DaVita

DaVita is the company Warren Buffett would describe as a toll bridge if he ever wrote about healthcare: roughly 200,000 US patients on chronic dialysis, three sessions a week, for 3-5 years on average until kidney transplant or death. Together with Fresenius Medical Care, DaVita controls about 70% of the US dialysis market — a regulated, low-growth oligopoly with brutal predictable cash flow.

Berkshire Hathaway owns 45% of the company as of the Q1/2026 13F. That position has not been Buffett's personal pick — Ted Weschler built it from 2011 onwards — but the Berkshire stamp matters because DaVita uses the share buyback as its primary capital return vehicle. Since 2014, the share count is down from 213 M to 81 M today, a 62% reduction. Berkshire participates passively (they have a standstill agreement capping their ownership), so each buyback raises Berkshire's percentage automatically. By 2027 that share could approach 50%.

The 2026 controversy is GLP-1. Ozempic, Mounjaro and the new orals lower diabetes prevalence and reduce the rate at which Type-2 diabetic patients progress to end-stage renal disease (ESRD). Bears argue dialysis patient population shrinks 1-2% per year structurally. The bull counter: lag between diabetes diagnosis and dialysis is 8-15 years, and the US patient pipeline through 2030 is already in the system. Plus DaVita is expanding internationally (Brazil, Colombia, China, Saudi) where GLP-1 access is far slower.

13F top holders Q1/2026: Berkshire Hathaway 45.1% (38.3 M shares), Vanguard 7.2%, BlackRock 4.5%. Berkshire has not bought any new shares since 2024 — they let the buyback do the work. They have a standstill agreement that prevents going above 49% without DaVita board approval.

What is interesting: Greenlight Capital (David Einhorn) shorted DaVita publicly at the Sohn Conference May 2026, framing it as a GLP-1 short and citing the international expansion as misallocation. Bill Ackman publicly disagreed and added 2.1% to his position. Stan Druckenmiller has been silent on DaVita but his 13F shows a 0.4% position (small for him).

Insider activity: CEO Javier Rodriguez sold $48 M of shares between October 2025 and March 2026 — large but in line with his 10b5-1 plan. CFO Joel Ackerman bought $1.2 M on the open market in February 2026 at $182 — the first insider buy since 2022.

Forward P/E of 11.6x against a 5-year average of 14.8x and 10-year average of 16.4x. EV/EBITDA at 8.2x vs healthcare-services median of 12.3x. Free cash flow yield at 9.4% (TTM) — meaning the buyback can continue at current pace for >10 years without leveraging up. Average sell-side price target $215 (range $175-$245); the Wells Fargo bull case at $260 assumes Diaverum hits margin targets by 2028 and GLP-1 ESRD impact is delayed to 2032+. Berkshire's implied cost basis is around $70 — they have a 185% unrealized gain that they show no signs of monetizing. Bear case $140 (-30%) assumes Einhorn is right on GLP-1 and CMS does a 3% cut in 2027.

1. Q3 2026: Diaverum integration close + first combined quarter — proof point for international thesis
2. November 2026: CMS CY2027 ESRD bundled-rate ruling — single biggest reimbursement input
3. Q1 2027: FLOW trial 5-year follow-up data — clarifies GLP-1 ESRD reduction trajectory

DaVita is a position you take when you trust the buyback math more than you fear the GLP-1 narrative. The Berkshire ownership structure does something quietly powerful: every dollar of buyback raises Berkshire's percentage, and Berkshire is not selling, so the float effectively shrinks faster than the share count. I find the asymmetry compelling at 11.6x forward — even if Einhorn is right and EBITDA flatlines for 5 years, the buyback alone produces 7-9% annual share-count reduction, so EPS still grows mechanically. I size DVA as a 3-4% position with an add-trigger at any single-day drop >8% on a GLP-1 headline (those headlines come every 3-4 months on average). The risk I would not ignore is CMS reimbursement — that is the only thing that breaks the thesis completely.