.

The inversion is real. The bear case is not '40% downside in a panic and you wait it out' — it's 'this gradually re-rates to a 16x mature-industrial multiple over 3 years while modernization disappoints and ISP penetration creeps.'

If I am right, the stock could be worth $58 within 3 years.

Lollapalooza Bias Check

Which biases are active in me, the analyst, right now?

Anchoring (HIGH) — The scorecard hands me a base IV of $125.18 and a current price of $77.08. The very framing 'price-to-IV ratio of 0.6157' anchors me to the conclusion that the stock is cheap. The reverse-DCF implied growth of 7.7% looks comfortably below the 14% base CAGR. Every input I'm given pre-disposes me to a 'Buy.' I have to discount this — anchoring is the single most active bias in this analysis. The IV figures are themselves model outputs, not facts.

Authority bias (MEDIUM) — Otis is the original elevator company (1853), bears the brand of an iconic founder, was carved out of the prestigious United Technologies, and is the global #1. I am inclined to trust the franchise narrative more than I would trust an unbranded equivalent. Buffett's See's framing [4] reinforces this — 'long-term competitive advantage in a stable industry' is irresistible language and I want Otis to be a See's. It might be; but the very desire is the bias.

Confirmation bias (MEDIUM-HIGH) — Once I formed the 'recurring Service annuity at a discount to IV' thesis early in this analysis, every subsequent fact I read (91% of profit from Service, 2.5M unit portfolio, IoT lock-in) confirmed it. I had to actively construct the inversion section to make the bear case visible. The China retrenchment, the three rounds of 'one-time' restructuring, and the 3.4x leverage are facts that don't fit the thesis cleanly — confirmation bias makes them easy to under-weight.

Social proof (LOW-MEDIUM) — Otis is a quality-compounder darling; many value-quality investors own it. I'm aware that the consensus among long-only value investors is constructive. This pushes me toward 'Buy' rather than 'Hold.'

Recency bias (LOW) — The recent 10-Q and 10-K data is fresh in my mind; older industry history (the 2008 China supercycle that built much of the installed base) is more abstract. I may be giving too much weight to recent China weakness and not enough to the long-arc industrial logic.

Deprival super-reaction (LOW) — Not strongly active. I don't own the stock; no commitment-and-consistency pressure to defend a position.

Net effect — The active biases all point toward 'Buy,' which is exactly why I'm calling it a 'Buy' with medium conviction rather than 'Strong Buy' with high conviction. The discount-to-IV is real but anchoring on the IV figure itself is the most dangerous error here.

10-Year Outlook

Same fundamental business model in 2036? Yes, almost certainly. Buildings will still need elevators; elevators will still need mandated maintenance; service routes will still benefit from density. Gen360-equivalent IoT will be standard rather than differentiating, but the underlying toll-bridge structure is unchanged.

Customer base larger? Probably yes. Global urbanization, aging European stock requiring modernization, and Indian/Southeast-Asian construction continuing all support installed-base growth. The wildcard is China — the world's largest installed base may plateau or shrink in service-portfolio terms over the decade. Net base likely +20-30% over 10 years.

Profit per customer higher? Likely yes, modestly. Service contract pricing escalates with inflation; mix shifts toward modernization (higher-revenue events); IoT enables tiered service pricing. Margins should expand 100-300 bps over the decade as UpLift productivity and procurement centralization bear out.

Moat wider? Probably stable, possibly slightly wider in developed markets (electronic architecture lock-in) and slightly narrower in emerging markets (local competition + ISP penetration). Net: similar moat width, deepened in some directions.

Single biggest threat over 10 years? A combination of (a) China new-build remaining structurally weak, shrinking the conversion funnel, and (b) Chinese OEMs (Otis Electric's local competitors, Hitachi-China, etc.) successfully exporting service capabilities to South Asia and Africa, eroding Otis's emerging-markets growth runway. Less likely but more catastrophic: a global product-liability event in a Gen2 or Gen360 family.

Risk of 'Too Hard' — The business is intelligible to a 12-year-old: Otis sells and services elevators worldwide; the servicing is the profit. The 10-year shape is recognizable. This is not a tech-adoption bet, not a regulatory-outcome bet, not a commodity-price bet, not an R&D bet. It passes Munger's circle-of-competence test.

CONFIDENCE: medium

Position Guidance

• Recommendation: Buy
• Conviction: Medium
• Target buy price: $75 (current $77.08 already in the buy zone; add aggressively below $70 = IV-low)
• Target trim price: $130 (above base IV of $125.18; full exit by $160 toward IV-high $162.50)
• Position sizing: Up to 4-5% of portfolio at current prices; up to 6-7% if it trades below $70. Cap at 7% given 3.4x net leverage and China overhang.
• Catalyst to add on: sub-$70 price with no fundamental break, or evidence of accelerated deleveraging.
• Catalyst to trim on: price above $130, OR a leveraging M&A deal, OR a Gen360 product-safety event.
• Re-underwrite trigger: maintenance portfolio unit count growth turns negative for two consecutive years, OR Service operating margin compresses by more than 200 bps from current levels.