ResMed Inc. (NYSE: RMD) — Priced for the GLP-1 Apocalypse, Still Compounding at Double Digits: A 22%-ROIC Franchise at Its Cheapest Multiple in a Decade

Independent Equity Research Date: June 28, 2026 · Analyst desk: Health Care / Medical Devices Price (6/26/26): $200.29 · Market cap: ~$29.0B · EV: ~$28.0B · FY-end: June 30 · CIK: 0000943819

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice and not a recommendation to buy or sell any security. The analysis that follows (sections 1–15) is deliberately position-free and carries no price target anywhere except inside this block.

Verdict: BUY / accumulate-on-weakness. Medium-high conviction. Fair-value zone ~$240–$280 (≈22–25× forward EPS of ~$11), with disciplined accumulation sub-$200 and aggressive interest in the $170s–$180s. Not a short at any price I can defend.

ResMed is a genuinely excellent business — ~22% ROIC, 60%+ gross margin, double-digit revenue growth, net cash, >100% free-cash-flow conversion — trading at ~19× trailing and ~17× forward earnings, the cheapest multiple in its decade-long public history (a valuation screen puts its P/E in roughly the 1st percentile of its own decade-long range). It got there because the market has decided GLP-1 weight-loss drugs will shrink the sleep-apnea pool. The problem with that thesis is that every data point ResMed and Eli Lilly have produced says the opposite over any horizon an investor can underwrite: ResMed’s own 2.1-million-patient claims analysis shows patients on both PAP and a GLP-1 are 11% more likely to start CPAP and resupply more often than PAP-alone patients, Lilly’s Zepbound-OSA campaign is free demand generation funneling patients into the diagnostic funnel, and the mask/resupply business — the recurring razor-blade engine — is still compounding 12–14%. The framing here is de-rated quality / falling knife into accelerating numbers: a quantitative factor-risk model reads RMD as a high-Quality, low-beta, negative-momentum stock with no Value loading and an R² of only 18% — i.e., a quality compounder the tape has abandoned on a single narrative, not a broken business.

What the market is pricing correctly: there is a real, long-tail (10-year+) technological-substitution risk that pharma eventually compresses the addressable OSA population, and ResMed’s growth has decelerated from the Philips-recall sugar-high to a more normal high-single-digit/low-teens cadence. What it is pricing wrongly: the timing. You are paying a sub-market multiple for a business that, on its own evidence, is a beneficiary of the very drug class supposedly killing it for the next several years, while the diagnosed-but-untreated pool (≈80% of the ~1 billion OSA sufferers worldwide) dwarfs any plausible GLP-1 cannibalization. Conviction flip-bull: a clean real-world print showing simultaneous PAP+GLP-1 prescribing accelerates new-patient starts (the “next-phase” analysis management flagged). Conviction flip-bear: the first hard quarter where US new-patient starts or resupply rates decelerate as GLP-1 penetration crosses a threshold — or a credible full Philips re-entry that takes share. Tag: “the disruption its own data calls a tailwind.”

📈 Stock Price Action — Five-Year Event Map

ResMed has round-tripped a full cycle in five years. The stock ran from the ~$180s in mid-2020 to a then-record ~$288 in September 2021 on the Philips-recall windfall, gave most of it back into the 2024 GLP-1 panic (low ~$160 in January 2024), recovered to a new all-time high of ~$291.57 on August 22, 2025, and has since de-rated ~31% to $200.29 (June 26, 2026) — its 52-week range is roughly $180–$291, and it now sits below its 21-, 50- and 200-day moving averages ($195 / $204 / $230). The defining feature is that the most recent leg down (Feb–May 2026, ~$256 → ~$190) happened through a strong earnings print: this is multiple compression, not a fundamental break.

1. The Philips windfall (2020–21): A foam-degradation recall pulled ~5M Philips devices and ~$1B+ of competitor revenue off the market; ResMed absorbed the demand and re-rated to a ~75× trailing P/E peak.
2. The rate-shock unwind (2021–22): As rates rose, the entire premium-multiple medtech cohort de-rated; ResMed also absorbed a ~1,000-bps-adjacent gross-margin squeeze from freight and component inflation (gross margin troughed near 56–57%).
3. GLP-1 panic #1 (2023–24): Lilly/Novo commentary on weight-loss drugs’ effect on OSA triggered the first disruption scare; RMD lost ~35% despite no fundamental deterioration.
4. Thesis repair (2024–25): Two years of continued double-digit growth, recovering gross margin, and ResMed’s own claims data reframing GLP-1 as net-positive carried the stock to a new record.
5. GLP-1 panic #2 + de-rating (2025–26): The oral-GLP-1 rollout reignited TAM fears just as the broad healthcare complex de-rated; Morgan Stanley cut to Equal-Weight (PT $230, 6/17/26). The Q3 FY26 print on April 30 (+11% revenue, +21% non-GAAP EPS) did not arrest the slide — a classic narrative-over-numbers tape.
6. Stabilization (mid-2026): The stock has based around $190–$200; KeyBanc maintains Overweight (PT $266, 6/22/26). Price moves are Fact; attributed causes are Interpretation.

1. Executive Summary

ResMed is the global leader in obstructive sleep apnea (OSA) and respiratory-care devices, the larger of a two-player flow-generator duopoly (with Philips Respironics) that became a near-monopoly in the US after Philips’s 2021 recall and 2024 consent-decree exit. The company sells cloud-connected CPAP/APAP/bilevel devices, a high-margin and faster-growing recurring stream of masks and resupply, and a residential-care software portfolio (Brightree, MatrixCare, MEDIFOX DAN). It compounds: revenue grew from $2.96B (FY20) to $5.15B (FY25) and $5.54B on a trailing-twelve-month basis, at ~10–15% per year, while sustaining a ~22% return on invested capital, ~59–62% gross margins, ~33–37% operating margins, net-cash balance sheet, and >100% free-cash-flow conversion. These are the financial fingerprints of a real, durable competitive advantage, not a narrative.

The stock trades at ~$200, down ~31% from its August-2025 record and at the cheapest valuation of its public life — ~19× trailing GAAP EPS (~$10.34), ~17× forward earnings, ~12.8× EV/EBITDA, and a composite valuation percentile in the 7th of its own decade-long range. The cause is singular: the market believes GLP-1 weight-loss drugs (tirzepatide/semaglutide and now oral agents) will durably shrink the OSA population and the CPAP installed base. The evidence to date contradicts this over any horizon shorter than a decade. ResMed’s own 2.1-million-patient and 1.7-million-patient real-world analyses show GLP-1 co-therapy increases PAP initiation and resupply; Lilly’s own messaging and its Zepbound-OSA marketing campaign treat the two as complementary and generate free demand for the diagnostic funnel; and the recurring mask/resupply business is still growing 12–14%.

The investable tension is therefore one of timing and probability, not business quality. The bear case is a genuine but long-dated substitution risk; the bull case is that a high-ROIC, net-cash compounder with an ~80%-untreated global TAM is being handed to buyers at a discount because of a disruption its own data calls a tailwind. The capital-allocation record is conservative and intelligent (growing buyback, sane bolt-on M&A, heavy reinvestment), with two governance caveats: a combined Chairman/CEO and a compensation plan that omits any explicit returns-on-capital metric. This memo carries no recommendation and no price target; it lays out the embedded expectations and the falsification tests for each side.

2. Business Overview

ResMed Inc., founded in 1989 and headquartered in San Diego, develops, manufactures, and markets medical devices and cloud-based software for the treatment of sleep-disordered breathing, respiratory insufficiency, and out-of-hospital care. The company employs more than 10,000 people and sells in roughly 140 countries. It reports two operating segments, recently re-named in line with its “2030 strategy”: Sleep and Breathing Health (formerly Sleep and Respiratory Care), which is ~88% of revenue, and Residential Care Software (RCS) (formerly Software-as-a-Service), ~12% of revenue.

