DaVita Inc. (NYSE: DVA) — A Levered Buyback Machine at Its Richest-Ever Price, Just as the Flywheel Loses Torque

Independent equity research note. Prepared 2026-07-04.

⚡ Claude’s Take

This is the author’s own independent opinion and general information only — not investment advice. The analysis that follows takes no position, sets no price target, and carries no BUY/SELL; this block alone states a view, clearly labeled.

Verdict: HOLD / trim-into-strength — a genuinely moaty, cash-gushing dialysis duopoly that has been melted up to its richest-ever valuation just as the engine driving its per-share story loses torque. Not a short. Risk/reward turns attractive again only well below spot — a defensible accumulation zone is ~$170–190 (≈12–13x the ~$14.65 midpoint of 2026 adjusted EPS and ~9x forward EV/EBITDA, its own mid-history multiples), versus ~16x / ~10.7x at the $234.91 all-time high. Conviction: medium.

DaVita is one of the cleanest business models in healthcare — half of a certificate-of-need duopoly (with Fresenius) over a life-sustaining, thrice-weekly, non-discretionary service, with a real Greenwald moat (local scale + physician-JV customer captivity) and ~11% ROIC against an ~8–9% cost of capital. But the operating business barely grows: US treatment volume was −1.1% in 2025, ESKD incidence has fallen ~18.6% over the decade, and management’s own long-term algorithm is 3–7% operating-income growth on a flat-volume base. The equity has nonetheless doubled in the first half of 2026 to an all-time high of $234.91 — an almost entirely idiosyncratic re-rating (FactorsToday R² only 12–23%; this was not a factor wave) that lifted the stock to the 98.5th percentile of its own ten-year price-to-sales history — the richest it has ever been. What the market is capitalizing is the financial engineering: DaVita has retired ~38% of its shares in five years (119.8M → ~64M), turning ~5% adjusted-OI growth into a guided +31% jump in 2026 adjusted EPS to $14.10–$15.20. That is real value creation — but it is a levered equity-shrink (à la AutoZone) on a static cash cow, and its potency halves as the price triples: the same ~$1.0–1.2B of annual free cash flow that retired ~15% of the float at ~$140 now retires only ~7–8% at ~$235.

The framing is therefore not “momentum darling” and not “falling knife” — it is a low-beta value stock in a momentum overshoot. At ~16x forward adjusted EPS the stock is not egregious versus the market; but on every own-history and cash-flow measure (P/S 98.5th percentile, ~24x trailing GAAP P/E, ~10.7x EV/EBITDA at the top of its range, P/FCF re-rated from ~4x to ~13x) you are paying peak price for a business facing a slow three-front squeeze on its commercial cross-subsidy — Medicare Advantage penetration, ACA-subsidy expiry, and the Marietta Supreme Court loss — plus a genuine (if distant) GLP-1 terminal-value risk, on 3.3x leverage with a −$651M book deficit and zero insider open-market buying. The easy re-rate is behind it, not ahead. What would flip me bullish: two-plus consecutive quarters of durable +1–2% organic volume growth with stable commercial mix (GLP-1/MA-mix fears falsified) — evidence the compounding is earned, not borrowed. What would flip me bearish (toward avoid/short-watch): a print showing treatments-per-normalized-day rolling over or revenue-per-treatment compressing on commercial-mix loss, with the multiple still at these levels. Tag: “The share-count shrink-ray meets a price ceiling.”

📈 Stock Price Action — Five-Year Event Map

DaVita has completed a full round-trip and then some over five years: from ~$122 in mid-2021, down to a $65.42 low on 9 Nov 2022 (−46%), grinding back through the low-$100s in 2023–24, to $234.91 at the 2 Jul 2026 close — an all-time high (intraday $236.51). The stock sits at the very top of its 52-week range of ~$101–$236.51 and is 0% off its high, having roughly doubled in the first half of 2026 alone. Price is stretched far above trend — the 200-day EMA is $159 and the 50-day EMA $193 — the signature of a fast, largely idiosyncratic re-rating rather than a broad-market move. (Fact: prices and EMAs from public 5-year daily price history, accessed 2026-07-04.)

Cycle narrative. (1) The 2021–22 collapse was a levered-defensive de-rating: rising rates compressed the multiple while wage inflation lifted cost-per-treatment and the Marietta Memorial v. DaVita ruling stoked reimbursement anxiety — a valuation, not a fundamentals, event. (2) From the Nov-2022 bottom, normalizing patient mortality and a volume recovery, plus a restarted buyback, drove a ~60% rebound. (3) Through 2024 into early 2025 the market rewarded the flywheel — steady revenue-per-treatment gains, cost control, and a rapidly shrinking share count producing double-digit adjusted-EPS growth, with Berkshire’s ~45% stake an ownership anchor. (4) The early-2025 peak unwound on the April-2025 ransomware incident and a cautious 2026 setup dominated by ACA-subsidy-expiry and Medicare-Advantage-mix fears. (5) A further slide to a ~$101 low on 14-Jan-26 marked peak pessimism. (6) The Q4-25/FY25 report in early Feb-2026 beat depressed expectations and reset sentiment. (7) The Q1-26 call (5-May-26) confirmed the turn — a “trilogy” beat (volume, RPT, cost) and a volume-guide raise from flat to +0.25–0.50%. (8) The final leg to the $234.91 ATH is a momentum-style melt-up in a low-beta name, powered by continued buybacks, the Fresenius-closure share-gain story, and a defensive bid — a move whose cause is interpretive but whose magnitude has clearly outrun the ~mid-single-digit growth in the underlying business.

1. Executive Summary

DaVita is the larger half of the US outpatient-dialysis duopoly — 2,657 US centers, ~200,500 US patients, ~28.7M treatments in 2025 — providing life-sustaining renal-replacement therapy to end-stage kidney disease (ESKD) patients three times a week, indefinitely. It is a superb cash annuity: revenue visibility is exceptional, the service is non-discretionary, and the moat is real. But it is not a growth business, and its equity has just been repriced as though it were.

The business. Consolidated FY2025 revenue was $13.64B (+6.4%), ~86% from US dialysis. The economics are a cross-subsidy: ~11% of US dialysis patients (commercial insured) generate ~26% of US dialysis revenue and the overwhelming majority of profit, while the ~68% government-paid majority (57% Medicare/MA) is treated at or below the fixed bundled rate. Every thesis on DaVita is, at bottom, a thesis on the durability of that thin commercial wedge.

The moat is durable but the price is contested. In Greenwald’s taxonomy the advantage is economies of scale plus customer captivity — local clinic density protected by certificate-of-need, and physician joint ventures (~30% of US dialysis revenue) that align the nephrologists who control referrals. It passes the share-stability and above-WACC-return tests (ROIC ~11.1% vs WACC ~8–9%). But the moat protects units, not price, and price is being taxed from three directions: Medicare Advantage penetration of the ESKD population (~25% in 2019 → ~37%+ post-2021 21st-Century-Cures), ACA enhanced-premium-tax-credit expiry (~$40M 2026 / ~$70M 2027 headwind), and the Marietta Supreme Court loss (2022) that lets employer/MA plans cap dialysis reimbursement.

Growth is low-quality; per-share growth is manufactured. US treatment volume was −1.1% in 2025 and is guided ~flat-to-+0.5% in 2026 (roughly half of the raise is Fresenius clinic-closure transfers). Reported growth is price/mix, transitional reimbursement (phosphate binders into the bundle, TDAPA expiring end-2026), lower-margin international M&A, and a decelerating value-based-care arm (IKC, first profitable year 2025 at $22M adj OI). The real engine is capital allocation: ~38% of shares retired in five years (119.8M → ~64M) at ~$1.8B/yr, converting ~5% adjusted-OI growth into a guided +31% 2026 adjusted EPS of $14.10–$15.20. Berkshire Hathaway owns 45.5%, capped by a standstill and selling into DaVita’s buyback to stay under it.

