UK biotech R&D: losing money is the business model

The single most important number in UK biotechnology research is a loss. Across the 228 UK biotech R&D companies that publish a full profit-and-loss, the typical firm runs a margin of around −12% — and roughly three in five are loss-making. In almost any other industry that would read as distress. Here it is the business model doing exactly what it is designed to do: a venture- or grant-funded company burns cash for years turning science into a drug candidate, and the burn is the plan. The firms that do make money aren’t better-run versions of the same thing. They are different businesses that happen to share the label — royalty vehicles booking milestone cheques, contract labs charging fees, and the UK research arms of global drug companies whose “revenue” is money moved between group entities. Rank this market by margin without separating those out and you learn nothing. Figures are approximate — verify against a company’s own accounts before relying on any single number.

Four businesses, one label

Everything below turns on a single distinction, so it comes first. The companies here fall into four groups whose economics have nothing to do with each other:

• Venture-stage drug developers — the archetype. Little or no revenue, deep losses funded by investors or grants, spending the money on clinical trials. This is where the −12% median (and the much deeper losses further down) comes from, and it is not a sign of failure.
• Contract research and manufacturing organisations (CROs / CDMOs) — genuine services businesses that run trials, discovery chemistry or cell-and-gene manufacturing for others, and charge fees for it. These have real revenue and real, mostly thin, margins.
• Multinational research arms — the UK laboratories of global drug companies. Their reported turnover is largely an intercompany recharge, and their “profit” is a transfer-pricing outcome, not a competitive result.
• Royalty and milestone vehicles — companies whose income is licence fees, milestone payments or royalties on a drug someone else now sells. These post enormous margins because there is almost no cost of sale — but it is not a trading margin you can compare to anyone.

Two more entries aren’t companies in the commercial sense at all: LifeArc (£182.6M) and Genome Research Limited — the Wellcome Sanger Institute (£164.9M) — are medical-research charities funded by royalty endowments and grant income. They appear in the size tables because they are large, but they are held out of every competitive read below and marked with a dash where a profit figure would mislead.

The giants: mostly not drug developers

The largest firms by turnover are, with one or two exceptions, not the venture biotechs the label conjures. They are manufacturing arms, contract labs, royalty vehicles and the UK subsidiaries of global pharma. The “What it is” column matters more than any number in the row.

Read across the profit column and the pattern is unmistakable. The two biggest profits in UK biotech — Astex’s £331.4M (an 82% margin) and Kandy’s £101.1M (a margin above 100%) — are not what a well-run drug company earns from selling medicine. Astex, owned by Japan’s Otsuka, books royalty and milestone income on cancer drugs discovered from its fragment-based platform; Kandy’s near-£101M profit on £100M of turnover is the fingerprint of a one-off milestone or intercompany gain following Bayer’s acquisition of its menopause programme, not a repeatable trading result. Illumina’s £2.36bn — more than a third of the entire market’s turnover on its own, and that is its FY2022 accounts; the years filed since are larger still — is genomics-instrument manufacturing, an altogether different industry that happens to file under the same heading. Meanwhile the businesses that actually sell research as a service — Labcorp, Oxford Biomedica, Pharmaron, IQVIA — are the ones losing money or scraping breakeven. The label “biotech R&D” is doing a lot of concealing.

The burn you don’t see at the top

The giants table is sorted by turnover, which is exactly why it hides the real archetype. The deepest-loss companies in UK biotech barely register a pound of revenue, because they are pre-commercial: all spend, no sales. The largest single-year losses here belong to EyeBio (−£238.4M), Centessa Pharmaceuticals (−£176.3M on £5.2M of turnover), Replimune (−£163.9M), Compass Pathfinder (−£135.9M), Verona Pharma (−£132.6M) and Genomics England (−£114.0M). These are not failing companies — several are among the most valuable and closely-watched drug developers in the country. They are simply mid-burn: investors have handed them cash to run clinical trials, and the loss on the page is the trial being run.

Put the two halves together and the market’s aggregate is striking. The 228 companies book £6.6bn of combined turnover but collectively lose about £1.3bn a year. Strip out just Illumina and Astex — the instrument maker and the royalty-fed Otsuka subsidiary — and the remaining net loss deepens to roughly £2bn. That is the shape of an R&D industry: capital flows in, science flows out, and profit is somebody else’s problem for a decade.

The shape of the market

The size distribution makes the two-population structure visible. Profitability climbs almost monotonically with size — from 14% profitable under £1M to 58% in the £25–100M band — but that is not because bigger biotechs are better run. It is because the large end is populated by the recharge subsidiaries, contract labs and royalty vehicles that have revenue, while the small end is where the venture burners live. The sub-£5M tier (137 companies) runs a median margin deep in negative territory: these are the trial-stage developers, and losing money is their normal state, not a red flag.

Where the money is — and why it isn’t a margin table

Screen this market mechanically for “best-run independents” — profitable, £5–100M, double-digit margins — and it returns a list that looks like a triumph and means almost nothing, because the highest margins on it aren’t trading margins at all.