Sleep and Breathing Health is the core. It comprises three economic pieces:

• Flow-generator devices (~52% of total revenue, ~$2.66B FY25): CPAP, APAP, and bilevel machines — chiefly the AirSense 10/11 platform — that deliver pressurized air to keep airways open during sleep, plus ventilators and oxygen products for respiratory insufficiency and COPD. Devices grew ~6% in the US and ~6–10% internationally in recent quarters; this is the slower-growing, lower-margin, “razor” portion.
• Masks and accessories / resupply (~36% of total revenue, ~$1.86B FY25): The “blades.” Every device in the installed base requires periodic replacement of masks, cushions, tubing, and filters. This stream grew 12–14% in constant currency in recent quarters, carries higher gross margin than devices, and is the most valuable, most recurring, most defensible part of the company. Resupply is increasingly orchestrated through ResMed’s digital channels (myAir, Brightree resupply, ReSupply programs).
• Digital health ecosystem: AirView (cloud platform for clinicians/providers to monitor and adjust therapy), myAir (patient engagement/compliance app), and home sleep-apnea testing assets (NightOwl/Ectosense, VirtualOx/VertuOx, Somnoware). The ecosystem is the connective tissue that lifts adherence — the single most important driver of lifetime patient value — and raises switching costs for clinicians and home medical equipment (HME/DME) providers.

Residential Care Software (RCS) is a portfolio of out-of-hospital healthcare software: Brightree (business-management software for HME/DME, pharmacy, home-infusion, O&P providers), MatrixCare (care management for senior living, skilled nursing, home health, hospice), HEALTHCAREfirst, and MEDIFOX DAN (German home/long-term-care software, acquired ~2022). RCS is ~12% of revenue, recurring/subscription in nature, and grew only ~4% in constant currency in the March 2026 quarter — dragged by the senior-living/long-term-care vertical and carried by MEDIFOX DAN. Management is reshaping the portfolio and targets a re-acceleration to high-single-digit revenue and double-digit operating-profit growth in FY27.

How it makes money: Geographically, the US, Canada and Latin America are ~58% of device-and-mask revenue and Europe, Asia and other ~42%. The business model is an installed-base razor/razor-blade: sell (or place, via HME) a flow generator, then monetize a multi-year stream of high-margin mask resupply, with software and data deepening stickiness. Revenue is predominantly product (devices + masks) with a growing recurring/subscription overlay (resupply + RCS). The recurring share — resupply plus software — is well over 40% of revenue and rising.

The digital ecosystem as connective tissue. What distinguishes ResMed from a commodity device maker is that the hardware is wrapped in a proprietary data and software layer. AirView is the clinician/provider command center — remote monitoring, troubleshooting, and over-the-air device-setting changes across the installed base — and myAir is the patient-facing engagement and compliance app. Brightree, which ResMed owns, is the back-office operating system for a large share of US HME/DME providers; that ownership is strategically potent because it positions ResMed at the billing and resupply nerve center of its own distribution channel. The cumulative effect is a flywheel: more connected devices → more adherence and resupply data → better algorithms and provider tools → higher adherence → more resupply and stickier clinicians. This is the mechanism behind the unusually high and durable margins, and it is the asset most resistant to commoditization by a re-entering Philips or a low-cost entrant, neither of which can replicate the data network quickly.

Verdict: A focused, global, installed-base-driven medical-device franchise with a high-margin recurring resupply core and a software adjacency. The revenue model is structurally attractive (recurring, sticky, cash-generative). The single most important business fact for an investor is that the resupply engine — not device sales — is the value driver, and it is still compounding double digits.

3. Industry Dynamics

The obstructive sleep apnea market is one of the more attractive structures in medical devices, defined by a vast under-penetrated patient pool, a consolidated supplier base, and durable reimbursement.

Market size and penetration. Sleep-disordered breathing is enormous and chronically under-diagnosed. ResMed cites ~1 billion people worldwide affected by OSA, of whom the overwhelming majority — on the order of 80% — remain undiagnosed and untreated. This is the single most important industry fact: the binding constraint on the category is diagnosis and treatment-initiation throughput, not market saturation. Anything that drives more patients to a physician — consumer wearables flagging apnea, GLP-1 demand-generation, rising clinical awareness of OSA’s links to cardiovascular disease, dementia, and Parkinson’s — expands the funnel. The category has compounded mid-to-high single digits structurally (devices ~mid-single-digit, masks ~high-single-digit per management’s market framing), with ResMed growing faster as it takes share.

Supplier structure — a duopoly that became a near-monopoly. Flow-generator manufacturing is a two-player game: ResMed and Philips Respironics. Philips’s June 2021 recall of ~5 million devices (polyurethane foam degradation), followed by a 2024 DOJ/FDA consent decree that kept it out of the US market, removed ~$1B+ of competing supply and handed ResMed both share and pricing latitude. Philips is only now signaling a possible US re-entry; as of the most recent ResMed call, management reports “nothing new” and no observed share loss. Barriers to entry are high: regulatory (FDA clearance, ISO/EU MDR), clinical-evidence requirements, manufacturing scale, an entrenched HME/DME distribution channel, and the installed-base/resupply lock-in. A credible third entrant at scale is implausible on any near horizon.

Reimbursement and channel. In the US, devices flow through HME/DME providers and are reimbursed by CMS and commercial payers, typically with a 90-day adherence requirement (≥70% usage in a 30-day window within the first 90 days) — a structural feature ResMed has turned to its advantage via compliance-driving software, since adherence both secures reimbursement and drives resupply. International markets (France, Germany, Japan, Australia, China) have their own reimbursement and tender regimes; ResMed runs market-specific ecosystems (e.g., a WeChat-integrated model in China). A live, lower-probability risk is payer-led utilization management (e.g., “Synapse”) changing the DME funding model; management characterizes OSA/CPAP as low-cost, low-acuity, high-ROI therapy that payers have little incentive to restrict.

Reimbursement, deeper. The reimbursement architecture is both a moat and a discipline. In the US, CPAP is a Medicare/commercial-covered durable medical-equipment benefit, typically rented over ~13 months before title transfers, with the 90-day adherence gate determining ongoing payment. This structure rewards exactly the capability ResMed has built — adherence-driving software and data — because non-adherent patients are a cost to the HME, and ResMed’s tools convert more patients past the gate. Mask/accessory resupply has its own established reimbursement schedule (replacement frequencies for masks, cushions, tubing), which underpins the recurring annuity. The risks are (a) payer-led utilization management compressing reimbursed frequencies or adding prior-authorization friction (the “Synapse” concern), and (b) competitive-bidding/pricing pressure in the DME channel. Both are real but bounded: OSA therapy is low-cost and high-ROI for payers (it reduces downstream cardiovascular, metabolic, and now dementia-linked costs), giving payers limited incentive to restrict it, and ResMed has operated profitably under stricter utilization-management regimes abroad (France, Germany). Internationally, reimbursement varies — tender-based in parts of Europe, mixed public/private in Asia, and increasingly cash-pay in China — which diversifies the regulatory risk across dozens of independent systems.

The pharma overlay (GLP-1). The defining new industry variable is the GLP-1 class. Tirzepatide (Lilly’s Zepbound) is FDA-approved for OSA (December 2024); oral agents are scaling. The bear framing is straightforward: if weight loss resolves OSA, the device TAM shrinks. The counter — developed in Section 4 — is that OSA is multi-factorial (age, sex, craniofacial anatomy, and weight), persists after significant weight loss, and that GLP-1 use empirically raises both PAP initiation and adherence in the near/medium term while generating demand for the diagnostic funnel. Industry-structurally, GLP-1 is best understood as a demand-generation and patient-flow accelerant today, with an uncertain very-long-tail substitution effect. The sizing matters: even if GLP-1s eventually cured OSA in, say, the obese subset most responsive to weight loss, that would address a minority of the ~1 billion-person OSA pool, while the diagnostic-funnel acceleration is pulling untreated patients — the ~80% who have never been diagnosed — into the system now. For the substitution effect to overwhelm the demand-gen effect, GLP-1s would need not only to resolve OSA durably (uncertain, given SURMOUNT-OSA showed residual disease in most patients even after meaningful weight loss) but to do so faster than the untreated pool is being newly diagnosed. No current data support that timeline.