Financial quality is high on cash, modest on returns. Free cash flow (DaVita-defined, after minority distributions) was ~$1.0B; CFO is a clean 2.5x net income. But ~31% of consolidated profit leaks to minority JV partners, ROIC is a positive-but-unspectacular ~2–3-point spread over WACC, the 2025 price/cost “trilogy” spread compressed (RPT +4.7% vs cost/treatment +5.9%), and buybacks have driven book equity to a −$651M deficit (P/B null, ROE meaningless). Leverage is a disciplined 3.34x with the maturity wall termed out to 2030.

Valuation is the crux. At $234.91 (ATH), market cap is ~$15.1B, EV ~$29.5B — ~10.7x EV/EBITDA (top of its six-year range), ~24x trailing GAAP P/E, ~16x forward adjusted EPS, and P/S at the 98.5th percentile of its own history (richest-ever). A reverse-DCF says ~$235 underwrites upper-end OI growth, benign MA-mix and ACA outcomes, no GLP-1 volume erosion, and the richest-ever multiple holding — i.e., the ~9x→10.7x re-rate has pulled years of buyback compounding forward into today’s price. The variant view: the market was too bearish on near-term volumes for years and has now over-corrected into paying a peak multiple for a low-growth compounder just as its buyback loses torque. No recommendation and no price target follow in the body; the single labeled opinion is in Claude’s Take above.

2. Business Overview

What DaVita does. DaVita Inc. is the larger of the two companies in the US outpatient-dialysis duopoly. Its core service is life-sustaining renal-replacement therapy — chiefly in-center hemodialysis, plus home modalities (home hemodialysis and peritoneal dialysis) — for patients with end-stage kidney disease (ESKD, formerly ESRD), whose kidneys have failed and who require dialysis roughly three times a week to survive, indefinitely, until transplant or death. As of December 31, 2025 the company operated 2,657 outpatient dialysis centers across 46 states and the District of Columbia, serving approximately 200,500 US patients, and contracted to provide inpatient dialysis in hospitals nationwide. It delivered 28,733,980 dialysis treatments in 2025.

Revenue segmentation. The company reports three segments, but one dominates:

• US dialysis — the overwhelming majority of revenue (~86%, $11.79B) and essentially all consolidated operating income. By revenue, US dialysis breaks down ~76% outpatient in-center, ~18% home-based, ~6% hospital inpatient (acute). FY2025 consolidated revenue was $13.64B, up 6.4% from $12.82B.
• DaVita Integrated Kidney Care (IKC) — the value-based/risk-based care arm, managing total cost of care for ~66,000 patients in risk-based arrangements plus ~9,400 additional patients, contracting with CMS (e.g., the Comprehensive Kidney Care Contracting / CKCC program) and payors for shared-savings and capitated revenue. FY2025 revenue $542M.
• DaVita International — 585 outpatient centers in 14 countries serving ~94,500 patients, FY2025 adjusted operating income $114M (revenue $1,346M). Ancillary lab and pharmacy services support the core.

How it makes money — the per-treatment economics. DaVita is paid per treatment, and the identity of the payor determines whether that treatment is profitable. The payor mix by US dialysis revenue is Medicare and Medicare Advantage 57%, commercial 32%, Medicaid and managed Medicaid 7%, other government ~3%. The economics are radically bifurcated: Medicare pays a fixed, bundled rate that is at or below the fully-allocated cost of care, while commercial-insurance rates run at multiples of the Medicare rate. The consequence is the single most important fact about this business: ~26% of US dialysis patient-service revenue — and a disproportionate share of profit — comes from only ~11% of patients, those with non-hospital commercial insurance. The Medicare-covered majority is treated at or below cost; the commercial minority cross-subsidizes the entire system.

Recurring nature. The revenue is about as recurring as any in healthcare services. ESKD is chronic and, absent transplant, permanent; a dialysis patient represents a multi-year, thrice-weekly annuity. Patients are sticky to a center by geography, nephrologist relationship, and the logistical difficulty of switching a life-critical scheduled treatment. The flip side is that the census only grows through new ESKD incidence and share shifts — both of which have stalled (§5). The business is a high-visibility, low-volume-growth cash annuity whose value is set by rate and mix per treatment, not unit expansion.

Verdict: A dominant, deeply recurring, but ex-growth service annuity whose entire profit rests on a thin, contested commercial-payor wedge. DaVita is the scaled leader in an essential, non-discretionary service with exceptional revenue visibility, but the model is a cross-subsidy in which ~11% of patients carry the economics — making franchise “quality” unusually sensitive to payor-mix erosion rather than to operational execution.

3. Industry Dynamics

Structure: a regulated duopoly over an essential service. US outpatient dialysis is one of the cleanest duopolies in American healthcare. DaVita (~2,657 US centers) and Fresenius Medical Care (~38% share) together operate roughly 80% of US dialysis facilities. The remainder is fragmented independents, hospital units, and a few mid-size chains. The concentration is durable: the industry is capital-, license-, and clinical-staff-intensive, and scale confers real cost advantages in supplies, drug purchasing, and fixed-cost absorption (§4).

Reimbursement: a government-set bundle that pays below cost. Since 2011, Medicare has paid for dialysis under a single bundled ESRD Prospective Payment System (PPS) rate per treatment, folding in the treatment plus most injectable/oral ESRD drugs and labs. The bundle is administratively updated annually and set at or below providers’ cost. Because Medicare (traditional + MA) is 57% of dialysis revenue, the government effectively fixes the price on the majority of units and leaves providers to make their margin on the ~32% commercial slice — a structurally fragile profit architecture with high operating leverage to any policy change touching the commercial book.

Epidemiology: the growth engine has stalled. The bull case for dialysis was always demographic — an aging, diabetic, hypertensive, obese population feeding a rising ESKD census. That tailwind has faded. Adjusted ESRD incidence has fallen ~18.6% over the past decade; ~131,564 people started ESRD treatment in 2023, roughly flat YoY (USRDS 2024). Prevalence stabilized in 2023 — but from lower mortality, not more new patients. DaVita’s own numbers confirm the stall: treatments fell 1.1% in 2025, and management’s path back to “at least 2% volume growth” is a clinical project (vaccination, GLP-1 adherence, middle-molecule dialyzers, missed-treatment reduction) not expected to bear fruit until ~2029.

The two policy pincers on the commercial cross-subsidy.

1. Medicare Advantage penetration (21st Century Cures Act, effective 2021). Before 2021, ESKD patients generally could not enroll in MA plans. Cures changed that — ESRD MA penetration jumped from ~25% (2019) to ~37% (end-2021) and keeps rising. MA plans reimburse dialysis below commercial rates and negotiate harder, so the mix-shift is a slow structural drag on revenue-per-treatment. A partial offset: CMS set a +6% ESRD MA rate for 2027 to correct prior underfunding — but the secular direction (more patients paying less than commercial) is unfavorable.
2. Enhanced premium tax credit (ePTC) expiration. The ACA-exchange subsidies that kept many patients in higher-paying commercial marketplace plans are expiring, a headwind DaVita quantifies at ~$40M (2026), ~$70M (2027), ~$10M (2028). Early-2026 enrollment is trending “slightly favorable” but with patients trading down to bronze plans (higher patient-pay, modest RPT drag).

