CVS Bought Every Piece of the Healthcare Chain. Owning the Risk Is the Part It Can't Price.

Walk into a CVS to pick up a prescription, and you are standing inside a much larger ambition than the one printed on the receipt. The company filling your statin also insures roughly tens of millions of Americans through Aetna, decides which drugs your plan covers through its pharmacy benefit manager, runs the primary-care clinic two towns over, and sends a nurse to your door for the home visit. CVS no longer wants to sell you the pill. It wants to own the entire path the pill travels — the insurance that pays for it, the rules that approve it, the doctor who prescribes it, and the building where you fill it.

The official story is that CVS pivoted to health because Amazon was coming for its pharmacy. That story is tidy, and it is wrong. CVS signed the Aetna merger agreement in December 2017² — a full six months before Amazon bought PillPack — and framed the rationale in its own SEC filings not as defense, but as becoming 'the new, trusted front door to health care.'³ This was not a flinch. It was a decade-long bet that the most valuable seat in American healthcare is the one that owns every piece at once.

The whole chain, bought one piece at a time

Start with the design, because the design is genuinely good. Healthcare in the U.S. is a chain of separate businesses that each take a margin and each blame the next one for the cost: the insurer, the PBM, the clinic, the pharmacy, the home-health visit. CVS's thesis is that if one company owns all of them, it can steer a patient toward cheaper, earlier, preventive care — and keep the savings instead of handing them down the chain. The Aetna deal closed November 28, 2018: roughly $70 billion in equity, and approximately $78 billion once you count the debt CVS assumed.¹ Five years later, CVS paid another $10.6 billion in enterprise value for Oak Street Health and its 169 senior-focused clinics, closing May 2023.⁶ By 2023, the rebranded Health Services segment — insurer-adjacent Caremark, Oak Street, Signify Health, and MinuteClinic under one roof — pulled in $186.8 billion of segment revenue inside a $357.8 billion company.⁷ The pieces are all there.

Here is the thesis, plainly: CVS isn't building a flywheel — it's buying one, and hoping it spins. The structure that lets Optum print money for UnitedHealth is the structure CVS has now replicated on paper. But there is a difference between owning the parts of a machine and being able to run it. The trouble lives in the one piece everyone underestimates: when you own the insurer, you stop earning a fee and start owning the risk.

Owning the risk is not the same as managing it

A pharmacy gets paid to dispense. An insurer gets paid to be right about the future — to guess, in advance, what its members will cost, and price the premium accordingly. Guess low, and you eat the difference. That is the seat CVS bought, and that is the seat it has not yet learned to sit in. In Q1 2024, the medical loss ratio on Aetna's Medicare Advantage business jumped to 90.4%, up from 84.6% a year earlier — meaning more than 90 cents of every premium dollar went straight back out as medical cost, roughly $900 million above what CVS had budgeted.⁸ By Q4 2024, the Health Care Benefits segment posted a $439 million adjusted operating loss.⁸ The pieces fit together beautifully. They are also, right now, hemorrhaging money out the insurance side.

This is the mechanism the bull case skips. Vertical integration was supposed to dampen medical risk — steer the patient to the cheap Oak Street visit before the expensive ER trip, and the loss ratio falls. Instead, CVS assembled the risk faster than it built the muscle to manage it. Oak Street itself was still losing money at deal close, with profitability not expected until 2025 at the earliest.⁶ The retail 'front door' that was meant to anchor the whole thing wobbled too: the 1,500 HealthHUBs promised by the end of 2021⁴ never arrived on schedule, and in November 2021 CVS announced it would close roughly 900 stores over three years instead.⁵ The flywheel's spokes are all installed. Nobody has yet shown it turns.

But isn't this just the early, messy part of a winning bet?

The fair objection is that integration always looks like this before it works. Optum spent years as a cost center before it became UnitedHealth's profit engine; Oak Street's clinics genuinely do bend cost curves for senior patients once they mature; and a blown medical loss ratio is a pricing problem, not a structural one — fixable in the next plan year. All true, and it is the strongest case for patience. But it concedes the whole point: the thesis is a long-dated option, not a confirmed flywheel. The structure being sound does not make the execution proven. CVS has demonstrated it can buy the insurer, the PBM, the clinics, and the home-health visits faster than it can price the medical risk those pieces create — and in healthcare, the company that owns the risk and misprices it doesn't get a slow, forgiving decline. It gets a $439 million quarter. The integration is real. The competence to run it is still being underwritten by shareholders, one bad loss ratio at a time.

CVS built the most complete vertically integrated health company in America, and the achievement is real — most rivals couldn't assemble half of it. But assembly was never the hard part. The hard part is the thing it bought when it bought Aetna: the obligation to be right about what millions of people will cost before they cost it. CVS out-acquired its own ability to price. The front door is built, the clinics are open, the rails connect. What's still missing is proof that owning the whole chain makes the math better instead of just bigger — and until the loss ratios come down, the most expensive health pivot of the decade remains an option nobody has cashed.