Capital-cycle read (Marathon). A supply-side lens reinforces the structural call. The classic capital-cycle danger — high returns attracting a flood of new capacity that mean-reverts margins — is muted here because the device-supply side is the opposite of a capital-cycle bust: Philips’s recall withdrew capacity, and regulatory/clinical/channel barriers deter new entrants, so the supply side is consolidating, not expanding. The capital is instead flooding into the adjacent pharma category (GLP-1s), which is where the Marathon caution should be directed — but that capital is, at least initially, expanding ResMed’s funnel rather than competing away its returns. The risk to watch is the longer-term one: that pharma capital, having built an enormous installed base of weight-loss patients, eventually reshapes the OSA disease pool itself.

Verdict: structurally attractive. A consolidated supplier base, high barriers, durable reimbursement, and an ~80%-untreated global pool make this one of the better device markets. The one structural overhang — pharmaceutical substitution — is real but long-dated, and on current evidence is net-additive to patient flow rather than subtractive. Net: a good industry, with a tail risk the market is over-weighting on timing.

4. Competitive Position

ResMed’s competitive advantage is multi-sourced and, unusually for a device company, shows up cleanly in the financials — the test Greenwald demands: if you cannot tie the “moat” to a financial outcome that would deteriorate without it, it is not a moat.

1) Scale-based cost advantage in a two-player market. As the larger of two flow-generator makers, ResMed amortizes R&D, regulatory, manufacturing, and supply-chain investment over the largest device and mask volume in the industry. The financial signature is a 59–62% gross margin and a 33–37% operating margin — and, critically, expanding margins (gross margin +290 bps year-over-year in the March 2026 quarter) driven by component-cost and manufacturing/logistics productivity that smaller players cannot replicate. Management’s stated goal is gross-margin accretion every year through 2030.

2) Customer captivity / switching costs. The deepest piece. ResMed’s flow generators are embedded in clinician workflows and HME/DME operations through AirView (remote monitoring/titration) and Brightree (the HME back-office system ResMed itself owns). A sleep physician or HME that runs its practice on ResMed’s data infrastructure faces real friction switching device brands. For the patient, compliance data, mask fit, and the myAir relationship create stickiness. The financial signature is the razor-blade resupply annuity: a ~$1.86B mask/accessory stream growing 12–14% off an installed base, with multi-year resupply rates management can measure to the basis point.

3) Intangibles — brand, clinical evidence, regulatory. Two decades of clinical evidence (CPAP as the “gold standard” frontline OSA therapy), an 80,000-completion CME program reaching 45,000 clinicians, FDA clearances, and the ResMed brand among sleep physicians and HMEs constitute a real intangible barrier. The newest expression is product differentiation in masks: the fabric-based AirTouch N30i and F30i Comfort/F30i Clear deliver a measured +6% 90-day adherence versus silicone equivalents — and because adherence is the master variable for lifetime value, this is “changing the basis of competition” (management’s phrase, but corroborated by the resupply growth).

Market-share stability test (Greenwald): ResMed’s share has been stable-to-rising for years, accelerated by the Philips exit, with no observed erosion from recent competitive launches. Stable/rising share + high, durable ROIC is the empirical hallmark of a genuine moat.

Earnings-power-value vs. asset-value (Greenwald). A useful frame for the quality of this moat: ResMed’s tangible book value per share is only ~$10–11 (the balance sheet is dominated by ~$3.0B goodwill + ~$0.4B intangibles from software acquisitions), yet the business earns ~$1.4B+ of net income and ~$1.65B of free cash flow. The gap between asset value (low) and earnings-power value (high) is the capitalized value of the franchise — the installed base, the resupply annuity, the clinical-evidence and regulatory barriers, and the brand — none of which sit on the balance sheet. A new entrant attempting to replicate ResMed’s earnings power would have to rebuild two decades of clinical evidence, an installed base measured in the tens of millions of devices, an HME/clinician data infrastructure, and FDA/EU-MDR clearances — a reproduction cost vastly above the book carrying value. That is the definition of a barrier to entry with a financial signature.

The resupply annuity, quantified. The single most important moat mechanism is the mask/accessory resupply stream (~$1.86B, ~36% of revenue, growing 12–14%). Its economics: every device placed creates a multi-year obligation to replace masks (every 3–6 months), cushions (monthly), tubing, and filters — a recurring, high-margin, low-cyclicality cash flow that compounds with the installed base regardless of the device-sales cycle. Because resupply gross margins exceed the device blended margin and the orchestration runs through ResMed’s own software (myAir/Brightree resupply), each incremental installed device is worth materially more in lifetime resupply NPV than its initial sale price. This is why management is willing to invest aggressively in adherence (the +6% from fabric masks) and demand-gen: adherence and installed-base growth are the two levers that compound the annuity. It is also why the GLP-1 debate ultimately reduces to a single question — does GLP-1 grow or shrink the installed base and resupply rate? — for which every disclosed data point currently says “grow.”

Competitive set and threats. (a) Philips Respironics — the duopoly partner, attempting US re-entry; the key watch item, though years of recall damage and a consent decree make a rapid share recapture unlikely. (b) Inspire Medical Systems (INSP) — hypoglossal-nerve-stimulation implant, a surgical alternative for the CPAP-intolerant minority; a niche substitute, not a mass-market threat, and arguably expands total OSA treatment. © GLP-1 pharma (Lilly, Novo) — the substitution tail risk and, paradoxically, the current demand-gen tailwind. (d) New entrants / AI — management is “productively paranoid” but reports no launch in the last 12 months that surprised it.

Direct comparison. Versus Philips, ResMed has scale, an intact reputation, a US-market monopoly window, and the resupply annuity. Versus Inspire, ResMed has cost, breadth, and the frontline-therapy position. Versus pharma, ResMed has efficacy (CPAP remains the most efficacious OSA therapy), the multi-factorial persistence of OSA, and — per its own data — complementarity rather than substitution over the investable horizon.

The Philips re-entry scenario, weighed honestly. The bear’s most concrete competitive argument is not GLP-1 but Philips: pre-recall, the two shared the device market roughly evenly, so a fully rehabilitated Philips could in principle reclaim 15–25 points of share. Three things bound that risk. First, time and trust: Philips spent years under a recall cloud and a DOJ/FDA consent decree; sleep physicians and HMEs that re-tooled around ResMed’s AirView/Brightree workflows do not switch back casually, and patient-safety memory is long. Second, the resupply lock-in: even if Philips recaptures some new-device share, ResMed’s installed base keeps generating mask/resupply revenue for years — the annuity is far stickier than device flow. Third, ResMed’s reinvested lead: the years of Philips absence funded an accelerated ResMed product cycle (AirSense 11, fabric masks, ecosystem) that widened the gap. A Philips return is a real multi-year watch item and would pressure device pricing at the margin, but a rapid share-recapture is low-probability; management reports no observed loss to date.

Verdict: a durable, multi-sourced competitive advantage — scale cost advantage + customer captivity + clinical/brand intangibles — validated by stable/rising share and ~22% ROIC. The honest caveat is the technological-substitution tail risk from pharma, which distinguishes this moat from an unassailable one: it is wide and deep today, with a low-probability/high-impact erosion scenario on a 10-year-plus horizon.

5. Growth History and Forward Opportunities

History. ResMed has compounded revenue from $2.96B (FY20) to $5.15B (FY25) — a ~12% CAGR — and to $5.54B on a trailing basis (to March 2026), with EPS growing faster than revenue throughout (operating leverage + margin recovery). Growth was overwhelmingly organic, supplemented by bolt-on software (Brightree, MatrixCare, MEDIFOX DAN) and home-sleep-testing tuck-ins. The shape of growth by piece in recent quarters: devices +~6% (US), masks/resupply +12–14% cc, RCS +~4% cc. Constant-currency group growth has run ~8% with headline (FX-aided) growth ~11%. This is high-single-digit/low-teens organic compounding — decelerated from the Philips-recall windfall years (FY22–23 saw >15% as ResMed absorbed Philips demand) to a more sustainable cadence.

Quality of growth. High. The mix is shifting toward the highest-quality revenue — recurring, high-margin mask resupply growing fastest — which is why EPS outgrows revenue and incremental operating margins run rich (incremental operating margin ~66% in FY25). This is the opposite of low-quality growth bought with dilution or margin: share count is falling, margins are rising, and growth is organic.

