Drug pricing policy: when a rule, not a molecule, sets the price

The first is statutory and narrow. The Inflation Reduction Act of 2022 obliged the health secretary to negotiate prices for a defined list of high-spend drugs, and the results of the first round took effect on 1 January 2026 for ten Part D medicines. The negotiated figures are called maximum fair prices, they run between 38 and 79 per cent below list, and they are projected to save the programme about six billion dollars in 2026 alone. This instrument is slow, legislated, and applies only to drugs that meet the statute's criteria.

The second is executive and broad. Executive Order 14297 directs agencies to tie US prices to what a basket of comparable developed countries pays - international reference pricing, arriving in a market that has never had it. Where the statute picks ten drugs, this aims at the whole branded market, and where the statute is durable, an executive instrument can be rewritten by the next signature.

The third is the enforcement mechanism, and it is a trade instrument rather than a health one. Manufacturers that have not agreed to reference-price terms face a hundred per cent tariff on patented imports and their active ingredients; manufacturers that agree receive a three-year suspension of it. By April 2026 seventeen such deals had been announced, covering an estimated 86 per cent of the US branded drug market. That is the number that tells you the mechanism worked as leverage.

The negotiation programme exists to reduce what one payer spends on a small number of very large drugs. Its design is deliberately conservative: a drug qualifies only after years on the market without generic competition, the list is short, and the price applies only within Medicare. That narrowness is the point - it produces a measurable saving with a bounded effect on the rest of the market, and it survives litigation better for being bounded.

Reference pricing exists to attack the gap itself. The United States has long paid multiples of what other wealthy countries pay for identical products, and every previous attempt to close that gap ran into the fact that manufacturers set prices country by country. Tying one country's price to the others' collapses that separation, which is why it produces a far larger reaction from the industry than a ten-drug list ever did.

The tariff exists to make the second one voluntary. A hundred per cent duty on patented imports is not a revenue measure; it is a price attached to declining a deal, and the 86 per cent coverage figure shows manufacturers paying it in concessions rather than in duty. That is the mechanism to understand: the tariff's success is measured by how little of it is ever collected.

Negotiated prices come from the Centers for Medicare and Medicaid Services as fact sheets and rule documents, with named drugs, named prices and named effective dates. This is the cleanest publication in the whole domain: a market on whether a specific drug is subject to a negotiated price at a specific date is settled by a document, not by an interpretation.

Executive actions appear in the Federal Register and on the White House's own channel. Their weakness as a settlement source is not availability but durability - an order can be amended, superseded or enjoined, and a market that spans a change is holding a different instrument at the end than at the start. Reading the order itself rather than the announcement of it is the difference between knowing what was directed and knowing what was said.

The deals are the hardest part to verify, and this is worth stating plainly. Individual manufacturer agreements have been announced without their terms being published, so the coverage figure is a count of announcements rather than an audited measure of price change. Independent analysts have repeatedly noted that the secrecy makes the realised effect difficult to assess. A contract that resolves on announced deals is answerable; one that resolves on actual price change frequently is not.