About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Drug Research Areas You Wish You'd Never Heard Of | Main | Scientific Retractions: A Growth Industry? »

August 10, 2011

The Economics of the Drug Industry: Big Can't Be Big Enough?

Email This Entry

Posted by Derek

I wanted to extract and annotate a comment of Bernard Munos' from the most recent post discussing his thoughts on the industry. Like many of the ones in that thread, there's a lot inside it to think about:

(Arthur) De Vany has shown that the movie industry has developed clever tools (e.g., adaptive contracts) to deal with (portfolio uncertainty). That may come to pharma too, and in fact he is working on creating such tools. In the meantime, one can build on the work of Frank Scherer at Harvard, and Dietmar Harhoff. (Andrew Lo at MIT is also working on this). Using simulations, they have shown that traditional portfolio management (as practiced in pharma) does achieve a degree of risk mitigation, but far too little to be effective. In other words, because of the extremely skewed probability distributions in our industry, the residual variance, after you've done portfolio management, is large enough to put you out of business if you hit a dry spell. That's why big pharma is looking down patent cliffs that portfolio management was meant to avoid. Scherer's work also shows that the broader the pipeline, the better the risk mitigation. So we know directionally where to go, but we need more work to estimate the breadth of the pipeline that is needed to get risk under control. Pfizer's example, however, gives us a clue. With nearly $9 billion in R&D spend, and a massive pipeline, they were unable to avoid patent cliffs. If they could not do it, chances are that no single pharma company can create internally a pipeline that is broad enough to tame risk. . .

That's a disturbing thought, but it's likely to be correct. Pfizer has not, I think it's safe to say, achieved any sort of self-sustaining "take-off" into a world where it discovers enough new drugs to keep its own operations running steadily. And this, I think, was the implicit promise in all that merger and acquisition growth it undertook. Just a bit bigger, just a bit broader, and those wonderful synergies and economies of scale would kick in and make everything work out. No, we're not quite big enough yet to be sure that we're going to have a steady portfolio of big, profitable drugs, but this next big acquisition? Sure to do the trick. We're so close.

And this doesn't even take into account the problems with returns on research not scaling with size (due to the penalties of bureaucracy and merger uncertainty, among other factors). Those have just made the problems with the strategy apparent more quickly - but even if Pfizer's growth had gone according to plan, and they'd turned into that great big (but still nimble and innovative!) company of their dreams, it might well still not have been enough. So here's the worrisome thesis: What size drug portfolio is big enough to avoid too high a chance of ruin? Bigger than any of us have.

Here's de Vany's book on the economics of Hollywood, for those who are interested. That analogy has been made many times, and there's a lot to it. Still, there are some key divergences: for one thing, movies are more of a discretionary item than pharmaceuticals are (you'd think). People have a much different attitude towards their physical well-being than they have towards their entertainment options. Then again, movies don't have to pass the FDA; the customers get to find out whether or not they're efficacious after they've paid their money.

On the other hand, copyright lasts a lot longer than a patent does (although it's a lot easier along the way to pirate a movie than it is to pirate a drug). And classic movies, as emotional and aesthetic experiences, don't get superseded in quite the same way that classic pharmaceuticals do. Line extension is much easier in the movie business, where people actually look forward to some of the sequels. Then there's all the ancillary merchandise that a blockbuster summer movie can spin off - no one's making Lipitor collectibles (and if I'm wrong about that, I'd prefer not to know).

Comments (47) + TrackBacks (0) | Category: Business and Markets | Drug Industry History | Who Discovers and Why


1. Anon on August 10, 2011 8:29 AM writes...

I think that the mistake that all these companies are making in portfolio management is selling off all the nonpharma divisions - generics, OTC, animal health, etc. These can serve as a cash flow baseline to at least maintain some stability in the times that pharma is dry. They can greatly assist in maintaining that portfolio diversity Munoz laments as being inadequate. Agreed that they are not high margin, but when the alternative is bust, what is wrong with low margin to stay in business?

Permalink to Comment

2. AKS on August 10, 2011 8:30 AM writes...

Lipitor collectibles, great idea! That sounds like the perfect thing for the Giant Microbes folks to get into:

Permalink to Comment

3. Calvin on August 10, 2011 8:34 AM writes...

I read many of the comments and was interested by Bernard Munos's reply to much of it. And I was very interested by this very comment he made. However, i think he might be missing a trick here. Just by spending oddles of cash it doesn't mean that Pfizer's portfolio was broad or diverse So I don't think that is the correct way of looking at it. I don't think expenditure correlates with breadth or at least it doesn't have to.

But what really surprises me is that Bernard doesn't take a view based on his 2009 Nature Drug Discovery paper where he showed that the output of companies remained constant over long periods and referred to the idea of "innovative capacity". I think that is pretty profound. If you can achieve that output in the most efficient way then you'll have the chance of making good returns and never overstretching yourself. Not too big and not too small.

But perhaps it could also be achieved by only working in certain therapy areas where you had strength (prior to mega merger fever it was recognised that many companies often had little therapeutic area overlap). Concentrate on what you are good at and where your expertise is. You could have a broad portfolio within a narrow set of therapy areas. Of course you'd have to makes sure you worked in areas that other companies didn't but perhaps that is where you could get your innovative edge.