Forward opportunities:

• Penetration of the ~80%-untreated pool — the largest and most durable driver. Awareness tailwinds (wearables, GLP-1 demand-gen, OSA–dementia/Parkinson’s evidence) feed the diagnostic funnel; ResMed’s home-sleep-testing assets (VirtualOx, NiteOwl, Somnoware) and PCP CME programs aim to convert that funnel.
• Masks/resupply share gains via fabric technology (N30i, F30i Comfort/Clear) at premium prices and higher adherence — directly compounding the annuity.
• GLP-1 as demand-gen — management explicitly frames GLP-1s as “a once-in-a-generation demand-gen opportunity,” with planned “investments and partnerships” to capture and channel the increased consumer awareness.
• International / China — AirSense 11 rollout across Latin America and a just-launched China market (premium, consumer-driven, WeChat-integrated ecosystem).
• RCS re-acceleration to HSD revenue / double-digit operating profit in FY27 via portfolio reshaping.
• Adjacency expansion — the Noctrix Health acquisition (restless leg syndrome, ~17M US sufferers, ~7% of adults globally) extends ResMed into the world’s third-most-prevalent sleep disorder through the same sleep-physician/HME channel.
• Management’s 2030 framework: high-single-digit revenue growth with earnings growth above revenue growth and annual gross-margin accretion — a credible, evidence-backed algorithm given the installed-base annuity and operating leverage.

Penetration math. The runway is best appreciated quantitatively. If ~1 billion people have OSA and ~80% are undiagnosed/untreated, the treated base is on the order of ~200 million globally, of whom only a fraction are on PAP therapy. ResMed’s installed base is measured in the tens of millions of devices. Even setting aside any TAM growth, simply converting a low-single-digit percentage of the untreated pool per year sustains high-single-digit volume growth for a decade-plus — before counting price/mix (premium fabric masks), resupply compounding, software, and new adjacencies (RLS). This is why management’s “high-single-digit revenue growth through 2030” framing is conservative rather than heroic: the binding constraint is funnel throughput (diagnosis and setup), not demand, and the company is investing precisely there (home sleep testing, PCP CME, demand-gen partnerships). The GLP-1 bear case is, in effect, an argument that this funnel will reverse — that the untreated pool will shrink faster than ResMed can convert it — which requires pharma to resolve OSA at a scale and speed no evidence supports.

Verdict: high-quality growth. Organic, recurring-mix-enriching, margin-expanding, and self-funded, with a long penetration runway. The debate is not whether the growth is high-quality (it is) but whether the GLP-1 tail risk eventually caps the TAM — a question about the 2030s, not the next several years.

6. Financial Quality

ResMed’s financials are, by the standards of any sector, excellent — and they are clean, which matters because the entire bear thesis rests on a future the numbers do not yet show.

Revenue and margins. Revenue $5.15B FY25 (+10%), $5.54B TTM. Gross margin recovered from a FY24 trough of 56.7% (freight/component inflation) to 59.4% FY25 and 62.2–62.8% in Q3 FY26 (+290 bps YoY) — a recovery management attributes to component-cost improvement and manufacturing/logistics productivity, with a 62–63% FY26 target and stated multi-year accretion to 2030. Operating margin ~32.8% FY25, 34.9% GAAP / 36.7% non-GAAP in Q3 FY26. EBITDA margin ~37%. These margins are rising, not eroding.

FY21 net margin/EPS/returns depressed by a one-time ~46% effective tax rate (litigation/tax settlement); normalize out for run-rate.

Operating leverage — why EPS outgrows revenue. The structural reason ResMed’s earnings compound faster than its top line is incremental operating margin: in FY25, incremental operating margin ran ~66% (every dollar of new revenue dropped ~66 cents to operating profit), versus a ~33% average operating margin. That gap is the signature of a business with a largely fixed R&D/SG&A base and a high-margin incremental product (mask resupply) — it is the mechanism behind management’s “earnings growth higher than revenue growth” promise, and it is evidenced, not promised: operating margin expanded from 27% (FY23) to 33% (FY25) to ~37% non-GAAP (Q3 FY26) while revenue grew. As long as the recurring resupply mix keeps rising and the cost base grows slower than revenue, mid-single-digit-to-HSD revenue growth converts to low-double-digit-plus EPS growth — the compounding engine the current multiple is failing to price.

Returns on capital — the moat made visible. ROIC of ~22% (FY25) against a weighted cost of capital in the high-single digits is the clearest evidence of durable advantage. Crucially, this is not a goodwill-flattered or goodwill-distorted figure: ResMed carries ~$3.0B goodwill + ~$0.4B intangibles against ~$8.8B assets, and even on invested capital including acquired goodwill it earns ~22%; ex-goodwill returns are higher still. This is genuine economic value creation, and it has been remarkably stable through a decade of cycles (excluding the FY21 tax anomaly).

Cash generation. Free cash flow was $1.65B in FY25 with >100% conversion of net income; Q3 FY26 generated $520M of FCF, again >100% conversion. The FY22–23 dip in FCF ($195M / $559M) was a deliberate inventory build during the supply-chain crisis (a working-capital timing effect, not an earnings-quality problem) and has fully normalized — net income to operating cash flow now tracks ~1.0–1.25×. There is no divergence between accounting earnings and cash; if anything, cash exceeds earnings.

Working-capital read (the FY22–23 inventory build). A disciplined reader should not miss why FCF cratered to $195M in FY22 and $559M in FY23 against net income of $779M and $898M. The cause was a deliberate inventory build — ResMed front-loaded components and finished goods during the semiconductor shortage to guarantee supply while Philips was out of the market (a rational, offensive move to capture share it could not otherwise have served). Inventory rose ~$300M+ in FY22 alone. As supply normalized that working capital reversed: FY24 saw a ~$172M inventory release and FY25 returned >100% FCF conversion ($1.65B FCF on $1.40B net income). The lesson: the FCF “weakness” was a timing artifact of a share-grab, not an earnings-quality problem — and it is now fully unwound. The cash-conversion cycle remains elevated (~175 days) by the nature of a global device/mask manufacturer carrying safety stock, but that is structural, not deteriorating.

Segment economics. The Sleep & Breathing Health segment (~88% of revenue) carries the company’s high device-plus-mask margins and is the profit engine. Residential Care Software (~12%) is a subscription business with its own margin profile but is currently the drag on growth (+4% cc), held back by the senior-living/long-term-care vertical even as MEDIFOX DAN performs well. Investors should size RCS correctly: at ~$620M of revenue growing low-single-digits, it is neither a needle-mover to the upside nor a thesis risk to the downside; the FY27 re-acceleration target (HSD revenue, double-digit operating profit) is an option, not a load-bearing assumption. The tax line is a modest, recurring headwind — the effective rate steps to ~21–23% in FY26 from prior high-teens on global-minimum-tax legislation (a ~1–2 point EPS drag), partly offset by a 10-year SGD/USD net-investment hedge adding ~$9M/quarter of net interest income.

Balance sheet. Fortress. As of March 2026: cash $1.66B, gross debt $664M, net cash ~$1.0B, net-debt/EBITDA −0.28×, current ratio 3.0×, interest coverage >30×. The company de-levered from $1.58B gross debt in FY23 (MEDIFOX DAN financing) to $664M today. There are no off-balance-sheet liabilities of concern; operating leases are modest and disclosed. The pending Noctrix acquisition ($340M cash) is easily absorbed without disturbing the net-cash position.

Dilution / SBC. Stock-based comp is modest (~$92M FY25, ~1.8% of revenue) and more than offset by buybacks — share count has drifted down (146.9M FY24 → 145.0M Mar-26). Accounting is conservative; the gap between GAAP and non-GAAP EPS is small (acquired-intangible amortization plus occasional deal/portfolio-review costs, on the order of $0.30–$0.50/share), unlike the large adjustment stacks common in roll-up medtech.

Verdict: economics improve with scale, and the quality is real. Rising margins, ~22% ROIC, >100% FCF conversion, net cash, falling share count, and minimal accounting noise. This is among the cleaner, higher-quality financial profiles in medical devices — which is precisely what makes the sub-market valuation notable.

7. Capital Allocation

ResMed’s capital allocation is conservative, returns-aware in practice (if not in its comp design), and has created value — with two governance caveats worth flagging.

Reinvestment first. Management’s stated first priority is organic reinvestment — R&D at ~6–7% of revenue and sales/marketing to fund demand-generation and the ecosystem. Given a ~22% ROIC, reinvesting in the core is the highest-return use of capital, and the financials show it compounding.

