Skip to main content

Business and Markets

Pfizer Versus BMS, Since 1989

I very much enjoyed this piece by Bernard Munos on Pfizer. It follows the same method as this classic article by Matthew Herper (updated here, commented on here and here): If you want to figure out how much a company is spending per new drug, take a reasonably long period, add up all the R&D expenses from the company’s financial statements, and divide by the number of drug approvals.

It’s not a perfect method – different companies may have slightly different accounting ideas about what goes into R&D, for one thing, as opposed to Selling, General, and Administrative (SG&A) but GAAP should (in theory) cut down on that variable. But its simplicity is a feature, not a bug. These companies really did spend that much money, and they really did get that many drugs out of it. So how have BMS and Pfizer been doing for their dollar?

Since its creation in 1989, Bristol-Myers Squibb has brought to market about the same number of new drugs as Pfizer—26 vs. Pfizer’s 25 (Figure 1). Their cumulative R&D spending, however, diverges widely. BMS did it for $56 billion, while Pfizer has spent $122 billion. Much of the difference—$62 billion out of $66 billion (94%)—was incurred during Pfizer’s post-1999 mega-merger era.

Munos points out that Pfizer’s tax-inversion deal with Allergan might save it two billion dollars a year. But using these figures, they’ve been spending four billion a year more than they might have, so which problem should be attacked first? I can tell you which one can have something done about it immediately, though – the taxes. Changes to R&D strategy are slow and uncertain, and if you’re the CEO of a company this size, you’re probably not going to be around when they finally kick in, anyway. The tax fix will also immediately make investors happy, whereas a gigantic retooling of R&D, along with the (likely unspoken) corollary that they’ve been doing it wrong all these years, will do nothing but sow fear and uncertainty. No, from Ian Read’s perspective, even if he bought into this line of thinking (and I’m sure he doesn’t), the current strategy is clearly the way he feels he has to go.

But that doesn’t mean that Munos is wrong. I think he’s right, and that Pfizer has been a massively inefficient research organization. The disruptions from all those mergers are surely a big part of the problem, as is the sheer size of the company. Changing the entire course of R&D there, though, even if you knew just what to do, would disrupt things yet again. Pfizer is in a hole, folks.

17 comments on “Pfizer Versus BMS, Since 1989”

  1. Chrispy says:

    A lot of this difference may come down to what counts as R&D. If you count the money that Pfizer spent on acquisitions as R&D then, sure, they look like they spend too much. But Pfizer is hardly the poster child for R&D spending — quite the opposite: their reputation is one of acquiring companies and gutting R&D. The money they spend on these companies should not count as R&D — it is an alternate route to the drugs. But the math should be a lesson to those companies actively dismantling their own internal R&D: purchasing your leads is more expensive.

  2. Dr. Manhattan says:

    I worked for both of this companies- BMS from 86 to 2001 and Pfizer from 2001 to 2007. I remember in the heady days just after the Warner Lambert Parke Davis merger, Pfizer actually spent money building up the R&D site in Ann Arbor and had daily 31 seat jet shuttle flights so scientists could go back & forth between Groton and Ann Arbor. That lasted for about a half dozen years and then the hammer fell on Michigan. Ann Arbor closed not that long after the purchase of Pharmacia. A lot of wasted money & talent in the whole R&D process which was indeed inefficient in some of the worst ways. The whole time, R&D was off balance as the hits just kept on coming. I remember that money and resources were taken from several areas to move torcetrapib forward, which was going to be combined with Lipitor, make tens of billions and save the company…

    The inflection point seemed to be when Bill Steere retired and Henry McKinnell took over. Things went downhill pretty fast, and McKinnell was forced out by the board (but with a $190 million severance package). I got outa town shortly after Kindler came on board, before the Wyeth acquisition.

  3. Anonymous says:

    Maybe there is a top limit to new drugs per organization, i.e. how many ripe therapeutic areas, good projects and targets are there at any given time? Move beyond and are we just wasting money chasing dreams…Is a sports team better if you double it’s roster size or do they just end up adding more marginal players?

  4. exGlaxoid says:

    I think both leadership and being too big can have effects on a company’s productivity. The big pharma companies used to have real scientists as CEOs and heads of R & D. Then everything started getting run by accountants and lawyers and most companies have gone down the tube since then. Same things is happening in tech companies, started by computer geeks, go well for a while, and then eventually the bean counters take over. And size can be great for screening programs, efficiency of scale in buying stuff, and sales force, but once R & D gets too big, spread across 100 sites, and run by bureaucrats, then it usually dies a slow and painful death. Glaxo and Burroughs Wellcome were both good companies, merged to form a still decent company, with only a few R & D sites and only 1 headquarter, had enough size to make a good library of compounds, and got just big enough to tackle large problems well. Once they merged with SBK, they got too big, had headquarters in multiple sites, had R & D sites all over the place, competing with each other for money, and more layers of management than was ever needed (about 20 levels). That created such a bog in R & D that nothing good could happen, sales and R & D spending became too big to be managed well, and there were over 500 VPs in the company. The entire USA only needs on VP, how can 500 VPs be useful? That was the beginning of the end for GSK. Pfizer is just following the same path, and now will eventually fail as well.

  5. Diver dude says:

    I was there for the BW/Glaxo “merger”. BW was the best company I ever worked for in 25 years in the biz and the combined entity never got close to the same levels of productivity.

  6. z says:

    I’m not sure about the details of how they calculate R&D spending, but why would you want to count the acquisition price tag into R&D rather than the cumulative R&D spending of both companies? Or at least discount it by roughly the ratio of R&D spending of the child company?

    I mean, this recent deal clearly points out that acquisitions are done for more reasons than just to acquire new drug leads. And there are price tags for those other reasons.

  7. Anon says:

    A better metric for total R&D spend would be total R&D costs PLUS BD&L costs – combined for the parent company PLUS all acquired companies.

  8. 235 East 42nd says:

    I think the blog post is largely on the money. However R and D at Pfizer has been retooled to an extent. Some compounds like Ibrance and Xeljanz have emerged and in another few years, the fate of things like their Staph Aureus vaccine etc will be known. Still Pfizer is too big for these R and D efforts alone to move the needle. Hence the Hospira and tax inversion deals to bulk up before an eventual split. The last two deals also seem to be done with the motivation that the mini Pfizer companies survive as independent entities post split.

  9. K says:

    I hate to point out the obvious, but this article is doing a typical apple-to-orange comparison here. Why is the number of the approved drugs important? The relevant number is the revenue, especially the revenue from those 26 or 25 drugs.

  10. Dave greg says:

    The relevant comparison is not revenue. The analysis is on research productivity which is not well measured by revenue which can be impacted by factors such as competitors, market access, and market size. Approvals is an adequate measure of research productivity.

    1. dp says:


      Revenue (or, better still, cash flow) is indeed the relevant (and superior) measure when estimating the value added of R&D to an organization: R&D productivity is (or should be treated as) shorthand for R&D ROI.

      As an investor (unlike, say, a patient), I don’t care about the number of drugs a company launches on account of its R&D activity. Rather I am desperate to have the R&D activity generate as much cash (beyond its expenditure) as possible – at least more than it costs.

  11. sgcox says:

    However, it risks Goodhart’s law: “When a measure becomes a target, it ceases to be a good measure.” Ideally pharma R&D exist to make better and safer medicines for which people will pay good money. Thus, making one first in class or best in class medicine is way more productive compared to many barely distinct entities (at least should be). Same as in academia, using number of papers or even papers in high impact journals as a metrics does not reflect the quality or impact of science.

  12. Emjeff says:

    Since GSK was mentioned in the comments, I am expecting Agilist to post momentarily about how “Sir” Andrew and the idiots there are embarking on a strong competitive paradigm to change the R&D landscape using synergy, and by the way it’s already having a great impact, blah, blah, blah. Always a job for a bootlicker…

  13. anonymous says:

    As a Jerk (rhyming with you know what) employee, we all use to make fun of BMS (but not Pfizer though!) if the folks are up to delivering drug etc.? Honestly it took a while and after when some of us left (or let go) to realize that somehow BMS kept chugging along and were doing something that we all missed! We all collectively became cocky after Januvia but are witness to BMS having last laugh at my previous company with Opdivo that is doing much better than Keytruda!

  14. Anonymous says:

    To Bristol’s credit they had the perseverance to continue with Yervoy when the early trials were not impressive. However, this was the result of desperation not insight as they had little else in the pipeline. The same desperation that makes small biotechs more successful than big Pharma. In reality, the efficacy of Opdivo and Keytruda look very similar. The key to success will be in determining the right combinations and tumors types. It will be a costly endeavour that will require Bristol to sell off most of the company to support the war with Merck, Roche and GSK. Good for investors, bad for employees.

    1. Anon says:

      “It will be a costly endeavour that will require Bristol to sell off most of the company to support the war with Merck, Roche and GSK.” I’m assuming Anonymous meant to indicate the war in oncology with Merck, Roche, NOVARTIS and PFIZER. GSK’s oncology was sold downriver to Novartis for a bag of magic beans. The current GSK Oncology R&D is rebuilding and will take 5-10 years before compounds are advanced enough to make it to market.

  15. Anonymous says:

    Bristol will have to narrow its focus to oncology and immunology biologics in order to survive. Their immediate competitors will be in the PD1 arena. The competition will be fierce and costly.

Comments are closed.