…and 5 more. Look at the 55–96% margins and you are looking at three things that are not comparable to each other or to anyone else. Ligand is a pure royalty aggregator — a 96% margin because there is essentially no cost of goods. IQVIA Biotech, Charles River, Curia and Pall are subsidiaries whose margin is set inside their parent group by intercompany pricing, not won in a market. And Amphista (58.5%, on shrinking headcount) and Autifony (55%, off a +1,310% revenue jump) are venture biotechs having a licensing year: a big pharma partner paid an upfront or milestone, it lands as revenue, and the margin is a one-off, not a business.

The genuinely instructive rows are the modest ones. Hvivo at 11.1% running human challenge trials, Futura Medical at 14.7%, and Bioascent at 24.6% doing discovery-services chemistry — these are companies earning an honest, sustainable margin from actually selling something. Nxera Pharma UK (the former Sosei Heptares) sits between the camps: a real operating R&D business at a modest 12.6%, but its revenue is still built substantially on milestones, upfronts and licensing rather than product sales — a steadier, better-diversified version of the deal model rather than a trading business. In UK biotech, a steady low-teens margin is a far rarer and more impressive achievement than a headline 90%.

Growth, read with care

Growth in this market is even more deceptive than margin, because the biggest jumps are almost never organic trading. They are deal events — a licensing upfront, a milestone, a first commercial supply — landing in a single year.

Ligand’s +77,880% is a royalty switching on, not a company scaling. Autifony and Amphista grew because a partner paid them, and their headcount barely moved (or shrank) — the tell that no operating business expanded underneath the revenue. Quell and Biomodal are deep in the venture-burn quadrant: revenue up sharply, still losing heavily, exactly as intended. Even the most reassuring row needs the same caution: Nxera Pharma UK (+176% to £78.2M, at a modest margin, headcount roughly steady) is a genuine operating drug developer — but its revenue is itself substantially milestone, upfront and licensing income, so the jump is better read as a well-diversified version of the deal model than as organic trading growth.

Market structure

Concentration looks extreme — the top five companies hold 55% of turnover, the top ten 66% — but the number is almost entirely one company. Illumina Cambridge alone accounts for more than a third of the market’s revenue, and it is a genomics-instrument manufacturer, not a drug-discovery firm. Take Illumina out and the “market” is not concentrated at all: it is a long, flat tail of small, mostly loss-making developers with no dominant player, because in a research industry there is no incumbent to dominate — the value is in the pipeline, not the P&L.

Vintage

The founding profile is young, and that is the point. The two largest cohorts — 67 companies incorporated in 2010–15 and 50 in 2016–20 — track the wave of UK venture-biotech formation over the last fifteen years, the spin-outs and platform companies that raised money to chase drug candidates. Older firms are scarce (just 12 predate 1990) because a research company’s life is finite by design: it burns capital, and then it is acquired, listed, licensed out or wound down. Longevity is not the goal, and vintage is no moat.

What the map shows

1. Losing money is the model, not the malfunction. The typical UK biotech research company runs a ~−12% margin and three in five are loss-making — because a venture- or grant-funded developer is supposed to burn cash turning science into a drug candidate.
2. The label hides four different businesses. Venture-stage developers, contract labs, multinational research arms and royalty vehicles all file as “biotech R&D” and have nothing economically in common — never rank them against each other.
3. The biggest profits aren’t drug companies earning drug margins. Astex’s £331M is royalties, Kandy’s £101M looks like a one-off milestone, and Illumina’s £2.36bn is instrument manufacturing — none is a repeatable trading result.
4. The high margins in the “best-run” screen are mirages. 55–96% figures are royalties, intercompany recharge or one-off licensing years; the honest achievement is a steady low-teens margin (Hvivo, Futura).
5. Growth is almost always a deal, not a business scaling. The largest jumps are licensing upfronts and milestones landing in a single year, with headcount flat behind them — even Nxera, the most business-like row, grew substantially on milestone and licensing income.
6. The market only looks concentrated because of one instrument maker. Remove Illumina and there is no dominant player — as you’d expect of an industry whose value sits in pipelines, not profit.

Methodology and caveats

This covers only the UK biotech R&D companies that publish a full profit-and-loss — 228 of a much larger register, so the many small or pre-revenue developers filing abridged accounts don’t appear, and the figures here are a partial view of a bigger field. Because the label mixes trial-stage drug developers, contract research and manufacturing firms, the UK arms of global drug companies, royalty vehicles and research charities, margins are not comparable across those groups, and we have flagged which is which rather than ranking them on one scale. A loss here is usually the intended cost of research, not a sign of distress; a very high margin is usually royalty, milestone or intercompany income, not a trading result. Research charities (LifeArc, the Sanger Institute) are held out of competitive reads and marked with a dash. Figures come from each company’s latest accounts in our dataset at the time of writing; where a newer filing has since landed (Illumina Cambridge’s row is its FY2022 year), the company page may show a more recent — and in Illumina’s case larger — figure, so if anything its share of the market is understated here. A few filers with overseas parents (Centessa among them) report in dollars or other currencies; their figures are shown as filed and may not be converted to sterling. Figures are approximate and business-type labels are directional — verify any specific number against the company’s own accounts before relying on it. This is analysis, not financial or investment advice.