M&A — disciplined bolt-ons in true adjacencies. ResMed has avoided large, dilutive, “diworsification” deals. The track record: Brightree (HME software, ~$800M, 2016), MatrixCare (care software), MEDIFOX DAN (German care software, ~$1B, FY22, debt-funded then rapidly repaid), home-sleep-testing tuck-ins (Ectosense/NightOwl, Somnoware, VirtualOx/VertuOx), and most recently Noctrix Health ($340M, announced April 2026, closing ~June 1, 2026) — a restless-leg-syndrome neurostimulation device (NIDRA, FDA de-novo) with ~$24M run-rate revenue, higher growth and higher gross margin than ResMed, flowing through the same sleep-physician/HME channel. The deals share a logic: adjacencies reachable through the existing channel and clinical relationships, bought at sizes that don’t threaten the balance sheet. No goodwill impairments of note. This is intelligent, channel-leveraged M&A — though investors should watch that the cadence of software deals (RCS, now growing only ~4%) earns its keep.

Shareholder returns — growing and self-funded. A $0.60/quarter dividend ($2.40/yr, ~1.2% yield, ~22% payout) raised regularly, plus a re-accelerated buyback — $175M/quarter recently ($262M total return in Q3 FY26 including the dividend), with a commitment to repurchase at least $175M/quarter through CY2026. Buybacks are funded entirely from net-cash FCF, not leverage. The buyback at ~17–19× earnings on a 22%-ROIC business is value-accretive; that management picked up the pace into the de-rating is a constructive signal.

Compensation and incentives — the governance caveat. The annual cash bonus is tied to adjusted net sales and adjusted operating profit (50/50); long-term PSUs are tied to absolute and relative total shareholder return. Return-on-invested-capital appears only in the plan’s menu of available metrics — it is not an actual performance metric. For a business whose entire investment case rests on durable high returns on capital, the absence of a returns-based incentive is a real (if common) weakness; it biases management toward growth and TSR over capital efficiency. The relative-TSR PSU is a partial mitigant. The other governance flag: Mick Farrell serves as combined Chairman and CEO (son of founder Peter Farrell), concentrating board leadership.

CFO transition. Long-time CFO Brett Sandercock (26 years, 55 quarters partnering CEO Farrell) is retiring, replaced by Aaron Blumer (most recently CFO of Exact Sciences; prior senior finance roles at 3M and Baxter). The transition is orderly and well-telegraphed; Blumer brings large-cap medtech and diagnostics experience. A new CFO is a modest execution risk and worth monitoring for any change in capital-return posture.

Insider behavior. Recent Form 4 activity (2026) is routine — option exercises and grant-related/10b5-1 sales paired with Rule 144 notices — with no discretionary open-market purchases observed. Neutral-to-slightly-soft signal; no insider conviction-buying into the de-rating, but also no alarming selling beyond the ordinary for a long-tenured team.

M&A scorecard (Marathon discipline lens). Judged by the capital-cycle test — does management deploy capital counter-cyclically and in its circle of competence, or chase asset growth at the top? — ResMed scores well. Its deals cluster in two buckets: (1) channel-adjacent clinical hardware/diagnostics it can scale through its existing sleep-physician/HME relationships (home sleep testing, now RLS via Noctrix) — high-confidence, on-strategy, small; and (2) care software (Brightree, MatrixCare, MEDIFOX DAN) that deepens channel control. The one area warranting scrutiny is bucket (2): RCS is now the slowest-growing part of the company, raising the fair question of whether the ~$2B+ cumulatively deployed into care software has earned its cost of capital. Management’s FY27 reacceleration plan and “disciplined portfolio management” language (code for pruning the laggard verticals) suggest they recognize this. Critically, ResMed has avoided the value-destroying move the capital-cycle warns against — a large, debt-funded, top-of-cycle transformative acquisition — and has instead kept M&A small enough that no single deal can impair the balance sheet. The MEDIFOX DAN debt was raised and repaid within ~2 years; net cash was restored quickly.

Verdict: capital has been allocated intelligently — high-return reinvestment, sane channel-leveraged bolt-ons, growing self-funded returns, conservative balance sheet — with the genuine caveats that the comp plan lacks a returns metric, the Chairman/CEO roles are combined, and the care-software bucket of M&A has yet to prove its returns.

8. Changes and Headwinds — Last Two Years

Strategic / corporate.

• CFO succession (April 2026): Sandercock → Blumer (ex-Exact Sciences). Orderly; the single most important recent personnel change.
• Noctrix Health acquisition ($340M, April 2026, closing ~June 1, 2026): entry into restless-leg syndrome — a new, adjacent, channel-aligned growth vector.
• Segment re-naming / “2030 strategy” (FY25): Sleep & Respiratory Care → Sleep & Breathing Health; SaaS → Residential Care Software. Presentation, not substance — but signals a strategic framing toward a broader “digital health ecosystem.”
• Buyback re-acceleration (FY26): from minimal to ≥$175M/quarter, into the de-rating.
• Product cycle: fabric-mask launches (AirTouch N30i, F30i Comfort/Clear) with measured adherence gains; AirSense 11 rollout to Latin America and China.

Competitive / regulatory.

• Philips remains largely absent from the US under its consent decree; signaling possible re-entry but “nothing new” per management. The key multi-year watch item.
• GLP-1 escalation: Zepbound’s OSA approval (Dec 2024) and Lilly’s dontsleeponosa.com DTC campaign; oral-GLP-1 rollout in 2025–26 — the proximate cause of the de-rating, but framed by management (with data) as demand-gen.
• Payer / funding-model risk (Synapse / utilization management): an emerging, lower-probability threat to the DME funding model; management downplays given OSA’s low-cost/high-ROI profile.
• Tariffs / geopolitical / FX: management notes fuel-cost and component-cost pressures and global-minimum-tax-driven tax-rate increase (effective rate to ~21–23% FY26 from prior high-teens), partly offset by supply-chain productivity and a 10-year SGD/USD net-investment hedge.

Headwinds.

• The dominant headwind is sentiment/multiple, not operations: the stock de-rated ~31% while fundamentals grew double digits.
• RCS underperformance (~4% growth) is a real, if small (12% of revenue), operational soft spot.
• Tax-rate step-up (global minimum tax) is a modest EPS headwind.

Verdict: on balance, these developments strengthen the thesis operationally while weakening it on sentiment. The strategic moves (Noctrix, buyback, fabric masks, China) are constructive; the CFO change is well-managed; the GLP-1 escalation is the source of the de-rating but is, on evidence, a tailwind. The one genuine operational watch item is RCS; the one genuine competitive watch item is a Philips return.

9. Risk Analysis (Risk Matrix)

ResMed’s risk profile is unusual: the business-execution risks are low (durable demand, net cash, high returns, no customer concentration of note), while a single exogenous risk — pharmaceutical substitution — dominates the entire debate and the valuation. The matrix below separates the near-term, evidence-testable version of each risk from the long-tail version, because conflating the two is precisely the error the market is making on GLP-1: pricing a low-probability/long-dated scenario as if it were imminent. The honest framing is that ResMed carries a low-probability, high-impact, long-dated terminal risk wrapped around a high-quality, low-volatility core — the inverse of a typical “cheap” stock, which usually pairs a fragile business with a near-term catalyst.

Catastrophic-loss assessment: negligible. Net-cash balance sheet, an essential chronic therapy with durable reimbursement, geographic diversification, and >100% cash conversion make a permanent capital impairment extremely unlikely barring a sudden, total pharmaceutical substitution that no evidence supports. The realistic downside is opportunity cost / sentiment (multiple stays low while the business compounds), not impairment.

10. Valuation Discussion (Embedded Expectations)

This memo states no price target. The purpose here is to make the market’s embedded expectations explicit and to test them against the evidence.

Where the multiple sits. At $200.29, ResMed trades at:

• ~19.3× trailing GAAP EPS ($10.34) and ~17–18× forward earnings (FY27 consensus EPS ~$11+);
• ~12.8× EV/EBITDA (TTM EBITDA $2.18B, EV ~$28B) and ~5.0× EV/Sales;
• ~1.2% dividend yield, ~22% payout, on a 22% ROIC.