Just a thought. Flame away

Permalink to Comment

4. NoDrugsNoJobs on August 10, 2011 8:38 AM writes...

I think in any event, no matter the strategy, the growth rate of the pharma industry was always unsustainable if for no other reason than what exponential growth implies. A continued growth rate of profits of, let's say, 12% would mean that profits double about every 6 years. One quickly runs up against real boundaries and corrections must occur. Add the many other problems and we (pharma) see considerable slowing even before swallowing the entire world but perhaps those other factors are simply the specific means by which untethered growth is ultimately tethered. Were it not by the means we face (patents, payor resistance, diminishing returns on incremental improvements, increased concern over safety, etc) perhaps it would be by others but the end result would be the same, a massive correction to a more sustainable long term effort.

Permalink to Comment

5. johnnyboy on August 10, 2011 8:56 AM writes...

I'm not going to get into the movie vs. pharma industry analogy - do the big brains behind that one come from McKinsey, by any chance ? The few points these two industries have in common are so completely overwhelmed by the massive differences between them that any conclusions drawn from this analogy are bound to be faulty (good old GIGO), ie. good for a MBA thesis but not much else.

I don't think it's that complicated: developing a drug is getting longer, more complex, more expensive, and more prone to failure. Therefore it stands to reason that you need the largest portfolio possible in order to have the few successes good enough to sustain your organization. In that light, the 'bigger is better' philosophy behind Pfizer, GSK, Sanofi et al is the correct one. Where Pfizer failed was in the execution. Their executives went too fast (too many mergers in too short a time, without the sufficient time to absorb the changes each merger brought, creating havoc and instability in the R&D programs), and their shareholders wanted results too quickly - I mean, the merger spree at PFE came into its own in 2000 with warner-lambert, and the company strategy started to be hailed as 'failing' around what time, maybe 2007 with torcetrapib ? Considering that it takes 10-12 years to bring a drug to market, is a 7-8 years trial of 'bigger is better' long enough to discount it as failed ?

I think the basic problem behind pharma's troubles is that the fundamentals of drug R&D are out of sync with the way capital works nowadays. The sustained effort required for drug development is too long, compared to the rapid turnover of executive teams (with each new team eager to "shake up the status quo" and "shift paradigms", ie. take rash, crap decisions) , and for impatient wall street investors, who think in quarters rather than decades.

Permalink to Comment

6. qetzal on August 10, 2011 9:13 AM writes...


I don't think it's that complicated: developing a drug is getting longer, more complex, more expensive, and more prone to failure. Therefore it stands to reason that you need the largest portfolio possible in order to have the few successes good enough to sustain your organization.

Except that means you're spending considerably more money to get fewer successes, so each success has to be even bigger. By that model, the traditional $1B blockbuster is inadequate. Now you need $10B blockbusters. Unfortunately, I don't thing that's realistically achievable.

Permalink to Comment

7. Ellen Clark on August 10, 2011 9:23 AM writes...

Glad you are keeping the topic alive, Derek. Really great comments here too. But mostly everyone is saying what went wrong and no one is offering any solutions out of this problem. It is easy to criticize and have hindsight about big pharmas woes but how are we going to stop this recent trend to outsource everything as a way to save the bottom line?

By the way I think the movie vs. pharma analogy is ridiculous too. Apples and oranges

Permalink to Comment

8. biotechbaumer on August 10, 2011 9:26 AM writes...

I agree with many of the sentiments above...particularly the last point by JohnnyBoy. The focus of the public markets are out of sync with the reality of drug development and its long product cycle. Achieving a certain rate of growth expected by investors may simply be unrealistic in this industry. Even if it is possible, it would occur over a long period of ups and downs as certain patents expire while other drugs are being approved. Public investors simply don't have the patience for that kind of business model and this is why we always get into the discussion on how to improve R&D efficiency. The reality may simply be that Pfizer is a great company, with tons of cash on hand, profitable and will continue to remain that way for a long time---people are just too short-sighted and quick to pull the panic trigger due to lofty growth expectations from public investors.

Permalink to Comment

9. Hap on August 10, 2011 9:36 AM writes...

Shouldn't you expect to see improved pipelines, though? You shouldn't expect to be getting products from mergers seven or eight years previously, but if there's an enhancement in discovery by having merged, you should expect to have seen more candidates.

The investment timelines and the drug development pipelines are likely not in register, but mergers might have more to do with investment timelines than drug discovery ones, which would mean that one would not expect them to be profitable for discovering drugs (they could be, but you wouldn't be able to expect it). That doesn't help you if your core business is discovering drugs.

Permalink to Comment

10. Student on August 10, 2011 9:37 AM writes...

Even if they had a broader base, they still have the same management across the company. If the company invested 1 billion in 50 start ups @ 20million a piece with the companies only involvement as SCIENTIFIC consultants from their R&D staff. I bet you will find a great deal of success (It is not as though big-pharma R&D isn't capable of generating those same ideas, its the limitations placed on those). I think it would really highlight the intrusiveness of management/executive decisions on pharmas profit and pipelines.

Permalink to Comment

11. RF on August 10, 2011 9:38 AM writes...

I've got to agree with AKS, there is probably a market out therefore them. You could even claim they were an educational toy.

I am, of course, not now plotting to make myself a cuddly ibuprofen.

