Skip to Content

Business and Markets

A Totally New Way to Finance Drug Discovery

Now, here’s a big idea. Thirty billion dollars worth of big idea. Andrew Lo, professor of finance at MIT (Sloan) and hedge fund manager, along with Jose-Maria Fernandez (Sloan) and Roget Stein (Sloan and Moody’s) propose raising that much money for discovery-stage oncology research. But he’s not running a fund-raising appeal for a charity; he wants to raise that money as an investment:

Here we propose a financial structure in which a large number of biomedical programs at various stages of development are funded by a single entity to substantially reduce the portfolio’s risk. The portfolio entity can finance its activities by issuing debt, a critical advantage because a much larger pool of capital is available for investment in debt versus equity. By employing financial engineering techniques such as securitization, it can raise even greater amounts of more-patient capital. In a simulation using historical data for new molecular entities in oncology from 1990 to 2011, we find that megafunds of $5–15 billion may yield average investment returns of 8.9–11.4% for equity holders and 5–8% for ‘research-backed obligation’ holders, which are lower than typical venture-capital hurdle rates but attractive to pension funds, insurance companies and other large institutional investors.

Here’s a Boston Globe story on the idea. Lo and his co-authors note the low productivity of drug research in recent years (which he doesn’t seem to think is a scam!), and its increasing costs. At the same time, there have been many scientific advances in areas that you might have thought would have helped, but here’s how he reconciles these trends:

Here we propose one explanation for this apparent inconsistency and a possible solution. Our proposed explanation is the trend of increasing risk and complexity in the biopharma industry. This trend can be attributed to at least two distinct sources: scientific advances and economic circumstances. That biomedicine is far more advanced today than even a decade ago is indisputable, but breakthroughs such as molecular biomarkers for certain diseases generate many new potential therapies to be investigated, each of which requires years of translational research at a cost of hundreds of millions of dollars and has a substantial likelihood of failure. Although such complexity offers new hope to the afflicted, it also presents an enormous number of uncertain prospects that must be triaged by researchers, biopharma business executives, investors, policymakers and regulators. . .the lengthy process of biomedical innovation is becoming increasingly complex, expensive, uncertain and fraught with conflicting profit-driven and nonpecuniary motivations and public-policy implications. Although other industries may share some of these characteristics, it is difficult to find another so heavily burdened by all of them.

Hard to argue with that. He then goes on to one of the same questions that’s been discussed around here – the effect of the stock market on a drug company’s behavior. If the quarter-by-quarter focus of most investors is inappropriate (or downright harmful) when applied to an R&D-driven company with timelines like the drug industry’s, and if private equity doesn’t have the cash to invest on that scale (or the willingness to take the expected returns even if things work out), what’s left?
That’s the idea here, to provide something that currently doesn’t exist. The idea is to finance things via securitization:

Our approach involves two components: (i) creating large diversified portfolios—’megafunds’ on the order of $5–30 billion—of biomedical projects at all stages of development; and (ii) structuring the financing for these portfolios as combinations of equity and securitized debt so as to access much larger sources of investment capital. These two components are inextricably intertwined: diversification within a single entity reduces risk to such an extent that the entity can raise assets by issuing both debt and equity, and the much larger capacity of debt markets makes this diversification possible for multi-billion-dollar portfolios of many expensive and highly risky projects.

These debt instruments will have longer time horizons, which can be tailored a number of different ways. Securitization can provide a whole range of different bonds to be issued, with different maturities and different levels of risk. The fund itself would earn its returns from the sale of whatever assets its funded projects generate – outright purchases by larger companies, milestone payments, royalties, whatever. With enough diversification, Lo et al. think that this could work, if some cost savings kick in as well:

Compared with the plethora of small pharmaceutical companies currently pursuing just one or two projects, these savings are especially important for a megafund. It is considerably harder to cull compounds efficiently in a small company because the livelihoods of the employees and management depend on the continued development of the company’s few compounds—in these cases, development tends to continue until the money runs out. With a megafund, this conflict is greatly reduced—capital can be more efficiently allocated to projects that are likely to succeed, and failing projects and compounds can be abandoned rapidly. In fact, for megafunds that have invested in a sufficient number of early-stage projects, it may be worthwhile to build and operate shared facilities for conducting preclinical studies motivated by the megafund’s projects. Such a ‘preclinical incubator’ could provide the megafund with valuable economies of scale as well as reduce duplicative costs in the industry.