On its own history, this is extraordinary: A decade-history valuation screen places ResMed’s composite multiple in the 7th percentile of its decade-long range, with the P/E in the 0.8th percentile (i.e., essentially the cheapest it has ever traded) and P/B in the 2.8th. Over the past decade RMD’s trailing P/E has ranged from the mid-20s to the 70s; ~19× is a generational low for the franchise. The de-rating is almost entirely multiple, not earnings: EPS has grown every year while the multiple roughly halved from its 2021–23 peak.

Versus peers. Against high-quality, single-product-category medtech compounders that trade at mid-20s to 30s+ forward P/E and mid-to-high-teens-plus EV/EBITDA, ResMed at ~17× forward / ~12.8× EV/EBITDA screens cheap on both an absolute-quality and relative basis, with comparable or superior ROIC and a cleaner balance sheet:

The discount is entirely the GLP-1 overhang. On every fundamental metric — margin durability, ROIC, cash conversion, recurring mix, balance-sheet quality — ResMed merits a premium-medtech multiple; it instead carries a sub-market one. The gap between “should trade at low-to-mid-20s P/E like its quality peers” and “trades at ~17×” is the entire investment debate expressed in one number.

Inspire as a read-through. That Inspire Medical (INSP) — the surgical hypoglossal-nerve-stimulation alternative, a far smaller and less profitable business — has historically commanded a richer growth multiple than ResMed underscores how much the market has discounted ResMed’s durability rather than its quality. The market is paying up for the OSA-treatment growth story in one vehicle while marking down the OSA-treatment cash-machine in another, on the same underlying disease. That is a valuation incoherence the bull case exploits.

Reverse-DCF / embedded expectations. At ~17× forward earnings with a high-single-digit grower, the market is implicitly underwriting one of two things: (a) a sharp deceleration of growth toward low-single-digits and/or margin reversal within a few years, or (b) a terminal-value haircut — i.e., a belief that GLP-1s structurally cap the TAM in the 2030s, pulling forward a lower terminal growth rate. A simple sanity check: a business growing high-single-digit revenue with EPS growing low-double-digits, ~22% ROIC, and >100% cash conversion would conventionally support a low-to-mid-20s multiple. The ~6–8 turns of P/E compression versus that “fair” level is the price the market is charging for the GLP-1 tail risk and growth-deceleration fear.

Put more explicitly: discount ResMed’s ~$2.0B of free cash flow forward at a ~9% cost of equity. To justify only ~$28B of enterprise value (the current level), one must assume FCF growth fades to ~low-single-digits within roughly five years and a terminal growth rate of ~1–2% — i.e., a business that essentially stops compounding shortly after the GLP-1 cohorts mature. To justify a “fair” ~$36–40B EV (a low-to-mid-20s P/E), one needs only ~6–8% FCF growth tapering to a ~3% terminal rate — squarely within management’s stated algorithm and well below the company’s historical growth. The reverse-DCF therefore frames the bet cleanly: the current price embeds a near-term growth collapse; the base case embeds the company merely doing what it has done for a decade, more slowly. The burden of proof the market has assigned to the bull is light; the burden it has assigned to the bear (prove the collapse) is heavy and, so far, unmet by data.

Scenario framework (illustrative; not targets):

• Bear: GLP-1 erosion proves real and near-dated; growth decelerates to low-single-digits, multiple stays ~14–16× depressed earnings → equity value materially below current (a high-$100s-to-low-$200s “value trap on narrative,” or worse if the TAM thesis breaks). The market is partly pricing this today.
• Base: management’s 2030 algorithm holds — high-single-digit revenue, low-double-digit EPS, annual margin accretion — and the multiple normalizes only partway (to ~20–22× as the GLP-1 fear fades). Forward EPS ~$11 × 20–22× → meaningfully above the current price.
• Bull: GLP-1 is confirmed as a durable demand-gen tailwind (simultaneous-prescription data turns positive), penetration of the untreated pool accelerates, RCS reaccelerates, and the multiple re-rates toward its quality-medtech peers (24–26×) → substantial upside on ~$11–12 EPS.

Verdict: The embedded expectation is pessimistic relative to the evidence. The market is underwriting a near-dated GLP-1 disruption and growth deceleration that the company’s own data and current results contradict; it is correctly pricing a non-zero long-tail substitution risk and a more-normal (post-Philips-windfall) growth cadence. The asymmetry, on the framework above, is favorable: the bear is partly in the price, while base and bull require only that the disruption not arrive on the market’s feared timeline.

11. Variant Perception

Consensus belief. The sell-side and market consensus has turned cautious-to-neutral: ResMed is a good business facing a structural, pharma-driven TAM threat that justifies a permanently lower multiple. The Morgan Stanley downgrade to Equal-Weight (PT $230, June 2026) crystallizes the “good company, capped upside, GLP-1 overhang” view; KeyBanc’s maintained Overweight (PT $266) is the dissenting constructive voice. Notably, both analyst price targets sit above the current ~$200 — consensus is cautious, not outright bearish, and the price has overshot even the cautious targets. This is itself a tell: when the marginal seller is pushing a stock below the cautious sell-side fair value, the selling is being driven by narrative and positioning (de-risking a “GLP-1 loser”) rather than by a fundamentals-based price target. The variant-perception opportunity sits in that gap — between a price set by narrative-driven flow and a value set by the cash flows the business is actually producing.

The strongest bull case. ResMed is a ~22%-ROIC, net-cash, double-digit compounder with an ~80%-untreated billion-person TAM, trading at the cheapest multiple in its history because of a disruption its own 2.1-million-patient data identifies as a tailwind. GLP-1s generate free demand, raise adherence and resupply, and (per Lilly’s own messaging and trials) work “better together” with CPAP; OSA’s multi-factorial etiology means it persists after weight loss; and the expanding OSA–dementia/Parkinson’s evidence broadens the medical and payer rationale for treatment. Mask/resupply — the high-margin annuity — is still compounding 12–14%. As the GLP-1 fear ages without showing up in the numbers, the multiple re-rates toward quality-medtech norms and EPS compounds underneath it: a double-barreled return.

The strongest bear case. GLP-1 drugs are a genuine, novel substitution technology, and the market is right to assign a terminal-value haircut. The current “tailwind” data reflect the early phase (newly-diagnosed, GLP-1-prompted patients still need CPAP); the late phase — patients who lose enough weight to resolve OSA and discontinue therapy, plus a generation that never develops weight-driven OSA — pressures the installed base and resupply annuity in the 2030s. Growth has already decelerated off the Philips windfall; RCS is stalling; a Philips re-entry could pressure device share and pricing; and a high-multiple stock with a credible long-tail threat can stay de-rated for years. Buying the dip risks a value trap on a real secular concern.

The 3–5 assumptions that matter most:

1. Does GLP-1 net-add or net-subtract patients/resupply over the next 3–5 years? (Bull: net-add, per RMD data. Bear: net-subtract eventually.)
2. Does the mask/resupply annuity keep compounding double digits? (The single most important financial KPI; currently yes.)
3. Does Philips meaningfully re-enter and take US share? (Currently no.)
4. Does the multiple normalize, or does the GLP-1 narrative permanently cap it? (The return-driver if the business performs.)
5. Does management’s 2030 algorithm — HSD revenue, EPS > revenue, annual margin accretion — hold?

Falsification evidence. Bull case falsified by: a quarter showing US new-patient starts or resupply rates decelerating as GLP-1 penetration rises; the “simultaneous-prescription” cohort analysis turning negative; a credible Philips share recapture. Bear case falsified by: continued double-digit mask/resupply growth alongside rising GLP-1 penetration; the simultaneous-prescription data confirming net-positive new-starts; multiple re-rating as the narrative ages.

The factor-positioning read. A factor-risk model characterizes RMD as high-Quality (+0.38), low-beta/defensive (BetaFactor −0.20), negative-Momentum (−0.29), with no Value loading and only an 18% R² — i.e., an idiosyncratic, story-driven, out-of-favor quality name, not a cyclical or a value play. The risk-adjusted track record is dead-money-to-negative over 1/3/5 years (y1 −21%, m6/m3 deeply negative annualized) with a lifetime −54% max drawdown. This is the empirical signature of a falling knife into accelerating numbers: the tape confirms consensus has abandoned the stock on the narrative, which is exactly the setup in which a quality franchise becomes mispriced — and the setup that can persist longer than fundamentals justify. The positioning supports the contrarian thesis without guaranteeing its timing.