Permalink to Comment

12. Student on August 10, 2011 9:39 AM writes...

Even if they had a broader base, they still have the same management across the company. If the company invested 1 billion in 50 start ups @ 20million a piece with the companies only involvement as SCIENTIFIC consultants from their R&D staff. I bet you will find a great deal of success (It is not as though big-pharma R&D isn't capable of generating those same ideas, its the limitations placed on those). I think it would really highlight the intrusiveness of management/executive decisions on pharmas profit and pipelines.
I mean...if everyone is cutting 10% of their R&D, I'm sure they have a couple billion laying around to play with.

Permalink to Comment

13. @8 on August 10, 2011 9:51 AM writes...

But is it the "public?" With steady profits from a drug over a stable/predictable timeline (patent expiration), should pharma stocks be the new GE, in terms of not being volatile? To me the idea is with no single person in management sticking around for the 15 years it takes to influence discovery and trials of drugs...yet still collecting massive bonuses for a job well done. The steak hadn't even been tasted yet!

Permalink to Comment

14. Anonymous on August 10, 2011 10:11 AM writes...

"the broader the pipeline, the better the risk mitigation.." - offset by the cost of complexity - another fundamental driver of operational results. Chunking it down into therapeutic areas, or into specialized sites is a strategy that proved out about as well as "big is better".

Does that mean smaller is better? Not necessarily. The work direction cited by De Vanry, et al, offers a glimmer of hope- essentially finding ways of better managing complexity. A long way to go in Pharma to approach hedge fund rocket science... And what better example to avoid a cliff or melt-down?

Permalink to Comment

15. Cha Ching $$$ Banker on August 10, 2011 10:27 AM writes...

Bigger is better for me and my bonus for financing the deal. Next time I think we will keep the poorly paid research people and offload the expensive top tier of management.

Permalink to Comment

16. RB Woodweird on August 10, 2011 11:23 AM writes...

#12 Student's comment suggests an interesting thought experiment. What if every drug candidate which reached an adequate intermediate stage in the pipeline was spun off into its own subsidiary company?

Permalink to Comment

17. johnnyboy on August 10, 2011 11:26 AM writes...

@7 : I think the only way you could achieve success is to have an institutional culture of stability. No mega-merger every two years, no musical chairs at the executive level, no constant internal reorganizations, middle management cut down to barebones, minimal hierarchy. Researchers being rewarded for doing research, not powerpoint presentations. I don't think i'm going out on a limb by saying that researchers like stability. They like knowing that their job won't be moved across the country at the flick of a management switch, they like to concentrate on a project until fruition, not being switched around programs like weather vanes. Develop a company with that culture, and you'll have happy, productive researchers. Unfortunately, in management culture, stability is a dirty word.

Permalink to Comment

18. SteveM on August 10, 2011 11:51 AM writes...

Re: #5 johnnyboy "movie vs. pharma industry analogy"

Actually the statistical analogy is valid. In both cases, risk is mitigated through diversification. The math is not hedge fund complicated.

See the book "The Flaw of Averages" by Sam Savage for a simple explanation. He even uses diversification of a movie investment portfolio as an example.

The underlying diversification relationship is the same with Pharma. The differences are the investment profiles and probabilities of success. Both of which are random variables. However drug development costs a lot more and success probabilities are a lot lower. But the analytical methodologies to examine them are the the same.

One of Savage's 7 Deadly Sins of randomness is neglecting the probability of ruin in the course of a random walk of a portfolio. I.e., the firms runs out of money before it ever hits the home run. That's what is happening to Pharma now. It's the law of large numbers that mitigates that.

Now Pharma has had Decision Analysis shops that for years have done the Monte Carlo simulation analysis of their portfolios. Their problem is that the parameters of the probability distributions for cost and success are obsolete.

What Munos is proposing is nothing new. He's argument for Bigness is just an implicit observation that historic probabilities of cost and success have to be adjusted. And once those adjustments are made, minimizing the probability of ruin will become much more expensive.

In that context, the fault lies with the Pharma management teams who'd rather not know...

Permalink to Comment

19. Cialisize Me on August 10, 2011 12:45 PM writes...

Copyright vs patent. It is time for a new category that combines the features of a patent with the extended protection of a copyright.

Right now you have brand name drugs on patent = expensive drugs, firehose of money coming in, versus later on generics = trickle of money. Goal would seem to be some way to extend and even out the money flow, allowing producers to recapture theier costs over a longer period. This could be expected to result in lower prices.

For drugs, which by nature have such an extended development time, why not create a copyright-type of protection for the *exact* molecule that gets approved, in effect a copyright for that exact molecular structure. Give it a 20 year life (or more) from approval date. All the rest of the IP protection around the molelcule would expire as usual with the patent.

So...competitors could use the teaching of the patent to make better molecules or develop some of the others in the patent. they would have to go through the std IND process. Since you know the competition would not go generic for a long time, there would then be incentive to compete and incentive to make better molecules.... your *improved* drug would not have to compete with generic until far in the future.

I think this would create incentives to make more innovative and improved drugs, allow companies to capture costs over a longer period, lower brean-name drug prices, and stimulate competition to actually create the best product as opposed to first to market.

My 2 cents worth, hope it was worth it. C.M.