Oral-binder move into the bundle (Jan 2025). Phosphate binders were folded into the ESRD bundle on 1/1/2025 via a Transitional Drug Add-on Payment Adjustment (TDAPA). This added revenue and matching cost (~half of 2025’s per-treatment cost increase; roughly margin-neutral), and the TDAPA is set to expire end-2026. It is a reimbursement-mechanics event, not a durable profit pool.

GLP-1s: a serious long-tail demand threat, correctly weighted as low-urgency. The strongest structural bear argument is that GLP-1 agonists and SGLT2 inhibitors slow the diabetes→CKD→ESKD progression that feeds dialysis. The evidence is real: the FLOW trial cut the composite kidney endpoint 24% (HR 0.76) and was stopped early for efficacy (NEJM 2024). Over a decade-plus horizon, widespread use plausibly bends the ESKD incidence curve further down. But timing defuses the near-term threat — uptake among highest-risk patients is low, ESRD-specific evidence is thin, cost is ~$1,000/mo, and the physiological lag is years. Read it as a genuine terminal-value risk that compresses the multiple, not a 3-year earnings risk.

Barriers to entry are high and stable: certificate-of-need laws restrict new facilities; scale is needed for supply/drug pricing and fixed-cost absorption; nephrologist JV structures are locked up by incumbents; and the below-cost bundle makes greenfield entry unattractive without an existing commercial book. New entry is effectively nil; the competitive action is share-shift between the two incumbents.

Verdict: A structurally decent but no-longer-good industry. The Marathon capital-cycle read is favorable on supply (no new capacity; Fresenius is closing clinics) but unfavorable on demand (declining incidence, GLP-1 terminal risk) and on the price of the units that matter (MA penetration and ePTC expiry eroding the commercial cross-subsidy, with Marietta tilting regulation against providers). This is a mature, regulated, demand-flat oligopoly where incumbents defend value rather than create it.

4. Competitive Position

Name the moat. In Greenwald’s taxonomy, DaVita’s advantage is economies of scale plus customer captivity, expressed locally and reinforced by regulation — not a brand, network-effect, or technology moat. Three mechanisms interlock:

1. Local scale / regulatory density (the primary moat). Dialysis is a local business: a patient goes to the nearest center that fits a thrice-weekly, life-critical schedule. In any metro, the operator with the densest footprint has the lowest cost-per-treatment (fixed-cost absorption, staffing flexibility, logistics) and the strongest position with local nephrologists and hospitals. Certificate-of-need regimes freeze that density in place. DaVita and Fresenius spent two decades assembling these networks; the below-cost Medicare bundle means a sub-scale entrant cannot profitably replicate them. It shows up as above-WACC returns (ROIC ~11%) on an essential service with no viable entrant.
2. Nephrologist joint ventures / MSO alignment (the captivity mechanism). DaVita holds controlling interests in JVs with the nephrologists, hospitals, and physician groups who refer patients, and those JVs represent ~30% of US dialysis revenue (noncontrolling interests are ~16% of US dialysis operating income). By co-owning the economics with referring physicians, DaVita aligns the very people who decide where a new ESKD patient starts dialysis. This is customer captivity engineered through ownership — hard for a rival to dislodge and hard for a regulator to unwind without disrupting patient care.
3. Supply/cost scale. As one of two ~$14B/~$20B-revenue buyers of dialyzers, ESAs, and bundled oral drugs, DaVita commands terms an independent cannot. Fresenius is vertically integrated (it manufactures machines and supplies) — a different flavor of the same advantage.

Pressure-test — durable? On supply, yes. There is no plausible new entrant, capacity is not being added (Fresenius is closing small US clinics and DaVita is capturing the transfers), and the JV structures are entrenched. The share map is stable — DaVita and Fresenius have swapped basis points, not markets, for years. In Greenwald’s market-share-stability test this passes cleanly, a hallmark of a real moat.

But the moat protects units, not price — and price is where it’s taxed. The durable local monopoly lets DaVita keep its patients; it does not let DaVita set the price on the patients that matter. The binding constraint is the payor:

• Marietta Memorial Hospital v. DaVita (SCOTUS, 7-2, 2022) — DaVita lost. The Court held that a group plan may impose limited outpatient-dialysis benefits uniformly without violating the Medicare Secondary Payer Act, even if the facially-neutral limit disproportionately burdens ESRD patients. Practically, this green-lit commercial/employer plans to cap dialysis reimbursement and steer patients toward Medicare — directly attacking the ~26% commercial wedge. The moat cannot defend against a legal change that lets the payer pay less for the same captive patient.
• Third-party premium assistance / American Kidney Fund. A meaningful share of commercial coverage is sustained by charitable premium assistance (notably via the AKF), which preserves the commercial cross-subsidy. Payors and legislators have repeatedly attacked this as steering (e.g., California’s AB-290, in litigation). A standing legal/regulatory risk to the exact revenue that makes DaVita profitable.

DaVita vs. Fresenius. DaVita is the sharper operator; Fresenius the more integrated competitor. DaVita is a US-centric pure-play with a strong commercial book, disciplined cost execution (5-yr total-cost CAGR ~2.6%), and an aggressive buyback. Fresenius is a globally diversified, vertically integrated manufacturer-plus-provider mid-turnaround, rationalizing its US footprint — closing sub-scale clinics whose patients DaVita is now absorbing. The dynamic is not a price war (the duopoly is too rational) but a slow share drift toward the better US operator. DaVita has the stronger US position and is a modest share taker in 2026.

Verdict: A genuine, durable, financially-visible moat that unambiguously protects the patient base but cannot protect the price. Name it — local scale + physician-JV captivity in a CON duopoly. It passes the share-stability and above-WACC tests, and DaVita is the better-positioned incumbent. But the moat’s value is capped by a payor environment that Marietta, MA penetration, and premium-assistance attacks tilt steadily against providers. A strong castle whose tax collector was handed more power.

5. Growth History and Forward Opportunities

Historical growth — flat volume, price-and-buyback-driven earnings. Revenue crept from $11.55B (2020) to $13.64B (2025), a ~3.4% CAGR that flatters the reality: treatment volume is flat-to-declining (US treatments −1.1% in 2025; −0.2% YoY in Q1-26), and essentially all top-line growth is revenue-per-treatment (RPT +4.7% in 2025) plus mix, international M&A, and the 2025 binder-bundle accounting, not units. Organic treatment growth has hovered near zero because ESKD incidence has fallen ~18.6% over the decade and post-COVID mortality thinned the census. Management guides 2026 volume to +25–50 bps — roughly half from Fresenius clinic-closure transfers, not underlying demand — and the path back to “at least 2% volume growth” is a clinical initiative it expects to pay off only around 2029. A long, execution-dependent, unproven bridge.

IKC / value-based care — a real milestone, but decelerating and low-multiple. IKC delivered its first profitable full year in 2025 (adjusted OI $22M, with a $46M Q4) on ~66,000+ risk-based patients, and posted the highest aggregate savings of any CKCC participant. Clinical results are credible (permanent access at start 35% more often, ~3x lower first-180-day cost). But three cautions temper it: (i) improvement is decelerating (+$20M OI guided 2026 vs ~$40–50M/yr prior); (ii) management explicitly does not expect a high-margin business; (iii) earnings are shared-savings/government-program dependent, lumpy and CMS-exposed. A complement that deepens patient capture, not a growth engine that re-rates the stock.

International — steady, but half-bought. International grew adjusted OI to $114M in 2025 (585 centers, 14 countries, ~94,500 patients), including the 2025 acquisition of Fresenius’s Brazil operations (~$94.3M) and prior Latin America deals. Growth is ~half M&A / half organic, with margins improving on fixed-cost leverage. Respectable and consistent, but capital-intensive, acquisition-fed growth in lower-margin geographies is not organic compounding, and it does not move a $13.6B company’s needle.