12. Fact vs. Interpretation Table

13. Open Questions

1. The “simultaneous-prescription” cohort. Management flagged a forthcoming analysis of patients prescribed PAP and GLP-1 at the same time. Does it confirm net-positive new-starts, or reveal substitution? This is the single highest-information-value upcoming data point.
2. Late-phase GLP-1 dynamics. What happens to resupply rates 3–5+ years into GLP-1 therapy, once meaningful weight loss has occurred? The current data are necessarily early-phase.
3. RCS trajectory. Does the software portfolio actually re-accelerate to HSD revenue / double-digit operating profit in FY27, or is it a structurally low-growth drag warranting divestiture?
4. Philips. Timing, scale, and product credibility of any US re-entry, and ResMed’s pricing/share response.
5. New CFO posture. Does Aaron Blumer maintain the conservative, net-cash, growing-buyback capital-allocation framework, or shift it (larger M&A, leverage)?
6. Margin ceiling. How much further can gross margin accrete (62–63% guided FY26; multi-year accretion claimed) against component/fuel/tariff pressure?
7. Payer utilization management. Does Synapse-type DME utilization management spread, and does it dent volumes or pricing?

14. What Must Be True

Bull case — what must be true:

1. The mask/resupply annuity keeps compounding ~double-digits (installed base + share gains + fabric-mask premiumization).
2. GLP-1 remains net-additive to patient flow and adherence through the late-2020s (the demand-gen and “better-together” thesis holds in the data).
3. Management’s 2030 algorithm delivers — HSD revenue, EPS growth > revenue growth, annual gross-margin accretion.
4. The multiple normalizes at least partway toward quality-medtech norms as the GLP-1 fear ages without showing in results.

Falsification test (bull): A single clean quarter in which US new-patient starts or resupply rates decelerate as GLP-1 penetration rises — i.e., the first hard evidence of cannibalization in the numbers — would break the near-term bull thesis. Likewise, the forthcoming simultaneous-prescription analysis showing net-negative new-starts.

Bear case — what must be true:

1. GLP-1 (especially scaled oral agents) durably resolves enough OSA to shrink the diagnosed/treated pool and pressure the resupply annuity within the investable horizon.
2. The early “tailwind” data prove to be a transient, early-phase artifact that reverses as cohorts mature.
3. Growth deceleration and/or a Philips re-entry compress both volume and the multiple, keeping the stock de-rated for years (value trap on a real secular concern).

Falsification test (bear): Continued double-digit mask/resupply growth alongside rising GLP-1 penetration over the next 4–6 quarters — plus the simultaneous-prescription data confirming net-positive new-starts — would falsify the near/medium-term bear case and leave only the speculative 2030s-TAM argument, which a ~17× forward multiple already over-discounts.

15. Source Appendix

Primary sources relied upon:

• ResMed Inc. FY2025 Form 10-K (filed 2025-08-08); Q3 FY2026 Form 10-Q (filed 2026-05-01); prior 10-Ks FY2021–FY2024; 10-Qs through Q3 FY2026 — SEC EDGAR (CIK 0000943819).
• ResMed Q3 FY2026 earnings call transcript (2026-04-30).
• ResMed DEF 14A proxy statement (filed 2025-10-02) — compensation, governance, board.
• Multi-year financial statements, profitability/credit ratios, enterprise value, and valuation multiples (FY2018–FY2025 annual + quarterly to Q3 FY2026), from company filings and standard market-data aggregators.
• Own-history valuation percentile screen; 5-year daily price/OHLCV history.
• Quantitative factor-risk model — factor loadings, risk-adjusted track record, relative strength, related stocks.
• Public analyst actions: Morgan Stanley (downgrade to Equal-Weight, PT $230, 2026-06-17); KeyBanc (maintains Overweight, PT $266, 2026-06-22) — via news feed; cited as third-party signals, not evidence.

No BUY/SELL recommendation and no price target appears in sections 1–15; the only position-taking content is the clearly-labeled “Claude’s Take” block, which is the author’s own independent opinion. This article is general information, not investment advice. The author may hold positions in securities mentioned. Do your own research.

APPENDIX A — Standard Diligence Questionnaire

ResMed Inc. (NYSE: RMD) — as of June 28, 2026

Supplemental to the research memo. Fact / Interpretation / Assumption labels applied where it matters.

General

What thoughtful questions have other investors asked about this company? The dominant question by far: will GLP-1 weight-loss drugs shrink the obstructive-sleep-apnea TAM and the CPAP installed base? Secondary questions: (a) how durable is the post-Philips-recall growth and share, and what happens if Philips re-enters the US?; (b) is the mask/resupply annuity’s double-digit growth sustainable?; © why does a 22%-ROIC compounder trade at the cheapest multiple in its history — is it a value trap or a gift?; (d) does the residential-care software (RCS) segment matter, and why is it stalling?; (e) how much further can gross margin accrete? The bull/bear debate is unusually binary because it hinges on a single exogenous variable (pharma) rather than on execution.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Neither — ResMed is a secular grower, not a cyclical. Margins are at a cyclical high (gross margin recovered from a FY24 freight/component trough of 56.7% to 62%+), but volume growth is in a normalized (post-Philips-windfall) phase, decelerated from the >15% recall years to high-single-digit/low-teens. Interpretation: earnings power is closer to mid-cycle than peak, with margin at the high end and growth normalized. Driven by external environment or internal actions? Both. Internal: share gains, fabric-mask premiumization, supply-chain productivity, demand-gen investment. External: the Philips recall windfall (now annualized), FX, GLP-1 demand-generation, reimbursement. How stable are revenues? Very. ~40%+ of revenue is recurring (mask resupply + software); the rest is an installed-base-driven device replacement/new-start stream. Through COVID, supply-chain crises, and two GLP-1 panics, revenue has grown every year. Outlook for products/services? Positive: an ~80%-untreated billion-person OSA pool, awareness tailwinds (wearables, GLP-1 demand-gen, OSA–dementia/Parkinson’s evidence), fabric masks, China/LatAm rollout, and a new RLS adjacency (Noctrix). The lone product risk is long-tail pharmaceutical substitution. How big will this market be — growing, shrinking, domestic, international? Growing mid-to-high-single-digits structurally; ~58% Americas / ~42% international, with international and China the higher-growth geographies. Assumption: GLP-1 does not durably shrink the TAM within the investable horizon.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Less on the device side (two-player duopoly that became a US near-monopoly after Philips’s recall/exit), with a new pharma-adjacent competitive vector (GLP-1) that is presently complementary. Mask competition exists but ResMed is differentiating (fabric). How profitable is the business (ROIC, ROE)? Very: ROIC ~22%, ROE ~25%, both well above cost of capital and stable for a decade (ex the FY21 tax anomaly). Fact. How profitable is the industry — competitors, barriers? High barriers (regulatory, clinical-evidence, scale, channel, installed-base lock-in); few competitors (Philips, niche Inspire). Industry profit pool is healthy and concentrated in ResMed. Can the business be easily understood? Yes — razor/razor-blade medical devices with a software overlay. The valuation debate (GLP-1) is harder than the business. Can it be undermined by foreign low-cost labor? No — regulated medical devices with clinical-evidence, IP, and channel moats; not a labor-cost-competed commodity. Do brands matter? Yes, among sleep physicians, HME/DME providers, and increasingly consumers (China/cash markets); the ResMed brand + clinical evidence is a real intangible. Nature of competition? Innovation (device features, mask comfort/adherence, software/ecosystem), clinical evidence, and channel relationships — not primarily price. Customers’ switching costs? High for clinicians/HMEs embedded in AirView/Brightree workflows and compliance data; moderate-and-rising for patients (myAir, mask fit, resupply convenience).