Permalink to Comment

20. BiotechTranslated on August 10, 2011 1:57 PM writes...

A few comments:

@#1: The reason why drug companies spin-off the non-pharma business units is because of earnings dilution. Sure, consumer healthcare might get you $2B in revenue, but the margins might only be 10%. If you average that into a pharma unit that produces $2B in revenue and a 30% margin, now your entire business has $4B in revenue, but only a 20% margin overall. Sell the consumer healthcare unit for $10B and funnel that money back into the pharma unit (through acquisitions or more R&D spending) and the hope is that you replace that $2B in revenue (at 10% margin) with $2B in revenue at 30% margin. Of course it rarely works out that way, but that's the thinking behind it.

As for the "big pharma isn't big enough", I think this has a LOT to do with the failure of risk mitigation strategies. You'll often see projects analyzed using an NPV (or something similar). They have a Phase I project and some revenue estimates. The problem is, they do a probability weighted NPV, which assumes there is a 40% chance of getting to Phase II and 20% chance of getting to Phase III, etc (these numbers are pulled from the historical rate of project progression). These probability numbers are TERRIBLE estimates. They don't take into account the unique aspects of the program itself. The only way those numbers would work is if the company had a pipeline that mirrored the mix of drugs that were used to put those probability numbers together. And even then it's a terrible estimate of risk.


Permalink to Comment

21. NoDrugsNoJobs on August 10, 2011 2:06 PM writes...

19: Cialisize Me - I think you hit it on the head. patent law needs to serve innovation and encourage R&D but where an industry is shedding massive R&D, its a sure sign that the incentives are not there. The 20 year period is not carved in stone, it is not a natural, immutable law - its made up. Its interesting to think that the same period covers those in fields with little investment and development time as those like pharma that are on the opposite end of the spectrum. A longer period to recoup investment via longer market exclusivity would actually allow for more stable prices and a more predictable investment landscape. Companies would have enough time with one drug on the market to predict revenue far enough out that they could truly support long term research decisions. Also, the prices would not need to be so high because the recovery period would be both more predicatble and longer. Right now, we are trying to fund long term activities like R&D for drug discovery with short term investment contracts (patents). Why are copyrights so long? Why are trademarks forever? Why are patents 20 years from filing? Does it all make sense given what we are witnessing by way of massive research reductions?

Permalink to Comment

22. SteveM on August 10, 2011 2:22 PM writes...

Re: 21 "Also, the prices would not need to be so high because the recovery period would be both more predicatble and longer."

Pharma will indeed do price sensitivity analysis when developing a business case for a molecule. But pricing has little if anything to do with the Payback Period after the fact Because R&D costs are totally sunk.

Pharma companies do rationalize drug prices based on development costs, but that's just an artifact to sway regulators. And it's also a PR play to the public. Like airlines "blaming high fuel costs" for raising fares.

The point is that Pharma and airlines will charge the price that their customers will pay regardless of the aggregate cost to provide the product or service.

You can be pretty certain that if a branded drug price is currently 4 bucks a tab, if the patent expiry date were suddenly extended, it would still be 4 bucks a tab.

Permalink to Comment

23. lynn on August 10, 2011 2:42 PM writes...

Naive thought here: Mergers are terrible because they are disruptive, lead to wholesale downsizing, loss of culture, etc [a la Mattina and everyone here] - but to get a pipeline, you don't need to merge - you can acquire [kill the M, adopt the A] good products or projects as they come along. This is the way Merck did things relatively successfully once their internal pipeline slowed down - until the Schering debacle. What if Pfizer had bought or licensed Lipitor and hadn't swallowed Warner-Lambert; or, in a smaller deal - what if they'd bought anidulafungin and not eaten Vicuron; or AZ and Novexel. Big Pharma killing small, rather than co-licensing, collaborating, letting the smaller companies continue to innovate. So, now it seems, that is what's happening [at Pfizer, GSK, Novartis, others]. Or what is claimed to be happening. Sort of what Student @10 and 12 said, Big Pharma supplying some money and consulting to provide the necessary drug discovery experience/expertise. You don't have to be Gigantic to acquire a broad pipeline; rather foster external stuff.

Permalink to Comment

24. NoDrugsNoJobs on August 10, 2011 2:44 PM writes...

SteveM - to some extent I agree with what you are saying but I don't think you can discount the absolute effect of unpredictable but short drug product lifetimes - The riskier an investment, the higher the payout for a wineer needs to be. Having said that, I think a collaborative pricing arrangement somewhat akin to what we see in Europe, Canada, etc is coming anyway, therefore, the pricing issue will probably take care of itself in the US as it has in much of the world already (like it or not). Europe allows 10 years of data exclusivity without any sort of patent, this represents an acknowledgement of the point I am making, only I think it does not go quite far enough. One can envision the entire patent system for pharma being thrown aside for a European-type exclusivity if the incentives are properly set. Patents really have nothing to do with safety and/or efficacy of drugs, they are a poor proxy for pharma policy. I could go on but will spare you....

Permalink to Comment

25. SteveM on August 10, 2011 3:01 PM writes...

Re: NoDrugsNoJobs "The riskier an investment, the higher the payout for a wineer (sic) needs to be."

My point is that pricing and project risk are taken into account when evaluating the business case a priori to full development. I.e., the forecasted pricing model has to yield a positive projected NPV.

Once the product is released, value based pricing is a marketing exercise in which sunk development costs are only used to back rationalize a target price. But that's it. Forecasted pricing is used in the decision to proceed. Actual pricing depends on what the market will pay. If a Pharma company spends an arm and a leg to develop another me-too SSRI now, they'll be sh*t outta luck regardless.