Home dialysis — a mix goal, not a growth driver. Home modalities are ~18% of US dialysis revenue (~15% of US ESKD patients treated at home), incentivized by CMS models (ETC/KCF). Home shifts modality mix and can improve outcomes/cost, but substitutes for in-center treatment rather than expanding the census — a quality/efficiency lever, not a volume story.

Phosphate-binder tailwind — transitional, expiring. The 2025 folding of oral binders into the bundle added revenue but roughly matching cost, and the enabling TDAPA expires end-2026. A one-time reimbursement-mechanics bump, not a durable line.

The real “growth” is per-share, manufactured by buybacks. The honest read of the growth algorithm is capital-allocation-led: guide adjusted OI to 3–7% on a flat-volume base, then convert that into adjusted EPS growth of +31% (2026 guide) via ~$1.8B/yr of repurchases — a large chunk bought directly from Berkshire (~45% holder) under a standing agreement. Shrinking the count against stable cash flow is legitimate value creation, but it is financial growth on an operationally static franchise, and it is finite (leverage already 3.3x).

Verdict: Low-quality growth. The underlying business is not growing in the way that matters — volume is flat-to-down, organic incidence is declining, and the “return to 2%” is an unproven ~2029-dated clinical project. What growth exists is transitional (binders), acquisition-fed and low-margin (international), decelerating and low-multiple (IKC), or manufactured below the operating line (buybacks). A mature cash cow whose per-share earnings can compound respectably for years — but calling it a “growth” business misreads the source of the earnings. The quality is in cash conversion and capital return, not the top line.

6. Financial Quality

Revenue composition and payor-mix economics. DaVita is a two-part company grafted onto one cash engine. FY2025 consolidated revenue was $13,643M (+6.4%) — US dialysis 86% ($11,793M) and ancillary 14% ($1,922M: International $1,346M, US IKC $542M, other $34M). The economics live entirely in US dialysis, and within it, in payor mix: 68% government (57% Medicare/MA, 7% Medicaid, ~3% other) and 32% commercial. Commercial pays a large multiple of the below-cost Medicare bundle, so the ~11% of patients on commercial plans generate ~26% of US dialysis revenue and the majority of segment profit. This is why the multi-year commercial-mix creep and the ACA-ePTC question dominate management commentary (a ~$40M 2026 ACA headwind was flagged on the Q1-26 call), and why a single point of commercial-mix erosion matters more than any operational lever.

The “trilogy” and margin trajectory. Management runs US dialysis on three metrics — treatment volume, revenue per treatment (RPT), and patient-care cost per treatment. In 2025 the spread compressed: RPT rose to $409.56 (+4.7%) (lifted by binders into the bundle plus normal rate increases) while patient-care cost per treatment rose faster to $273.34 (+5.9%) on wage and other cost growth. Consolidated GAAP operating income fell 2.2% to $2,044M, while adjusted operating income rose 5.7% to $2,094M (the gap is ~$25M of direct cybersecurity charges plus normalization). GAAP operating margin ~15.0%, EBITDA margin ~20% ($2.73B). By segment, adjusted OI was US dialysis $2,109M (+1.1%), ancillary +$117M (IKC turned GAAP-positive; International scaled), corporate −$133M. The core dialysis business is a low-single-digit organic grower with structurally flat-to-slightly-rising margins; the 2025 EPS growth came from below the operating line, not from operating leverage. Management’s own framing — a 2.6% five-year total-cost CAGR and a “3–7% OI growth” algorithm — confirms a GDP-plus operator, not a compounder.

Quality of earnings — two real leakages (and one data artifact to ignore). (1) Anchor to reported figures: GAAP diluted $9.84 total / ~$9.51 continuing-ops and DaVita-reported adjusted diluted $10.78 (2024 $9.68, +11%). (A third-party aggregator’s “extraordinary gain ~$639M / $13.89 EPS” is a data-mapping artifact — no such line exists in the 10-K.) (2) Noncontrolling-interest leakage is real and large: DaVita consolidates majority-owned JV centers, so of $1,079M total net income, $332M (≈31%) belongs to minority physician/partner interests, leaving DVA common with $747M — every consolidated figure (revenue, EBITDA, EV/EBITDA) overstates what DaVita shareholders own by roughly a third; ~$324M left as cash distributions. (3) DaVita’s own FCF definition is the honest one: it reports FCF of $1,024M (2024 $1,162M) after subtracting NCI distributions and all capex — versus naïve CFO−capex of ~$1,311M; use ~$1.0B as free cash flow to equity. On the positive side, earnings are not diverging above cash: CFO of $1,887M is 2.5x net income to common (driven by ~$715M D&A and the NCI add-back); SBC is small (~$140M) and non-distorting.

Returns on capital, balance sheet, maturity wall. ROE is meaningless — DaVita carries a shareholders’ deficit of −$651M (2025, vs +$121M in 2024), the mechanical result of cumulative buybacks exceeding retained earnings, which also renders P/B null. The relevant measure is ROIC ≈11.1% against a WACC in the ~8–9% area — a positive but unspectacular ~2–3-point spread that has been roughly stable. Balance sheet: financial debt principal $10.34B (net ~$9.6B), cash $676M plus $24M STI, $1.5B undrawn revolver — company-defined leverage 3.34x consolidated EBITDA, comfortably inside the 3.0–3.5x target. (This corrects the ~4.5x figure that appears if ~$2.5B of operating leases is wrongly counted as debt.) Debt expense jumped to $580M (+23%) at a 5.51% weighted rate — a rising fixed charge consuming ~28% of operating income and the single largest constraint on the levered-buyback model. The maturity schedule is heavily back-loaded (2026–29 only $109–150M/yr; then $4.49B in 2030 and $5.35B thereafter); the July-2025 refinancing (new Term Loans A-2/B-2, $1.5B revolver) termed out near-term risk, so refinancing is a 2030 event.

Verdict — do economics improve with scale? Only weakly. DaVita has the scale advantages of the #1/#2 US operator — purchasing power, fixed-cost absorption, a clinical/data apparatus — but the numbers show a business whose operating margin and ROIC are essentially flat with scale (ROIC ~9–11% across five years), whose price/cost spread compressed in 2025, and which surrenders ~31% of profit to JV partners. Cash conversion is strong and clean, and leverage is prudently managed at ~3.3x — but reported growth is manufactured below the operating line. This is a high-cash, GDP-plus, capital-adequate business — not a high-quality compounder.

7. Capital Allocation

The strategy is one thing: a levered equity-shrink machine. DaVita is the purest large-cap expression of the AutoZone playbook — take a slow-growing, cash-generative, moaty business, hold leverage at a target band, and route essentially all free cash flow (plus the incremental debt that constant-leverage-on-growing-EBITDA permits) into buybacks. The arithmetic is stark: basic shares outstanding fell from 119.8M (2020) to ~74M (end-2025) — ~38% retired in five years, at a 5–9%/yr pace. In FY2025 DaVita repurchased 12,678,623 shares for $1,788M at an average of $140.09 — 9.29M open-market at $138.98 and 3.39M bought directly from Berkshire at $143.11 — after $1,389M (2024) and $286M (2023). Authorizations were topped up +$2.0B in Sep-2024 and again +$2.0B in Aug-2025, and repurchases continued into 2026 (3M shares in Q1, 2M more since). There is no dividend and never has been. Retiring ~13% of the diluted count in 2025 is what converted ~5% adjusted-OI growth into ~11% adjusted-EPS growth — and the further shrink to ~62M average shares is what drives the guided +31% 2026 adjusted-EPS jump.