Now, this idea is fascinating, but it raises several big questions. Readers with some knowledge of the financial markets will have noted that this whole securitization-and-repackaging process was one of the main engines of destruction during the recent financial crisis. (I continue to recommend Michael Lewis’s The Big Short for details on this). Vast amounts of mortgage-backed securities were generated, and their risks were, it is fair to say, poorly evaluated. The paper explicitly addresses this problem, with suggestions on how to keep things from getting out of hand, but this is something that will have to be watched carefully. Securitization, the authors note, can almost be too efficient a way to raise capital.
The rest of the article is a detailed look at the idea through the lens of portfolio theory, along with some simulations of how it might have worked in the past. I strongly recommend that anyone who finds this idea interesting check out these details, but they’re beyond the scope of an already-lengthy blog post. I note that Felix Salmon has looked at this from a financial writer’s point of view. His take is that this is quite possibly a worthy idea, but he has doubts. For one thing, the proposed “megafund” might find it difficult to pay investors in its early years, and might be forced to make some bad decisions in order to do so. He’s also skeptical that the training-set period used for Lo’s simulations is representative of what we might expect in the future.
But Salmon’s biggest objection is that the idea might well prove unworkable even to test. A smaller version of such a fund would lose many of its advantages; it has to start off large or not start at all. And he’s not so sure that anyone can raise that kind of money for something that’s as big a change as this would be. There is, in chemical terms, too high an activation barrier.
I’m still thinking about all this myself; there’s a lot to think about. Take a look and see what occurs to you – I think that I can guarantee that you’ll have some strong opinions, because this is one of those proposals that it’s hard to be neutral about.
Update: Nature Reviews Drug Discovery weighs in with a detailed assessment. The article is cautiously optimistic, but wonders if more money will really do the trick.

26 comments on “A Totally New Way to Finance Drug Discovery”

  1. Wile E. Coyote, Genius says:

    Where is the economy of scale for reduced development costs for this large effort? Programs are scattered about amongst a large number of little efforts, each duplicating resources that may be shared if co-located, such as in a pharma company. No critical mass. Sounds to me a lot like another version of the FIPNet model, where everything is outsourced. I think the jury is still out on FIPNets.
    Also, I’d hope that there are serious scientists with extensive drug development expertise and experience high up in the organization that could veto moves made by the MBAs that don’t understand drug development.

  2. Morten G says:

    Hmmm… maybe. My idea has been for a long time that biotech/small pharma sell too cheaply to big pharma for the sector as a whole to be profitable. But if the investors were better organized and invested in more companies they would be better at evaluating the actual value of something that looks successful (or like it will be successful). Big Pharma would have to pay more.

  3. Phagonizer says:

    I lost it at “With a megafund, this conflict is greatly reduced — capital can be more efficiently allocated to projects that are likely to succeed, and failing projects and compounds can be abandoned rapidly.” Like that happens in the megafund that is a big pharma company

  4. bbooooooya says:

    “I’d hope that there are serious scientists with extensive drug development expertise and experience high up in the organization that could veto moves made by the MBAs that don’t understand drug development.”
    What’s not to understand? For every 10,000 molecules you make 1 of them makes it to market. All you have to do is choose the right disease (say Alzheimer’s, that seems promising). the reports I have from McKinsey and BCG say so, so it must be true. If we want 2 drugs, we’ll make 20,000 compounds.
    Drug discovery is so easy.

  5. weirdo says:

    This is a fabulous idea. No, it really is. It would be better if Big Pharma had skin in this game, too, but that’s another business model (and another discussion).
    How many of us in Big Pharma have worked on projects to a certain point only to have the “strategy” change? Good assets are put on the shelf because the current mood does not support further development. Out-licensing? Good luck. BD guys don’t get bonuses for out-licensed assets, and the immediate assumption by potential partners is that there is something wrong with the asset — after all, if it’s any good, why is it being out-licensed?
    Now, here is essentially an organization with the funding of a Big Pharma (through a certain point at least — Phase 1? Phase 2a POC?), but who can go shop the cool assets they find to the highest bidder. If they are any good, they will do fine. More than fine, I’m guessing.
    And, sorry Morten G., regardless of what you read by Kevin Kinsella, Big Pharma generally overpays for most assets. There are few bargains out there, and it is NOT the job of Big Pharma to make sure venture capitalists “get theirs”.

  6. Shanedorf says:

    I’m sure that this will be very good for the money guys, but it won’t change the success rate of drug development. Seems like all they are doing is switching pockets, not unlike the tsunami of deals that moved the risk away from Big Pharma to their CRO partners. The fundamental problem is never addressed, just who pays for it.
    New accounting / securitization methods just lead to the next bubble, and the money guys walk away with their bonus checks leaving a trail of failure behind. Pretty sure I’ve seen this movie before. Lather. rinse. repeat.

  7. homer says:

    Hang on…this sounds like setting up another Big Pharma, but with the R&D done in small, incorporated units. That is J&J (or similar). I love the idea, but have to ask myself why would it be more successful than Pfizer, Merck, J&J right now? They issue stock, bonds, pay dividends, invest in multiple angles, expecting some to pay and some to be write-offs. I guess the current plan here by Prof. Lo is to have Finance as a separate corporate entity–like a giant Venture Fund, leading investments external R&D programs rather than bear the load of internal fixed costs. Hmmmmm.

  8. milkshake says:

    funding pools of biotech startups has been done before – from what I hear it does not work so great because necessary expertise for supervising investments into a larger pool of biotech startups is hard to get – you depend on outside expert opinions but even then you can easily get fed propaganda how all these projects are looking great (and in the meanwhile irreproducibility/contradicting results problems are not adressed and failing drug candidates are kept in late-stage clinic trials artificially long so that the top management has time to cash out their stock option before the whole thing comes crashing down).
    Like with any other investment, you need to know very well the people in charge of the company that you are investing in so that you can gauge their integrity. It helps to give them the right long-term incentives, not just a pile of stock options. It helps if the people in charge of the biotech you are investing in have a large amount of their own money and their family money already sunk in the company and stand to lose great deal if the company fails – this helps to keep them frugal and honest

  9. Esteban says:

    Securitization itself is not inherently risky. The problem with the mortgage-backed securities was that many were given AAA ratings based on a flawed risk assessment model. I think it’s fair to assume that the proposed endeavor would not be getting a AAA from any ratings agency.
    What I’d really like to see is a change to how compound advancements get done. I’d love to see someone try a prediction markets-type approach where drug discovery scientists get to cast votes in some manner. Yes, there will be bias towards one’s own projects, but that bias should cancel out across projects and so it will be the teams that make the best case for their compound(s) to the wider community of scientists in the organization that will carry the day.

  10. Anonymous says:

    This will just lead to ‘shots on goal’ gone wild!!
    The more money, the more targets will be progressed but the likelihood of any one project succeeding will decrease exponentially (if your top 3 ideas don’t work, why do you think number 30 will?? STOP!!).
    It is exactly like selling mortgages to people who will never pay them back. Sub-prime projects will dominate yet the hype for many years will be about how brilliant all these projects are. Some people will get out early and make a lot of $$$. Everyone else will lose – but I guess at least the taxpayer won’t have to pick up the cheque this time…..

  11. Twelve says:

    So what’s to keep a researcher from keeping the really great projects out of the pool and having them resourced separately, while the dreck gets peddled to these guys? Would there be a mandatory all-or-nothing commitment to participation?

  12. Yancey Ward says:

    Count me greatly skeptical. Pharmas are not limited to equity financing, and never have been.
    What I see is a big pile of money that will be pissed away, should it ever actually get accumulated on the scale Lo seems to need to get started. I will give you one guess where the start up funds will come from.

  13. portia.vz says:

    How much money do we spend on Iraq and Afghanistan each week?
    Eventually, the government is going to have to step in and provide seed money and/or structure a private-public partnership. Consider it a public investment.

  14. Milton Weedman says:

    Want to hear a well know fact? Wall Street makes something like 80% of their profits from fees. Fees for setting up the fund, fees for maintaining the fund, fees for selling the fund.
    So beyond the revenue bankers will make in selling these new securitized product to pension funds, I don’t see how this is any different than Pfizer buying or investing in twenty small companies. Yes, Pfizer can issue bonds too.
    THIS IS A SCAM! Due to low interest rates on most govt debt and a correcting global economy (the BRICS are dying), these shills for Wall Street are looking to line their own pockets. BEWARE!

  15. drug_hunter says:

    Intriguing idea. I’m a little unclear, though, on why this structure would have a significant effect on the largest source of failure in cancer, namely, our lack of understanding of disease biology. When I look at the best cancer shops, e.g. Novartis, I think they are already applying the biological knowledge we do have in a pretty efficient way. (No, I don’t work for Novartis.) Yes, I do understand that it would be easier to kill boneheaded projects, and that will help a little — but it will not help that much because we will still be left with lots of projects that all look interesting but have no clinical validation, and most of them will ultimately fail in the clinic.

  16. SteveM says:

    This is just another variation on intelligent portfolio management. Most portfolios fail because junk projects are not killed early enough.
    The argument is simple enough, small independent bio-techs will run a bad idea into the ground and they go broke. And the cash that chased the idea just disappears into the investment netherworld.
    The mega-fund supposedly overseeing many initiatives would kill the weak ones sooner and rather than have the investment cash disperse, it would be retained and redirected in the bio-tech domain.
    That ain’t rocket science. As mentioned by others, that has been the historic Big Pharma portfolio management model. The sticking point in securitization would be paying back bond holders when development timelines are a decade or more. I.e., how to handle cash flows?
    In the context of investor short attention span, that may be just a fact of life that can’t be circumvented. Which implies that long lead time R&D investing is being made practically infeasible regardless of the business model that would try to support it.

  17. newnickname says:

    Apologies in advance: I didn’t read the articles. I suggest “Manhattan Projects”.
    Before medicine, I’d “Manhattanize” solar energy to the point of actual implementation. Just solar, NOT wind or anything else. Photovoltaics, energy storage and distribution.
    After 3 or 4 years (maybe 5), that problem would be solved and THEN we could Manhattanize the search for solutions to particular problems in health and medicine.
    As the first medical Manhattan Project, I suggest finding a drug for the treatment of delusional thinking.

  18. Hap says:

    16: The last problem will get solved, eventually – if you can’t make money doing anything useful, eventually you just won’t make any money or do anything useful (other than, maybe, surviving) until someone finds a new game to play.
    Probably won’t help us much, for an expansive definition of “us”.

  19. MattF says:

    Q1: Who’s in charge?
    Q2: Who makes money if it succeeds and who loses money if it doesn’t?
    Q3: Who bears the risk?
    Q4: Who bears the liability?
    Q5: Who owns the patents?

  20. sciencemonkey says:

    @17: I agree, oil won’t last forever so we need to get on that energy problem. Next Manhattan project should definitely by low-cost, easy, reliable, and widely utilized birth control methods for both genders.

  21. MBA Hater says:

    One of the best and brightest at work – again. Why the F$%^ these people even have jobs just baffles me. The securitizaton experiment was one giant failure with morgages.

  22. SteveM says:

    Re: 17 It’s not like DoE has been sitting on its haunches. I supported the DoE Office of Energy Efficiency and Renewable Energy as a contractor. The technology breakthroughs have been marginal due to the limiting nature of the science.
    People have a tendency to assume a “Moore’s Law” rate of technology development is possible for any domain if only enough money were thrown at it. But energy R&D is not computing. The physics and chemistry of energy are well understood, or rather the constraints of physics and chemistry are well understood. Thermodynamics is what it is and the probability of synthesizing transformationally energy efficient materials is very small. Moreover, you can’t miniaturize a BTU like you can miniaturize a data packet.
    I’m not knocking investment in energy R&D but I’m not anticipating any breakthroughs either that would make it competitive with fossil fuels.
    Back to my points about a bio-tech mega-private equity fund, the investor mindset right now is to throw billions at IT start-ups with the hope of Facebook returns in only a couple of years. The patience for Pharma-like return profiles just ain’t there. Hence, reactive slashing and burning would probably not be attenuated because the R&D portfolio was owned and managed by private equity rather than by Big Pharma. Those guys would probably be as stupidly ruthless.

  23. darwin says:

    The Big Short- Volume II

  24. Cellbio says:

    Late to the party, but can’t help but weigh in even though prior comments cover most everything.
    The part that I find amazing is the lack of any true appreciation for the human element. It is a reasonable argument to point out that small company staff drive their projects through a long march of death because early termination conflicts with preservation of salary, but this is a thin analysis. While true in some cases, I know of many alternative examples, lived through three myself, where far less than all the money was spent and the team said, “sorry, really nothing here, take your money back’. However, it does happen that teams act in a manner influenced by self preservation.
    It happens in big pharma, as we all know, where projects persist due to career aspirations and self-survival instincts. But somehow, just magically, this will not happen for the people doing the same work in support of the mega-fund? How is that? Oh, I know, once again the whiz kid business guy who ran a bunch of number through a spread sheet has an idea of how to limit the excesses of the barely civil research staff. Because we know, the wise guys that dream up these funds would never be so animal-like to have the decisions they make on the mega-fund portfolio be influenced by human nature. Of course, recognizing the unpure nature of research teams, they’d set predetermined metrics and place them in the work process charts and review them at portals and strive to kill fast and celebrate killing projects whose merits and challenges they never needed to waste their time to comprehend. Heck, we could even invoke martial arts analogies for our new crack team of managers (power, Ninjas…no, blackbelts, that’s it!) Too bad pharma never tried this.
    Come on. Like small company folks don’t ask themselves all the time: what is most likely to work, what has the greatest chance for near-term value demonstration.
    I am also guessing that the modeling is greatly influenced by Avastin. I wonder if that one drug was omitted from the model if the outcome would have been the same (apologies if the paper does has a sensitivity analysis).

  25. This initiative would seem to have some merit not least provided much needed finance to an area in shortage. BUT
    They are arguably superimposing a square financial peg in a round scientific hole, perhaps the hole rather than the peg needs some work.
    The idea of spreading risk using portfolio theory has the central supposition that there will be enough winners to overcome the losers and therefore pay investors back. Debt investors first and then the equity gang. The evidence he gives to support this are the costs of research compared to the enormous size of the rewards of success plus extrapolation of past historic performance in the field. All seems logical enough, but is it?
    Innovative success in drug discovery perpetually sets the barrier for later success higher. The statins made a fortune, the next challenges in heart disease (still the developed worlds biggest killer) are proving much harder to crack. This is arguably because to move death rates the challenge may well involve addressing biological complexity of the cardio system as opposed to the much simpler single contributory factor.
    To place this is in context of an investment case take a simple business opportunity process;
    1. spot the opportunity
    2. assess the risks
    3. devise a plan to exploit the opportunity which states how the risks are to be mitigated
    Investment decisions based on the outcome are made and those viewed positively go into execution (it is more complex than this but you get the idea).
    Compare this with the case made by the megafund promoters and you see there are big holes in numbers 2 &3.
    The use of securitization would be a neat way of expanding the pool of money using portfolio theory but there is a hidden iceberg. A quote from the heat of the credit crisis ‘Portfolio theory protect your herd of pigs against the loss of a few runts but is no vaccination for swine flu!’
    In the case of cancer research biological complexity is swine flu. Until the industry spells out clearly what the central scientific challenges are and how they intend how they intend to overcome them to allow probability of success to be viewed in context then they are basically asking to play by different rules compared to other industries (possible exception oil exploration but their success rate is much higher). ‘Trust us and leave it the experts, we are better than we used to be.’ is not that compelling an investment case.
    Perhaps more compelling:
    • Biological function and malfunction (disease) is driven by proteomics.
    • Proteomics operates through complex protein interaction systems.
    • New tools can interpret these systems.
    • New physics tools are needed to supply key real time in situ interaction data.
    • Using these cancer stem cells can be differentiated from their healthy equivalents and struck down selectively.
    • Network pharmacology tools have been developed capable of establishing those compounds with the correct multi target interaction attributes needed for efficacy.
    • Cells communicate in complex ways and this could also be analysed. This provides a basis for harnessing adult stems cells in degenerative disease and potentially the immune system.
    Some of this is arguably underway already but unless properly framed, in a framework of – these are the challenges and hence the risks and this is how they are to be addressed, it is difficult to see how and indeed why a beleaguered pots of financial and public capital should make a special case given the hugely high historic failure rates.
    Indeed it could be argued that the research bias toward gene studies followed by pathway analysis is actually undermining the case. This is the methodology that has so frequently tried and failed and what is the investment case for more of the same?
    Genetic studies have an important role but must be placed in the context of how biological function is derived and how cancer treatment will manifest. Geron telomerase cancer story has joined the crammed graveyard of victims of the simple single protein or gene approach.
    Lets hope new money is attracted to the industry but perhaps more work is needed on the investment case.

Comments are closed.