One thing the market does hate is uncertainty. Especially regulatory uncertainty because its effects can instantly implode a business case. Regarding your ruminations about future IP models, it wouldn't surprise me if Pharma is holding back in some context because of the political pathology that is playing out in Washington.

Permalink to Comment

26. AnonyMouse on August 10, 2011 3:28 PM writes...

Interesting discussion. A few people have pointed out the incompatibility between the long drug discovery & development cycle and the quarter over quarter growth needs of public investors. So my question is...where do you expect to find that "patient" investors who will be ready to sink in a billion dollars and tell you to contact him ten years later with result of your work? I really don't think such an investor exists. Unless, and here is a radical thought that I am myself not convinced about but I will throw it out there...

The most "patient / passive" investor potentially possible is the government itself. Maybe drug R&D should cease to be a private sector / for profit industry. Let's have a government organization (a behemoth one if we ever get there) go the full monty - from fundamental research to approving and distributing/selling drugs. Have a payroll tax be the source of funding for this organization. Being not for profit they can ensure that price gouging is not done when drugs get on the market.

Sure there are many disadvantages I could think of in having the public sector take on drug R&D. But it comes back to the fundamental issue that I just can't see a solution existing for...the drug R&D cycle is way too long for "for-profit investors" who'd rather put their money in sectors such as technology, internet, etc where the returns are much faster (and usually larger).

Permalink to Comment

27. Rick on August 10, 2011 4:11 PM writes...

After having spent the past 3 years studying economics and how it works in the pharma industry (assisted by a kind neighbor who also hapened to be a former head of the econ departmennt at MIT) I can assure you that we are only just now beginning to find out what we haven't known for the past 2 decades or so. Analogies like those to the movie industry are only the earliest stabs at finding useful analogies to the risk factors that overwhelmingly define the economics of the drug industry.

As Bernard correctly said, this is just beginning to get serious scholarly study; it's at the cutting edge of research. However, in my opinion, one thing is already clear, many of the economic fundamentals of this business differ significantly, quantitatively and/or qualitatively, from the industries upon which standard microeconomic and financial theory depend. As a result, pharma managers have been dutifully using tools they were taught in manager school, such as risk management or cost-benefit analysis, in ways that are fine for washing machine manufacturing, for example, but are, as these studies are pointing out, inappropriate for the pharma industry. Not only does this mean that the approaches were based on flawed assumptions, but that the reactions to crises were equally, if not more, inappropriate.

Before anyone accuses me of being another MBA basher, I'm NOT saying these guys are stupid. I'm saying they may well be doing the best job they know how in totally good faith based on what they've been taught is good management, but what they've been taught doesn't work the same way in pharma as they were taught in the classroom or case studies.

What this all means is that essentially we have been engaged in big experiments over the past 20 years, either as the researcher or the lab rat, in whether standard management tools, developed and refined for other industries, work in the pharma industry and we seem to be in the early phase where a lot of the hypotheses are wrong. Among the subjects we need to think differently about are: how the same research functions on different scales (both up-scaling and down-scaling); the fungibility of resources; the use of risk management and NPV analysis in managing early stage research; non-monetary assessments of "value" in the 10-12 year "cash-destroying" phase of drug discovery.

The sad thing is the number of jobs and potentially productive avenues of research destroyed by (in retrospect) overly confident and widespread adoption of management approaches that, it now seems, weren't well suited to this industry.

Permalink to Comment

28. Rick on August 10, 2011 5:35 PM writes...

AnonyMouse, #27, A variant of your idea on how to fund drug research that is more "patient/active" was described in Gary Pisano's book, Science Business, which I will, again, highly recommend to the followers of this blog. He called his model "Venture Philanthropy" and suggested it for very similar reasons, and more, to yours.

Even better, an example of this model has recently shown promising results in a case near and dear to the heart of our Dear Blog Leader: Vertex's discovery and development of a novel treatment for cystic fibrosis. It received significant support from the Cystic Fibrosis Foundation, which also played a relatively active role in overseeing the project (vis a vis more hands-off VCs). From my outsiders perspective, it seems that this project was a great success, both in terms of reaching its milestones and, so far, financially. Derek probably can't comment directly on this project, but perhaps he can redirect us to independent analysis of how this project worked vis a vis the more "traditional" models of financing early stage drug research.

Permalink to Comment

29. Rick on August 10, 2011 5:37 PM writes...

AnonyMouse, #27, A variant of your idea on how to fund drug research that is more "patient/active" was described in Gary Pisano's book, Science Business, which I will, again, highly recommend to the followers of this blog. He called his model "Venture Philanthropy" and suggested it for very similar reasons, and more, to yours.

Even better, an example of this model has recently shown promising results in a case near and dear to the heart of our Dear Blog Leader: Vertex's discovery and development of a novel treatment for cystic fibrosis. It received significant support from the Cystic Fibrosis Foundation, which also played a relatively active role in overseeing the project (vis a vis more hands-off VCs). From my outsiders perspective, it seems that this project was a great success, both in terms of reaching its milestones and, so far, financially. Derek probably can't comment directly on this project, but perhaps he can redirect us to independent analysis of how this project worked vis a vis the more "traditional" models of financing early stage drug research.

Permalink to Comment

30. Terry on August 10, 2011 6:01 PM writes...

medical DR plays the most important role in new drg discovery, we and other DRs are just audience to watch a process.

