A $400bn drug merger is being talked about, and AstraZeneca's own shareholders hated the idea

The Financial Times reported that AstraZeneca and Bristol Myers Squibb have been in talks about a merger. Reuters followed with a person familiar confirming discussions took place. Neither company has said anything: AstraZeneca declined to comment and Bristol Myers did not respond.

That is the entire confirmed basis for a story about a roughly $400 billion combination. It is worth stating plainly, because the market reacted as though the deal were closer than the reporting supports.

The arithmetic is straightforward. AstraZeneca carried a market value of about $264 billion and Bristol Myers about $133 billion as of 2 August. Combined, the result would sit among the largest drugmakers in the world by capitalisation.

The reaction is the interesting part, because it went the wrong way. AstraZeneca shares fell around 7 per cent in early trading and closed down roughly 9 per cent in London — the company's steepest one-day decline since 2020. Bristol Myers moved very little by comparison.

A buyer's shares often soften on deal news, since shareholders are being asked to fund a premium. A fall of that size is something else: it reads as shareholders rejecting the strategic logic rather than quibbling about price. Analysts quoted in coverage described themselves as unclear on the rationale, given that AstraZeneca was not obviously in need of a transformative acquisition.

There is also a specific regulatory problem sitting inside the portfolios. Both companies sell major cancer immunotherapies — Bristol Myers has Opdivo, AstraZeneca has Imfinzi — and those are the same class of drug rather than complementary assets. Overlap of that kind is what competition regulators examine first, and it would be examined in several jurisdictions at once.

The broader argument against mega-mergers in this industry is not new and is not about any one deal: consolidation tends to reduce the number of independent research programmes, and the promised savings arrive more reliably than the promised innovation. Whether that applies here is exactly the question shareholders appeared to be asking.

It is also worth separating the two things a merger of this size would actually do. Combining balance sheets and cutting duplicated overhead is arithmetic, and it works. Combining research pipelines is not arithmetic, and the record on it is mixed: programmes get cancelled during integration, scientists leave, and the acquiring company inherits obligations it did not choose. Whichever of those a shareholder expects to dominate largely determines whether they read the report as opportunity or as risk, which is one explanation for a move this sharp on news this thin.

What is unresolved is nearly everything that matters. There is no announced structure, though reporting suggests any deal would mix cash and shares. There is no price, no timetable, and no confirmation that talks are continuing rather than concluded. Talks of this size collapse routinely and are reported at every stage, which is why a share price can move nine per cent on something neither party has acknowledged exists.