Is it wise to buy back at these prices? Mixed. DaVita bought at an average ~$140 in 2025; that vintage looked expensive against its own valuation percentile then and is now marked at ~$235 — so those repurchases worked. The deeper critique is potency and fragility: at ~$235 the same ~$1.0–1.2B annual FCF retires only ~7–8% of the float versus ~15% at ~$140, so the per-share engine structurally weakens as the stock rises. And EPS growth depends on continued share shrink funded partly by debt, into a business with flat operating margins, rising interest expense ($580M and climbing), and a ~31% NCI leakage — a thin, levered residual on top of ~$10B of debt and a −$651M book deficit. It compounds per-share value only while ROIC stays above the after-tax cost of the incremental debt; that spread is real today but not wide.

M&A and the two ancillary bets. Beyond buybacks, capital went into two adjacencies. International grew revenue +37.8% to $1,346M (Latin America, notably Brazil) — a scaled, GAAP-profitable business (Q1-26 OI $30M) diversifying away from US reimbursement risk. IKC, the value-based build, was a multi-year cash drain that finally turned GAAP-positive in 2025 (+$22M) but remains lumpy and Q4-weighted (Q1-26 loss −$19M). Sensible, on-strategy uses of capital, but small relative to the buyback — optionality, not the thesis.

Leverage philosophy and incentives. Management targets 3.0–3.5x and sits at 3.34x — disciplined; the July-2025 refinancing termed out the wall to 2030. The incentive design reinforces the per-share strategy: CEO Javier Rodriguez’s 2025 total comp was $18.2M (down from $21.8M), with the annual bonus on Adjusted Operating Income and Adjusted Free Cash Flow (paid 92.9%) and long-term equity 60% PSUs on cumulative 3-year Adjusted EPS + Relative TSR. Notably, there is no explicit ROIC or return-on-capital metric — pay rewards adjusted-EPS growth and TSR, both of which buybacks and leverage flatter directly. Aligned with the chosen strategy, but it does not independently police capital efficiency, and it pays management to keep shrinking the count regardless of price.

SEC filings sweep & insider read. The trailing 5-year corpus (5× 10-K, 15× 10-Q, 51× 8-K, DEF 14A, 300+ Form 3/4/5) yields a clean event map. The material 8-Ks are the April 14, 2025 cybersecurity incident (Interlock ransomware; ~$25.2M direct charges in Q2-25 that exclude the unquantified business-interruption drag — the main reason GAAP OI fell while adjusted OI rose; systems restored), the July 2025 Seventh Amendment / refinancing, and the Aug-2025 +$2.0B buyback authorization — no litigation, restatement, or auditor 8-K of note. The insider read offers no bullish confirmation: across 150+ Form 4s since Jan-2024, management activity is entirely routine grants (code A) and tax-withholding (code F) — zero open-market purchases (code P) by any officer or director, including through the ~24% decline to ~$114 in early 2026. Separately, every Berkshire Form 4 is a sale (code S) (e.g., 1.22M shares at $149.84 in May-2026), but this is not a conviction exit — it is the mechanical output of the April-2024 Share Repurchase Agreement, under which DaVita buys shares from Berkshire so its stake does not breach the standstill cap as the float shrinks. Berkshire held 30,100,585 shares = 45.5% at March-2026, essentially unchanged in proportion — selling to stand still, not to leave.

Verdict: Competent execution of a deliberately aggressive, defensible-but-fragile strategy. Management has allocated capital coherently — hold leverage at ~3.3x, term out maturities, seed two reasonable adjacencies, and funnel the rest into a relentless buyback that retired ~38% of the float. This is not “intelligent capital allocation” in the Buffett buy-it-cheap sense; it is financially-engineered per-share compounding that has at times overpaid and that leans on ever-rising debt service and a flat-ROIC core. It creates value while the ~2–3-point ROIC-vs-cost-of-debt spread holds and reimbursement is stable — and unwinds quickly if either breaks. Rational and shareholder-oriented, but not conservative — and not confirmed by insider buying.

8. Changes and Headwinds — Last Two Years

Verdict up front: the last two years modestly weakened the structural thesis while the equity re-rated to an all-time high on operational execution — a widening gap between price and durable fundamentals. None of the changes is an acute break, but the cumulative direction of reimbursement, payor mix, and secular demand is negative, and the balance sheet has less slack.

(a) The April 2025 ransomware attack. On 2025-04-12 DaVita disclosed (8-K, 2025-04-14) an Interlock ransomware incident that encrypted portions of its network; the intrusion began ~2025-03-24. Dialysis kept running on contingency protocols, but back-office/operational functions were disrupted and >1.5TB of data — including PHI of 2,689,826 individuals — was exfiltrated. Direct incident cost was ~$13.5M in Q2-2025 (with ~$25.2M of total related charges booked; the unquantified business-interruption impact is separate). Interpretation: the operational disruption is resolved and the direct cost immaterial to a $13.6B-revenue company; the residual risk is the multi-year litigation/notification tail (class actions filed) and the demonstration that a single-network operator running ~2,657 centers is a systemic cyber target.

(b) Oral phosphate binders into the ESRD bundle (TDAPA), effective 2025-01-01. CMS folded oral binders into the bundle, reimbursed transitionally via TDAPA at 100% of ASP plus a $36.41/month add-on. This lifted reported RPT in 2025, but patient-care costs rose in near-lockstep — a roughly margin-neutral pass-through, not a profit tailwind — and the TDAPA window expires end-2026, after which whether CMS rebases the bundle to fully cover binder cost is an open question. A headwind casually mislabeled as a tailwind.

© Medicare Advantage mix and 2026 rate/utilization pressure. Post-Cures-Act, ESKD patients enroll freely in MA, which reimburses below commercial; a rising MA share of the ~57% “Medicare & MA” bucket is a slow structural drag versus the shrinking commercial book. Management noted in Q1-2026 that commercial and MA mix were ~flat sequentially — mix has stabilized recently — but the multi-year trend remains adverse and 2026 MA plan margin/utilization pressure is a watch item. Offset: CMS’s +6% ESRD MA rate for 2027.

(d) Marietta v. DaVita (SCOTUS, 2022). Decided 7-2 (Kavanaugh): a plan that limits outpatient-dialysis coverage uniformly does not violate the Medicare Secondary Payer statute. This legitimizes MA/employer “dialysis carve-out” designs that reimburse dialysis at a fixed low multiple of Medicare and push ESKD patients toward government coverage — directly attacking the commercial cross-subsidy. The FY25 10-K still lists it as an unresolved risk with uncertain payor uptake. A slow-acting but strategically central negative.

(e) International M&A. DaVita acquired Fresenius’s Brazil, Colombia, Chile, and Ecuador operations (154 clinics, ~30,000 patients, ~$300M; 2024) to become the largest LatAm provider; international is now ~94,500 patients / 14 countries. Sensible redeployment of US free cash into a fragmented, growing, less-reimbursement-concentrated market — but it adds FX, sovereign, and integration risk and is not yet a needle-mover.

(f) IKC / value-based care inflection. After years of losses, IKC revenue rose to $542M (FY25) from $504M (FY24) and swung to +$22M operating income from a −$18M loss. Removes a chronic drag and validates the “kidney care beyond the chair” strategy, though still a small fraction of profit.

(g) Leadership/board. Javier Rodriguez remains CEO; board turnover was routine (director Charles Berg did not stand for re-election in 2025). No destabilizing change.

(h) GLP-1 secular-demand debate. The 10-K explicitly names GLP-1/SGLT2 therapies as able to slow CKD progression and reduce future ESKD incidence/dialysis demand. A real long-duration terminal-value question, but no measured volume erosion yet — a 5–15-year overhang, not a 2026 event.