Permalink to Comment

31. Arthur De Vany on August 10, 2011 9:17 PM writes...

Excellent discussion of the "wild" (Mandelbrot's term) statistics of both these industries. Where the bigness analogy may fail is that with the stable Paretian distribution of outcomes, a larger firm will have a larger variance than a smaller one, where larger means more drugs. With these skewed, extreme event dominated outcomes, the variance rises with the size of the sample. See my UCLA/Harvard talk on the similarity of these industries, available on my web site.

Permalink to Comment

32. DrSnowboard on August 10, 2011 11:27 PM writes...

Isn't the part of the patient investor also being played by the large charities, Gates, WHO, Wellcome Trust? You might argue that is because they are not trying to make money...just gain health outcomes, and that lends them an objectivity that market responsive big pharma doesn't. Perhaps the only way for pharma research to be successful is for investors to realise they may well lose money? After all, for every biotech success story there are many that never made it to profitability for a whole host of reasons. What I'm reading in these comments is that at least the 'biotech' investor knows it is a high risk gamble, whereas the pharma exec thought they could string a chain of black swans together and make a flying city..

Permalink to Comment

33. Joel West on August 11, 2011 2:47 AM writes...

Clearly one difference between drugs and movies is the size of the bets. Maybe 10 movies have grossed (theatrical showing) more than $1 billion, none more than $3 billion. Meanwhile Lipitor brings in $10b/year.

Also, the big studios bring out a dozen or more theatrical movies a year, not counting TV. How many drug companies release more than 10 NMEs a year? (I’m guessing none, since the FDA seems to approve about 20-30 NMEs every year).

PS: I'm guessing Munos is referring to the 2000 Research Policy article, although Harhoff guessed it was the 2000 Journal of Evolutionary Economics.

Permalink to Comment

34. Bernard Munos on August 11, 2011 3:20 AM writes...

Anon (#1) and StevenM (#18)

The diversification question is another example where the application of standard financial theory to the pharma industry is leading some companies to their ruin. The textbook argument is that companies should not diversify because investors can do this themselves in the stock market at a much lower cost. So, as the argument goes, rather than incur the high transaction costs involved in diversifying into other lines of business (e.g., bankers' fees, disruption costs, greater complexity, etc) companies should be "pure plays".

Sound reasonable enough, unless you face a significant probability of ruin in the normal course of your business. In pharma, most products (except blockbusters) never recover their costs. The median peak sales is about $425 million. The viability of a company hinges on its ability to come up with blockbusters often enough to stave off the prospect of ruin. Companies that hit a dry spell are gone... unless they have ancillary businesses that can tide them over until their luck turns around.

You'd think that big pharma pure plays with a lot of staying power would be impervious to that, but they are not. Given enough time, they will all hit a dry spell and disappear in yet another M&A. Remember Syntex, Sandoz, Squibb, Glaxo, SmithKline & French, Marion, Merrell, Aventis, Astra, Zeneca, Sterling, Wellcome, Beecham, Boots, Upjohn, Rorer, Pharmacia, Organon, Robbins, Parke-Davis, etc?

On the other hand consider companies with a modest NME output that managed to keep themselves afloat thanks to their non-pharma business: Allergan, Bausch & Lomb, Boehringer-Ingelheim, Merck KGaA, Baxter, J&J, Abbott, Carter-Wallace, Bayer.

Unfortunately, by divesting their non-pharma businesses, companies make it easier to run what's left, but they lower their probability of survival. Roche, Bristol-Myers Squibb, and Pfizer have stacked the odds against themselves. Novartis, which has been criticized for spreading itself, is likely to outlive its detractors. Lilly would have done itself some good, had it learned to better manage its medical device business. Solvay, UCB, American Home Products, ICI/Zeneca, Hoechst thrived for decades, but promptly disappeared once they became pure plays.

Diversification is not a cure-all. You still need good management, but it helps bring stability to an industry that inherently lacks it.

Permalink to Comment

35. Bernard Munos on August 11, 2011 4:23 AM writes...

The important analogy between pharma and movies is that they both lose money on their standard product. Their business model is inherently unsustainable and only rescued by the random occurrence of blockbusters. Ten years ago, we did not understand that, but we now do, thanks to the work of scholars, and books that have popularized the concept of 'black swan'.

On the face of it, what could be easier than predict the success of a movie, or a new tune? Try it!

There are other blockbuster-dominated businesses: oil & gas exploration, mining, publishing, investing, to name a few. They really call for different management tools, but this was not recognized until recently. They don't teach that in business schools, or at least not when I was there. The MBA toolbox is more geared to mundane businesses. In the hands of over-confident non-scientists, it has been pretty toxic to drug innovation.

Is change in the offing? I fear it is not. HR groups move slowly. Their leadership pipelines are stuffed with more of the same (with a few happy exceptions).

Joel West (#36)
The papers I was referring to are "Technology policy for a world of skew-distributed outcomes" by Scherer/Harhoff, and several working papers that Scherer has on his web page.

Permalink to Comment

36. provocateur on August 11, 2011 6:29 AM writes...

I have two comments to make to Mr. Munos
1) you obviosly like Novartis and its way of doing research.Then you have the answer right in front of you.Novartis has actually comitted to increase R&D in the coming years which is in direct contrast to your call of slashing R&D.This is classic double talk from you.You like something and dont want to replicate it.Is it because of your conflict of interest as you are a consultant now.
2)I am a huge fan of history.You worked at LLY and it faces one of the worst crisis in the coming years because of its 'patent cliff'.Why should I listen to you at all because your past 'recipes' hv not exactly cured the 'disease'?