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Yes — the installed base of flow generators (the source of the resupply annuity), the clinical-evidence base, the AirView/Brightree data network, and brand are economically valuable but not capitalized. Tangible book understates economic value. Off-balance-sheet liabilities? None material; operating leases modest and disclosed. No pension/litigation overhang of note (the historical tax/litigation item that hit FY21 is resolved). How conservative is the accounting? Conservative. Small gap between GAAP and non-GAAP EPS (acquired-intangible amortization + occasional deal costs, ~$0.30–0.50/sh); >100% FCF conversion; no aggressive revenue recognition or capitalization. Cash exceeds earnings. How CapEx-hungry? Light — CapEx ~2% of revenue (~$35M/quarter); FCF conversion >100%. An asset-light, high-return model.

Capital Allocation & Management

How much FCF, and how is it used? ~$1.65B FCF FY25. Priority order: (1) organic reinvestment (R&D ~6–7%, S&M); (2) bolt-on M&A in channel-aligned adjacencies; (3) growing dividend (~$2.40/yr, ~22% payout); (4) re-accelerated buyback (≥$175M/quarter). All self-funded from net-cash FCF. Significant acquisitions recently? Noctrix Health ($340M, RLS, closing ~June 2026); VertuOx (home sleep testing, FY25); MEDIFOX DAN (~$1B German care software, FY22). All bolt-ons; no transformative/dilutive deals. Buying back shares? Yes, re-accelerated; share count drifting down (146.9M → 145.0M). Fact. Issuing large amounts of stock to insiders? No — SBC modest (~1.8% of revenue), more than offset by buybacks. Compensation policy? Cash bonus on adjusted net sales + adjusted operating profit (50/50); PSUs on absolute + relative TSR. Caveat (Fact): no explicit ROIC/returns metric — a governance weakness for a high-ROIC franchise. Motivations of management? Long-tenured, founder-family-led (Mick Farrell, Chairman & CEO, son of founder Peter Farrell); mission-driven culture (“improving lives”). Combined Chair/CEO concentrates leadership; orderly CFO succession (Sandercock → Blumer, ex-Exact Sciences) underway.

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — US C-corp, common stock, NYSE-listed (also CHESS Depositary Interests on the ASX historically; the primary listing is NYSE). Standard 1099 dividend; no K-1. Dividend policy? Quarterly cash dividend ($0.60/qtr, ~1.2% yield), raised regularly, ~22% payout — conservative and growing. How profitable is the business? Among the most profitable in medtech: ~27% net margin, ~22% ROIC, ~37% EBITDA margin. Is net income diverging from cash from operations? No — operating cash flow ≥ net income (>100% FCF conversion); the FY22–23 inventory build (a deliberate supply-chain hedge) has normalized.

Risks & Downside

What factors would cause the stock to decline? (1) Evidence of GLP-1 cannibalization in the numbers (new-starts/resupply decelerating); (2) a credible Philips US re-entry; (3) margin reversal (component/fuel/tariff); (4) RCS continued stall; (5) the GLP-1 narrative simply persisting and keeping the multiple compressed regardless of results; (6) a misstep in the CFO transition or a shift to aggressive M&A. Risk of a catastrophic loss? Very low — net cash, essential chronic therapy, durable reimbursement, geographic diversification, >100% cash conversion. The realistic downside is multiple-stays-low / opportunity cost, not impairment. Chance of a total loss? Negligible. No solvency path; would require a sudden, total pharmaceutical substitution unsupported by any current evidence.

Recent News & Events

Has the business environment changed recently? Operationally, no — Q3 FY26 (Apr 30, 2026) delivered +11% revenue, +21% non-GAAP EPS, +290 bps gross margin. Sentiment changed sharply: the stock de-rated ~31% from its Aug-2025 record on renewed (oral-)GLP-1 fears and a healthcare-sector de-rating; Morgan Stanley cut to Equal-Weight (PT $230, Jun 2026), KeyBanc maintained Overweight (PT $266). Significant acquisitions? Noctrix Health ($340M, RLS), announced April 2026, closing ~June 1, 2026. Change in accounting policies? None material; segment re-naming (Sleep & Respiratory Care → Sleep & Breathing Health; SaaS → Residential Care Software) with no change to reported segment financials. Effective tax rate stepping up to ~21–23% on global minimum tax. Recent changes — new markets, facilities, management? AirSense 11 launched in China and Latin America; fabric-mask launches (N30i, F30i Comfort/Clear); CFO succession (Sandercock retiring → Aaron Blumer); buyback re-accelerated; a 10-year SGD/USD net-investment hedge executed (Feb 2026).

APPENDIX B — Source Appendix

ResMed Inc. (NYSE: RMD) — Research as of June 28, 2026

All figures reconciled to primary filings where possible. SEC filings are primary and authoritative; market-data figures are cross-checks. Management commentary (transcripts) is treated as hypothesis, validated against filings and data.

Primary — SEC filings (EDGAR, CIK 0000943819)

1. FY2025 Form 10-K — filed 2025-08-08 (period ended 2025-06-30). Segment revenue (Sleep & Breathing Health vs Residential Care Software), product revenue (devices vs masks & other), geographic split, competition, regulation, risk factors. https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000943819
2. Q3 FY2026 Form 10-Q — filed 2026-05-01 (period ended 2026-03-31). Quarterly revenue, margins, balance sheet (cash $1.66B, gross debt $664M, net cash ~$1.0B), share count (~145.0M).
3. Form 10-K FY2021–FY2024 — filed 2021-08-17, 2022-08-12, 2023-08-11, 2024-08-09. Multi-year revenue/margin/returns history.
4. Form 10-Q (all quarters FY2022–Q3 FY2026) — quarterly trend, gross-margin recovery, segment performance.
5. DEF 14A proxy statement — filed 2025-10-02. Executive compensation (annual cash bonus on adjusted net sales + adjusted operating profit; PSUs on absolute + relative TSR), board composition, combined Chairman/CEO structure.
6. Form 8-K — 2026-04-30 (Q3 FY26 earnings release + CFO transition announcement); buyback authorization disclosures.
7. Form 4 / Rule 144 — 2026 insider filings: routine grants/option exercises and 10b5-1/144-paired sales; no discretionary open-market purchases observed.

Primary — Management commentary

8. ResMed Q3 FY2026 earnings call transcript — 2026-04-30. Source for: GLP-1 real-world data (2.1M-patient and 1.7M-patient cohorts), Noctrix Health acquisition, fabric-mask adherence (+6% 90-day), gross-margin drivers, Philips commentary, payer/utilization-management commentary, FY26 guidance (GM 62–63%, SG&A 19–20%, R&D 6–7%, tax 21–23%), 2030 algorithm, capital return ($262M in Q3), CFO transition.

Market & valuation data (cross-check; reconciled to filings)

9. Multi-year financial statements, profitability/credit ratios (ROIC ~22%, ROE ~25%, net cash), enterprise value, and valuation multiples derived from company filings and standard market-data aggregators (annual FY2018–FY2025 + quarterly to Q3 FY2026).
10. Own-history valuation context: composite valuation percentile ~7th, P/E ~1st percentile of the stock’s decade range; trailing EPS ~$10.34, price $200.29 (2026-06-26). Five-year daily price/OHLCV history (beta ~0.74).
11. Quantitative factor-risk positioning: high-Quality (+0.38), low-beta/defensive, negative-Momentum (−0.29), no Value loading, R² ~18%; risk-adjusted track record (1-yr ~−21%, lifetime max drawdown ~−54%); idiosyncratic vol ~18.7%.

Third-party — qualitative / signal (cited as signal, not evidence)

12. Analyst actions: Morgan Stanley downgrade to Equal-Weight, PT $230 (2026-06-17); KeyBanc maintains Overweight, PT $266 (2026-06-22).
13. Clinical literature referenced by management (independently verifiable): Thorax (UK ~2M-adult OSA–dementia study); GeroScience meta-analysis (OSA +33% dementia, +45% Alzheimer’s risk); JAMA Neurology (CPAP and Parkinson’s risk reduction); Eli Lilly SURMOUNT-OSA (tirzepatide for OSA; Zepbound OSA approval Dec 2024).

Data notes / caveats

• Enterprise-value snapshots can lag the live price; EV recomputed at spot (~$28.0B) for valuation.
• FY2021 net income/EPS/returns depressed by a one-time ~46% effective tax rate (litigation/tax settlement) — normalized out of run-rate analysis.
• GLP-1 cohort statistics are company-sourced (management); treated as hypothesis pending independent corroboration.