(i) DOJ / anti-kickback / AKF litigation. In Feb-2025 DaVita paid $34.5M to settle FCA kickback allegations (a qui tam by ex-COO Kogod). A DC Attorney General antitrust investigation into DVA/Fresenius donations to the American Kidney Fund is ongoing. Offsetting positive: the 9th Circuit (Apr-2026) struck down California’s AB-290 as unconstitutional, protecting a mechanism that keeps commercial patients on commercial plans.

9. Risk Analysis (Risk Matrix)

Catastrophic / total-loss risk. A permanent-impairment scenario is unlikely given the essential, non-deferrable nature of dialysis, duopoly scale, and ~$1B+ of annual free cash generation that comfortably services $580M of interest. The realistic tail is not bankruptcy but equity de-rating: a step-down in the commercial cross-subsidy (aggressive Marietta-style carve-outs plus continued MA-mix shift) compressing margins on top of 3.3x leverage and a full valuation — a combination that could halve the equity without threatening solvency. A true total loss would require a simultaneous reimbursement shock and refinancing freeze, a low-probability compound event.

10. Valuation Discussion (Embedded Expectations)

Reconciled market cap and enterprise value. DaVita reported ~64.2 million shares outstanding as of 5 May 2026 (Q1-26 10-Q cover), already reflecting Q1 repurchases (including from Berkshire). At the $234.91 close (2 Jul 2026) that is a market capitalization of ~$15.1 billion. Adding net financial debt of ~$9.6B (or ~$12.6B including finance leases) and ~$1.8B of minority interest yields an enterprise value of ~$27–29.5 billion. Roughly 45% of DaVita’s enterprise value is debt — this is a levered equity, and small EV moves lever hard onto it. (A stale aggregator snapshot of “~$8.4B market cap” derives from an earlier ~$113–160 price and share basis; do not use it.)

Multiples — rich against DaVita’s own history, not against the tape.

The cleanest tell is the own-history valuation percentiles: P/S at the 98.5th percentile of DaVita’s own ~10-year history, P/E at the 80.5th, composite 89.5th (P/B is null — negative book equity from years of debt-funded buybacks). On P/FCF the stock re-rated from a persistent ~3.5–5x (2020–24) to ~13x. Every one of these is an own-history statement: DaVita is not expensive versus the S&P — a mid-teens forward P/E on a defensive cash generator is unremarkable cross-sectionally — it is expensive versus every price it has ever commanded for itself. Against its closest peer, Fresenius Medical Care trades around ~6–7x EV/EBITDA; DaVita now commands a clear premium to the other half of the duopoly.

The GAAP-vs-adjusted gap deserves a flag. FY2025 GAAP diluted EPS from continuing operations was ~$9.51 (total ~$9.84), putting the trailing P/E near 24x. Management guides 2026 adjusted diluted EPS to $14.10–15.20 (Q1-26 call), which drops the forward multiple to ~16x. The ~$4–5-per-share wedge between GAAP and adjusted (discontinued-ops drags, restructuring, technology-investment and amortization add-backs) is large enough that the “16x” deserves the same skepticism as the “24x” — the truth of the earnings power sits between them.

Embedded expectations — what ~$235 is underwriting. Run the price backwards. At ~10.7x EV/EBITDA (top of its band) on a business whose management targets 3–7% operating-income growth, the equity math leans heavily on the buyback — whose per-share potency is structurally halving as the stock rises: 2025’s repurchases shrank the count ~15% when the average price sat near $100–160; at ~$235, the same ~$1.0–1.2B of annual FCF retires only ~7–8%, and flat/rising interest expense (~$145M/quarter) plus a 3.5x leverage ceiling cap any debt-funded acceleration. To justify the price you must believe all of: OI compounding at the upper end of 3–7%; MA-mix and ACA-ePTC-expiry RPT headwinds staying modest; no GLP-1-driven volume erosion; and the multiple holding at its richest-ever. In effect, the ~9x→10.7x re-rate has pulled several years of buyback compounding forward into today’s price.

Scenario framing (no price target).

• Bear: Volume goes flat-to-negative as GLP-1/SGLT2 slow ESKD incidence; commercial-to-government mix erosion and ACA bronze-plan migration compress RPT; cost/treatment (+5.9% in 2025) keeps outpacing price; OI flat to down; the multiple reverts toward its historical 8–9x EV/EBITDA. The scenario the current price most cavalierly dismisses.
• Base: OI grows ~3–4%, RPT +1–2%, volume +~0.5% (Fresenius transfers helping); ~$1.0–1.2B annual buyback retires ~5–7% of shares → adjusted EPS +~7–9%; multiple roughly held near 15–16x forward.
• Bull: Fresenius closures plus active share capture push volume +1–2%; binder/TDAPA reimbursement nets positive; MA-mix stabilizes; IKC keeps improving; OI +6–7% → adjusted EPS +12–15%, and a “de-risked compounder” narrative re-rates toward ~18x.

What the market has right, and what it may not. Correctly priced: DaVita’s resilience — a near-duopoly in a non-discretionary service, a proven cost-and-buyback machine, and genuine H1-2026 operational outperformance. Potentially mispriced: the market is extrapolating the H1-2026 momentum and the buyback flywheel at the exact moment the flywheel loses torque — while the GLP-1 secular-volume question is a real multi-year tail and MA-mix/ACA-expiry are live 2026–27 RPT headwinds. The 98.5th-percentile P/S is the market’s way of saying the easy re-rating is behind, not ahead. (No price target; no recommendation.)

11. Variant Perception

Consensus. After a decade dismissed as a value trap — a slow-growth, heavily-levered, regulatory-hostage dialysis operator threatened by MA reform and, latterly, GLP-1 demand destruction — consensus has, in six months, flipped to “de-risked defensive compounder.” The 2026 melt-up encodes a market that now believes the secular-decline fears were overdone, that the buyback flywheel plus Fresenius share gains delivers durable high-single-digit-plus EPS growth, and that Berkshire’s ~45% ownership validates the franchise. The low beta (~0.6) and defensive-quality profile made it a rotation destination.

Strongest bull case. DaVita is half of a rational US duopoly with genuine pricing/scale advantages, throwing off ~$1.0–1.3B of FCF that retired ~42% of shares since 2020. The Q1-26 print showed all three levers (volume, RPT, cost) beating simultaneously, with management raising volume guidance from flat to +0.25–0.50% — the first tangible evidence the GLP-1 doomsday case is not imminent. Fresenius closures hand DaVita organic share, IKC is inflecting toward profitability, binder bundling is a potential RPT tailwind, and even flat OI compounds into high-single-digit EPS via buybacks. On this view the re-rate simply corrects a decade of unwarranted discount.

Strongest bear case. Underlying volume is structurally flat and faces a real secular threat: GLP-1/anti-obesity drugs should, over years, slow the incidence funnel that feeds dialysis. Simultaneously, RPT faces a vice of MA-mix migration and ACA-ePTC expiry pushing commercial patients to bronze/Medicaid coverage, while cost/treatment is already climbing (+5.9% in 2025). The company carries ~$9.6–12.6B net debt at 3.34x and negative book equity, so the buyback that drives the whole per-share story is both the source of the leverage and losing potency at a tripled price. Overhangs persist — Marietta-lineage litigation, DOJ/AKF scrutiny, the 2025 ransomware tail. Layer a 98.5th-percentile P/S and ~24x trailing GAAP P/E onto a low-growth, levered, secularly-questioned business, and the risk/reward has plainly inverted from where it sat at $101.

The 3–5 assumptions that actually matter.

1. Treatment-volume trajectory — does the +0.25–0.50% 2026 guide mark a durable inflection, or a Fresenius-transfer-flattered pause before GLP-1 erosion shows up in new-starts?
2. Revenue-per-treatment durability — can RPT hold +1–2% through MA-mix loss and ACA-ePTC expiry, or does commercial mix roll over in late-2026/2027?
3. Buyback potency at elevated prices — does management keep buying at ~$235 (weaker accretion), and does FCF/leverage even allow the same pace?
4. Multiple persistence — is 10.7x EV/EBITDA / 98.5th-percentile P/S a new floor, or a peak that mean-reverts the moment growth disappoints?

Factor-positioning read (the offsides question). FactorsToday frames DaVita as a cheap, low-beta, anti-growth value name (Growth beta −0.57, Value +0.34, Market +0.60, BetaFactor −0.21) that clusters with dividend-aristocrat and low-volatility baskets (SDY, NOBL, SPLV) — the buyback functioning as a synthetic dividend. Yet this “boring” profile just produced a ~+61% quarter and ~+108% six-month run to an ATH, with model R² of only 12–23% and 38% idiosyncratic volatility — the move is overwhelmingly DaVita-specific, not a factor wave. That cuts both ways. On the melt-up, consensus looks offsides long: a low-beta value stock does not “earn” a doubling on ~mid-single-digit fundamental growth — the re-rate is almost entirely multiple expansion (P/S 62nd→98th percentile). On the secular-decline fear, consensus was arguably offsides short for years — volumes did not collapse, and 2026 guidance ticked up — the relief the melt-up monetized. Synthesis: the market was too bearish on near-term volumes and has now over-corrected into paying a peak multiple for a low-growth compounder just as its buyback engine loses torque.

What would falsify each side. Bull falsified by: a print showing treatments-per-normalized-day rolling over, or RPT compressing on commercial-mix loss — evidence the volume/rate turn was transient. Bear falsified by: two-plus consecutive quarters of genuine +1–2% organic volume growth with stable commercial mix — evidence the GLP-1/MA-mix erosion is not materializing and the “compounder” re-rate is earned, not borrowed.

12. Fact vs. Interpretation Table

13. Open Questions

1. Does the H1-2026 volume turn persist, or fade once the Fresenius clinic-closure transfers are fully absorbed (roughly two-thirds of a year of new-starts by year-end)?
2. How fast does commercial mix actually decline through 2026–27 as ACA enhanced-premium-tax-credits expire and patients trade to bronze/Medicaid — and does the +6% 2027 ESRD MA rate meaningfully offset it?
3. What replaces the TDAPA phosphate-binder mechanism at end-2026 — does CMS rebase the bundle to fully cover binder cost, or is there a margin step-down?
4. What is the true normalized earnings power given the ~$4–5/share GAAP-to-adjusted wedge and the ~31% NCI leakage — is “adjusted” flattering, or fair?
5. How long can the levered buyback run before leverage or ROIC-vs-cost-of-debt spread forces a slowdown, and will management keep buying at ~16x forward / all-time highs?
6. Does GLP-1/SGLT2 adoption begin to show up in new-start incidence within a 3–5-year window, ahead of the market’s apparent assumption of “no effect”?
7. What is the ultimate cost and liability tail of the April-2025 ransomware breach (2.69M PHI records, class actions)?

14. What Must Be True

For the bull (owning at ~$235 works out):

• US treatment volume must inflect durably to +1–2% organic (not just Fresenius transfers), validating the ~2029 clinical-growth bridge earlier than expected.
• Revenue-per-treatment must hold +1–2% through the MA-mix and ACA-ePTC headwinds — i.e., the commercial cross-subsidy must prove sticky despite Marietta and premium-assistance pressure.
• The buyback must keep compounding EPS at a high-single-to-double-digit pace and the market must keep paying ~16x forward / 98.5th-percentile P/S — the peak multiple must persist.
• GLP-1/SGLT2 must not visibly dent ESKD incidence within the investment horizon.
• Falsification test: two consecutive quarters of treatments-per-normalized-day rolling over, or a step-down in commercial mix/RPT, with no offsetting rate win — the volume/rate turn revealed as transient.

For the bear (the stock de-rates from here):

• Volume stays flat-to-negative and/or GLP-1 begins to bend incidence; commercial mix erodes on ACA-ePTC expiry and MA migration; cost/treatment keeps outpacing RPT.
• The multiple mean-reverts from ~10.7x toward its historical ~8–9x EV/EBITDA as the “compounder” narrative disappoints — a de-rate that outweighs buyback accretion.
• Rising interest expense and 3.3x leverage constrain the buyback pace, removing the per-share engine at exactly the wrong price.
• Falsification test: two-plus consecutive quarters of genuine +1–2% organic volume growth with stable commercial mix and RPT — evidence the secular-erosion thesis is wrong and the re-rate is earned.

The analysis above takes no position, sets no price target, and makes no BUY/SELL recommendation. The single, clearly-labeled exception is the Claude's Take block at the top, which is the author’s own independent opinion. Source appendix follows as Appendix B.

APPENDIX A — Standard Diligence Questionnaire

DaVita Inc. (NYSE: DVA) — as of 2026-07-04. Supplemental to the memo. Labels: F = Fact, I = Interpretation, A = Assumption.

General

What thoughtful questions have other investors asked? The recurring institutional questions center on (1) the durability of the commercial cross-subsidy (~11% of patients / ~26% of revenue) against Medicare Advantage penetration, ACA-subsidy expiry, and Marietta; (2) whether GLP-1/SGLT2 drugs will erode the ESKD-incidence funnel and when; (3) the sustainability and per-share potency of the buyback at all-time-high prices; (4) the true normalized EPS given the GAAP-to-adjusted gap and ~31% minority-interest leakage; and (5) Berkshire’s ~45.5% stake and its selling into the buyback. The Q1-26 call was dominated by volume-trajectory, mortality-normalization, and commercial-mix questions.

Cyclicality & Earnings Nature

Cyclical high or low? Neither in the industrial sense — dialysis demand is non-discretionary and largely acyclical. But earnings are near a relative high on the trilogy: RPT +4.7% and a favorable H1-2026 cost/volume beat, with adjusted EPS guided +31% on the buyback (F). Margins are mid-range, not peak (I). External environment or internal actions? Both — reimbursement (external, government-set on 68% of revenue) sets the ceiling; cost discipline and capital allocation (internal) drive the delta (I). Revenue stability? Extremely high — a chronic, thrice-weekly, multi-year patient annuity; the risk is rate/mix, not volume volatility (F/I). Market size — growing/shrinking? Flat-to-shrinking on incidence: US ESRD incidence −18.6% over the decade; DaVita treatments −1.1% in 2025 (F). International adds modest growth. Terminal risk from GLP-1 (I).

Business Quality & Competitive Moat

Industry more/less competitive? Stable duopoly (DaVita + Fresenius ≈ 80% of US facilities); no new entry; if anything Fresenius is retreating (closing clinics), so competition is easing on supply while payors press on price (F/I). How profitable (ROIC/ROE)? ROIC ~11.1% vs WACC ~8–9% — a positive but unspectacular ~2–3-point spread (F/I). ROE is meaningless (negative book equity from buybacks) (F). Industry profitability / barriers? High barriers (CON, scale, JV lock-up, below-cost bundle deters entry); moderate profitability capped by government pricing (I). Easily understood? Yes — pay-per-treatment, cross-subsidy economics are transparent once the payor mix is understood (I). Undermined by low-cost foreign labor? No — care is delivered locally in-person; not offshorable (F). Do brands matter? Minimally — patients choose by location and nephrologist relationship, not brand; the “brand” that matters is the physician JV (I). Switching costs? High for patients (life-critical scheduled care, geography, physician tie) — customer captivity is a core moat element (I).