I want an honest reply.Were you wrong then or now?
Let me see your honesty in your arguments and see your potential for self reflection.

Permalink to Comment

37. anon on August 11, 2011 6:41 AM writes...

Mr. Munos
In your answer for reforming R&D, I see elements of Mr. Taleb's black swan.But unlike markets where you can bet in the extremes in the hope of cashing on a 'black swan' occurence, R&D Projects cannot be assigned risk because unlike markets you cannot hedge!I see doom in your argument as I see for the simple reason that every project is intself a black swan in pharma r&d.

Permalink to Comment

38. Rick on August 11, 2011 7:24 AM writes...

Thanks to both Bernard (#37, 38) and Arthur (#32, 38) for weighing in. In light of your comments and your clear scholarly interest in this subject matter, I have a couple of questions:

1. (At the risk of being accused of MBA bashing) How long before the information you are discussinig will become part of the curriculum in economics, finance and B-school classes? Like the idea that the earth orbits the sun in Galileo's time, these seem radically opposed to what is currently accepted as the "standard model" of the economic universe. Until curricula change, how can we expect the people emerging from these classrooms to adjust their approaches in highly risky enterprises?

2. How much study is being devoted to understandidng the relationship between scale and productivity at the smaller end of the scale? You and others have spoken at length about the non-linearity (or perhaps inverse relationships???) between size at the large end of the scale, making the still-counterintuitive point that bigger isn't always better. At the smaller end, it seems to me that there remains this idea that pharma is like house-painting, wherein 1 painter can do a job in 2 days that 4 painters can do in one-half day. For us researchers, it seems pretty obvious one "standard scientist unit" (SSU) working full time in proximity or 2 SSU working half-time or at two widely separated sites can NOT discover a drug in 2X days that two SSU working in proximity can discover in X days. It may often be the case that 1 SSU working full time or 2SSU working half-time (or full-time at two separate facilities) will never, ever in all of eternity, discover one drug whereas 2 SSU can discover and develop it in 10-12 years. But this idea of a critical mass seems to elude many (perhaps most) people who have the greatest power to influence industry strategy. It even seems to be woven into the fabric of corporate M&A strategies. I suspect that a plot of the relationship between some measures of size of effort (e.g. dollars, headcount, non-human resources, hours) will be sigmoid with a jaw-droppingly small "linear" range between the extremes. If I'm right, this would have significant implications for what tomorrow's healthcare economists and MBAs should be taught.

Permalink to Comment

39. TRB on August 11, 2011 8:27 AM writes...

I read with interest the ongoing dialogue on portfolio management. What has become apparent is our collective insistance to force a solution into a particular silo. In reality, the problem with Pharma pipelines is a multifactorial where faulty decisions perpetuate the degradation of pipeline portfolios. Should we have a dialogue on what is wrong with the current model of drug discovery? At 300K an FTE/yr, have we priced ourselves out of the market (China is ~100K per FTE and India +60K)? Given the economics, how long should it take to deliver a development candidate? Will outsourcing really solve the fundemental problems in a high-risk drug discovery environment?

I would argue that we have no lack of innovative ideas. We have a problem with execution of an idea to marketed product. As scientists, we are far too open-ended with our commitment to projects. We cannot take 5+ years to deliver a drug candidate.

Permalink to Comment

40. CMCguy on August 11, 2011 10:31 AM writes...

#27 Rick I know what you are saying however I would tend to bash MBAs as majority being "Stupid" for not recognizing "what they've been taught doesn't work the same way in pharma as they were taught in the classroom or case studies." That axiom has been obvious for a long time but rather than adapting most continued to push the misaligned approaches that exacerbated the problems. Maybe these new models, or historical perspective Pharma will be taught to future MBAs but they way the industry is going will that enlightenment be relevant.

Of course I cast same accusations on most Lawyers and even MDs for not "understanding R&D" then counter that most in R&D, especially at Big Pharma, have little clue on functioning in other areas we like to criticize (although view most scientist types probably more amenable to learn and readily adapt to such tangential knowledge).

Permalink to Comment

41. anon#1 on August 11, 2011 10:57 AM writes...


I see that you generally agree with me. However some of your examples of diversified companies that became pure plays then disappeared aren't entirely accurate. AHP, later Wyeth, was hugely diversified (think Libby's, Advil, Jiffy Pop, Robitussin, 3-in-1 Oil, Chapstick, rattlesnake antivenin, vaccines, Fort Dodge animal health, Wyeth nutritionals, etc which were only a few of their nonPharma products over the years) and did narrow their focus over time. Just prior to Pfizer's purchase, they maintained small molecule, biotech, nutritional, vaccines, OTC, and animal health at Wyeth. The reason they were gobbled up by Pfizer is that they were diversified, especially the vaccines and biotech, which is what Pfizer did not have. The tragedy was that the Wyeth CEO got greedy and saw the $$$$ coming to him for a sellout and convinced the board of this. Now we see Pfizer looking to sell off animal health and OTC, which they previously sold from their own portfolio pre-Wyeth, then crowed about how great it was to acquire these when they bought Wyeth. The rudderless portfolio mgmt at Pfizer to me is just stunning.