Financial Condition & Balance Sheet

Assets not on the balance sheet? The physician-referral relationships and JV alignment are economically valuable but not capitalized; goodwill ($7.58B) and intangibles carry the acquired networks (I). Off-balance-sheet liabilities? ~$2.5B of operating/finance leases (clinic real estate); contingent legal exposures (Marietta follow-on, DOJ/AKF, ransomware class actions) (F). How conservative is the accounting? Mixed — clean cash conversion (CFO 2.5x net income), but a large GAAP-to-adjusted wedge (~$4–5/share) and heavy reliance on adjusted metrics warrant scrutiny; NCI consolidation overstates headline figures by ~31% (I). CapEx-hungry? Moderate — ~$576M/yr (~4% of revenue), declining; maintenance-heavy with modest development. Not a capital sink (F).

Capital Allocation & Management

FCF and its use? ~$1.0–1.3B/yr FCF, routed almost entirely to buybacks (~$1.8B in 2025, including incremental debt); no dividend, ever (F). Philosophy: hold leverage at 3.0–3.5x, shrink the share count relentlessly (I). Significant acquisitions? Fresenius Latin America ops (~$300M, 2024–25); otherwise bolt-ons. M&A is secondary to buybacks (F). Buying back shares? Yes — ~38% of shares retired in 5 years; the central capital-allocation act (F). Issuing shares to insiders? Modest SBC (~$140M, ~1% of revenue); routine grants, no egregious dilution (F). Compensation policy? CEO Rodriguez 2025 total comp $18.2M; bonus on Adjusted OI + Adjusted FCF; LTI 60% PSUs on 3-yr Adjusted EPS + Relative TSR — no ROIC metric (F). Aligned to the per-share strategy but does not police capital efficiency (I). Management motivations? Paid to grow adjusted EPS and TSR, which buybacks/leverage flatter directly; zero insider open-market buying even through a ~24% decline is a neutral-to-mild-negative tell (F/I).

Valuation & Market Data

ADR/MLP/K-1? No — ordinary US common stock, NYSE-listed (F). Dividend policy? None; capital return is 100% buyback (F). How profitable? GAAP net margin ~5.5%; adjusted operating margin ~15%; EBITDA margin ~20% (F). Net income diverging from CFO? No divergence of concern — CFO is a healthy 2.5x net income to common, driven by D&A and the NCI add-back; earnings are not running ahead of cash (F).

Risks & Downside

What would cause the stock to decline? Commercial-mix erosion (MA/ACA/Marietta), a volume roll-over (GLP-1 or mortality), a reimbursement cut, a buyback slowdown, or simple multiple mean-reversion from the 98.5th-percentile P/S (I). Catastrophic-loss risk? Low — essential service, duopoly, ~$1B+ FCF services $580M interest comfortably; the realistic tail is a ~50% equity de-rate (cross-subsidy step-down + leverage + full multiple), not insolvency (I). Total-loss risk? Very low — would require a simultaneous reimbursement shock and refinancing freeze (I).

Recent News & Events

Has the environment changed recently? Yes, on several fronts: the April-2025 ransomware breach; the Feb-2025 $34.5M DOJ FCA settlement; the Jan-2025 phosphate-binder bundle (TDAPA, expiring end-2026); the Apr-2026 9th-Circuit strike-down of CA AB-290; a raised 2026 guide (adj EPS $14.10–15.20) on the Q1-26 beat; and a doubling of the stock to an all-time high in H1-2026 (F). Significant acquisitions? Fresenius Latin America ops (2024–25) (F). Accounting-policy changes? None material; the binder-into-bundle change is a reimbursement, not accounting, event (F). Recent changes — markets/facilities/management? International expansion (Brazil/LatAm); Fresenius clinic-closure share capture; ongoing AI/technology infrastructure investment (ScheduleHub, proprietary EMR); CEO/board stable (F).

APPENDIX B — Source Appendix

DaVita Inc. (NYSE: DVA) — research date 2026-07-04. Primary sources over secondary. Access date 2026-07-04 unless noted.

Primary — SEC Filings (EDGAR, CIK 0000927066)

• Form 10-K FY2025 (filed 2026-02-11) — business description, segment/payor-mix data, RPT and cost-per-treatment, risk factors, debt schedule, buyback Note, NCI.
• Form 10-K FY2021–FY2024 — five-year trend base.
• Form 10-Q Q1-2026 (filed 2026-05-05) — share count on cover (~64.2M), Q1 balance sheet, leverage.
• Form 10-Q corpus (15 filings, 2021–2026) — quarterly trend.
• Form 8-K, 2025-04-14 — cybersecurity (Interlock ransomware) incident disclosure.
• Form 8-K, 2025-07-17 — Seventh Amendment / debt refinancing (Term Loans A-2, B-2, $1.5B revolver).
• Form 8-K, 2025-08-20/21 — +$2.0B share-repurchase authorization.
• Form 8-K (quarterly earnings) — 2026-02-02 (FY25), 2025-10-29, 2025-08-05, 2025-05-12.
• DEF 14A (2026 proxy) — CEO/NEO compensation, incentive metrics, Berkshire 45.5% ownership & standstill.
• Form 3/4/5 (300+ insider filings) — insider transaction read; Berkshire Form 4 sales; no officer/director open-market buys.

Primary — Company Earnings Calls & Releases

• DaVita Q1-2026 earnings call transcript (2026-05-05) — 2026 guidance (adj OI $2.15–2.25B; adj EPS $14.10–15.20), volume-guide raise, RPT/mix/ACA commentary, buyback pace, leverage 3.34x.
• DaVita Q4-2025 / FY2025 earnings call & release (2026-02-02) — FY25 results, segment detail, IKC/International.
• DaVita Q1-2026 results release (2026-05-05) — newsroom.davita.com.

Primary — Regulatory & Legal

• Marietta Memorial Hospital Employee Health Benefit Plan v. DaVita Inc., No. 20-1641 (U.S. June 21, 2022) — supremecourt.gov.
• CMS CY2025 ESRD PPS Final Rule — phosphate-binder TDAPA (100% ASP + $36.41/mo add-on), effective 2025-01-01.
• DOJ press release, Feb-2025 — DaVita $34.5M False Claims Act kickback settlement — justice.gov.
• 9th Circuit ruling on California AB-290 (Apr-2026, struck down) — KPBS/CalMatters coverage.
• 21st Century Cures Act (ESRD MA enrollment, effective 2021) — Congressional Research Service R46655.

Secondary — Industry & Clinical Data

• USRDS 2024 Annual Data Report — ESRD incidence (−18.6%/decade), ~131,564 new 2023 starts, prevalence stabilization.
• FLOW trial (semaglutide in CKD), NEJM 2024 — composite kidney endpoint −24% (HR 0.76), stopped early for efficacy.
• JAMA / JAMA Health Forum — ESRD Medicare Advantage penetration (~25% 2019 → ~37%+ 2021).
• Trade press: HIPAA Journal / TechTarget (ransomware breach detail, 2.69M PHI records); Fierce Healthcare (DC AG / AKF antitrust probe); Sidley / Cleary Gottlieb (Fresenius LatAm acquisition).

Quantitative Data (reconciled to filings)

• Third-party aggregated financial data (three statements, profitability/credit ratios, enterprise value, valuation multiples, per-share data) — reconciled to EDGAR.
• Own-history valuation percentile ranks (P/S 98.5th, P/E 80.5th, composite 89.5th; P/B null).
• Five-year daily price history / EMAs / beta (price-action map).
• Quantitative factor model — factor loadings (Growth −0.57, Value +0.34, Market +0.60), risk-adjusted return leaderboard, specific volatility, related stocks.