Permalink to Comment

42. TJMC on August 11, 2011 11:43 AM writes...

# 40 CMCguy - I agree with the caution that general MBA programs did not frame Pharma/Biotech properly in the past. May be changing - but does that help now? The "black swan" event is a recent metaphor that MAY prove of use to guide strategy. However, much of the sturm and drang between these groups seems to be rooted in two things.

One is whether you can measure useful aspects of R&D. Development seems to be apt, while the argument rages on for Discovery and preclinical. I think that any inability to create useful measures just reflects a failure of creativity - whether on the science or the MBA sides.

The second is the timeline to measure across. Many rail against recent stupid moves. While much of that comes from legitimate pain, the same folks argue that past practices were best. This is despite the indication that practices from the 1990's may have contributed to the recent dry spell (apart from pricing and regulatory which act far more quickly.) Make your mind up or at least be consistent.

Permalink to Comment

43. Rick on August 11, 2011 1:51 PM writes...

TJMC, #42
The answer to your question, "...does that help now?" is of course no. But let's face it, there really is no business solution that can possibly help now. Guys like me (PhD researcher and research manager with 15+ years experience, laid off in the past couple years) are, on average, beyond hope of ever returning to their former careers. We're yesterday's news, industrial waste, collateral damage, rotting stinking useless hulks of half-digested offal, about as useful as [insert epithet here] and no one who is in any position of real power gives a flying f*** at a rolling donut about it.

I know it's hard, but it's better if we abandon the search for short-term solutions that might help us and focus on solutions that might bear fruit for the generation-after-next (Yes, I'm saying the situation is generally hopeless for the current generation of younger researchers, or "remainders"). The battleship is just too big and too committed to barreling at flank speed in the wrong direction. Therefore, in my opinion, we'll be more productive if we focus our energies on trying to affect change at more fundamental levels in the hope that they will make a difference long after we are dead, rotting and blown away by the wind. For me, this means using what we've learned to rewrite the curriculum that taught us to screw things up. Then, future generations can move along to making new mistakes.

Permalink to Comment

44. HFM on August 11, 2011 4:15 PM writes...

What #1 said.

It doesn't matter how big your portfolio is if all your risks are correlated. There are going to be industry-wide downturns - right now, the easy targets are gone and the hard targets are still hard - and even if you own the whole industry, you're going to feel it.

I'm not sure the MBA goons are wrong about research being a cost sink, at least for the next handful of years. Of course, what they plan to do about it is less than helpful - rearrange the deck chairs, collect their bonuses, leave. Not sure what the right answer is, though. Go into hibernation, sell vitamins, and let public funding / academics carry the torch? Turn into a VC fund, or a clinical trials contractor?

Permalink to Comment

45. pharmawannabe on August 11, 2011 7:37 PM writes...

Drug Discovery Today had an editorial piece about alternative business models for drug discovery

Permalink to Comment

46. Bernard Munos on August 15, 2011 3:05 AM writes...

To provocateur (#36)

Sorry for this delayed reply. I was off-line for a couple of days.

I like Novartis because it incarnates the ethos of the pharma business: do what's right for the patients. Sadly, this has often been a casualty in the quest for blockbusters. Novartis does dazzling science, focuses on breakthroughs, and has fostered an innovation culture that has made it the most innovative company since it was created in 1996. It is evidently doing something right. If you look at its R&D spend per NME, however, it comes out at about $4 billion (calculated over the last 10 years). This is a lot better than most big pharmas, but it is still a lot, and ought to be a concern to its CEO. How do you show support for R&D while at the same time bringing down R&D spend per NME? Mr. Jimenez' answer seems to be to raise overall R&D spend while asking his R&D managers to tighten their belts and make do with less (as I have heard from Novartis employees). It is clever, and consistent with what I have been recommending. Contrary to what you suggest, I am not against R&D spending, only against useless R&D spending. I throw in this category clinical trials of compounds that have none of the hallmarks of breakthrough treatments, efforts to skew R&D to produce 'successors' to existing blockbusters (with most often disastrous results), money expended to regiment R&D through six sigma, etc. If you have followed my posts on this blog, or read my recent papers in InVivo or Science Translational Medicine, you should know that I have been arguing these points consistently. No double-talk, and no conflict of interest there.

As for Lilly, it is a big company that has done a lot, probably more than the rest of the industry combined, to enable open innovation. Yet, it was not an easy, or painless transition. Some people, including the former CEO, never really got it. Some still don't. When I published my Nature paper on open-source drug R&D in 2006, the concept was alien to the industry, and most people had little use for it. Today it's everywhere. It takes a lot of persistence, as well as a thick skin, to get a company to change its business model. Lilly's troubles do not stem from my 'recipes' as you put it, but from sticking far too long to a success formula that was failing, and made worse by the interference of non-scientists with the R&D decision-making process. This resistance to change, even as patent cliffs were looming, is surprising, but unique to Lilly. Novartis stands out again, however, as one company that went against the grain, and refocused early on patients and breakthroughs, while many competitors wasted valuable time chasing blockbusters and other fads. You've got to give it credit for showing leadership.

Permalink to Comment

47. Bernard Munos on August 15, 2011 9:48 AM writes...


In my previous post (#46), the following sentence towards the end of the second paragraph should read: "This resistance to change, even as patent cliffs were looming, is surprising, but NOT unique to Lilly".

Sorry for the mistake.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry