New proposals for funding pharmaceutical research and development are of interest to me since the current, market-, and high-profit-driven system has failed to produce affordable drugs for diseases afflicting many of the world’s citizens. So when I heard that Andrew Lo, a well-know MIT economics professor, hedge fund founder, and financial engineer (see his bio in Wikipedia), and colleagues had published a new scheme in Nature Biotechnology, I gave it a read [“Commercializing biomedical research through securitization techniques” (Fernandez et al. 2012)]. At the center of their scheme is a “megafund”: a large, diversified portfolio of biomedical projects (like 150) at all stages of development which would be supported by the $5-15 billion fund, whose investors would own equity (shares) and/or debt (bonds) and repaid when the projects’ products are licensed. I understand how the megafund reduces risk by funding a large pool of projects and how the potential licensing/royalty income of the successful projects are assets, but not how “financial engineering” makes the risk low enough to attract investors. The authors explain: “As a result, these programs are more readily packaged for a much larger population of investors via financial-engineering techniques such as ‘securitization’: a financing method in which a pool of investment capital is raised by issuing equity as well as several classes of bonds that differ from each other in their risk-reward profile to a diverse population of investors, and where the funds are used to invest in various assets that serve as the collateral for the bonds.” OK?
Of course, the authors, being good engineers, provide numbers and examples to fill in the outlines of the megafund. They point out the differences between their fund and a venture capital fund (the latter doesn’t use debt), a major pharma company (uses debt but avoids early-stage projects), and a mutual fund (invests only in public companies) and that existing examples are the firms that buy future royalty streams, the biggest being Royalty Pharma (Royalty Pharma) which has assets of $8 billion of which half is securitized debt based on the royalty income of approved drugs. The authors also address some of the details wherein the devil (challenges) dwell, such as, how the fund would be structured and managed, and present a “detailed simulation of a hypothetical funding vehicle for cancer drug development programs.” Cancer is used because, as the authors note, it is a major cause of mortality in the US and rest of the world, it is more than 200 diseases (therefore has multiple targets although recently there is evidence each individual’s cancer may have 100s of targets), and there are a couple of databases of cancer drug development programs, successful and not. After plugging in lots of numbers and assumptions (like for development costs and phase transition probabilities) and turning the crank of their simulator many times (like 500,000 per), they present results of two stimulations, both of which yield respectable, if not attractive, returns on investment ranging from 7.2 to 11.4% from funds ranging from $2.5 to 15 billion over a ten-year life. This all sounds good to me since my investment in Click and Clack’s Capital Deprecation Fund which is not doing well.
But before going on to my real interest, the application of the megafund model to drug development for global diseases, I pick a few minor and two major nits with the scheme. Here are a few operational challenges the fund’s management may face:
- identifying the fundable projects and/or promising principal investigators;
- executing the many licensing deals, especially those with the notoriously uncooperative university technology transfer offices;
- designing and implementing criteria for terminating as early as possible the many projects unlikely to succeed; and
- managing the drug development process as the projects transition from academia or biotech companies into contract research organizations/companies which are needed to do the translational work most cost-effectively.
However, two nits are bigger. The first is mentioned in passing by the authors, that the megafund may use/need “credit enhancements” to quell queasy investors which include credit-default swaps, government guarantees, and tax incentives. This sounds like the witches brew that led to the recent financial mess to me. My second nit is that the valuation of the fund’s assets requires competent analysis of lots of data on the performance of those assets, like how many projects will succeed, how long it will take and cost, and what their products’ revenue will be. I can’t imagine the bond rating agencies (whose professional reputation took a big hit after the 2008 melt-down) will be able to figure out what the fund’s assets, and therefore their bonds, will be worth. Clearly, the pharmaceutical industry and its legion of investment analysts has had a tough time figuring out which products will be successful as shown by the a large number of late stage failures recently (FierceBiotech article). The authors reply to this problem is: “greater risk, even unknown risk, can often be managed effectively through more sophisticated financial engineering.” I’d like to know more about managing the unknown or at least where these guys are putting their retirement savings. I did not find much response to the article by those in the finance biz (probably not readers of Nature Biotechnology), but did note some eye-brow-raising about its practicality in an opinion piece by Felix Salmon of Reuters (Reuters opinion).
Could the megafund idea be applied to the funding of drug and other product development for global health? Lo and his colleagues imply that it could, in that may be attractive to large institutional investors like pension funds, sovereign wealth funds, and foundations which want managed risk, don’t mind lower returns, and, sometimes, want to “do good” with their investing. Along those lines, Bill Gates, whose foundation has been a major source of funding for improving global health, mentioned creating an international infrastructure fund among other ways to encourage private investment in development at the 2011 G20 Summit ( G20 report summary). As for suitable projects to be funded, although research on rest-of-world/neglected diseases receives a fraction of overall US government biomedical funding, there are thousands of projects underway throughout the world (14,570 according to the G-Finder database) and some in later stages of development, for example, those managed by the grant-funded product development programs (PDPs) like Medicines for Malaria Venture and the Global Alliance for TB Drug Development. In addition to new product development, a megafund for global health could also fund research to make existing vaccines and drugs, like injected cancer chemotherapies, useable in under-resourced settings. The major problem is that the megafund’s potential revenue stream would depend on the world, or more specifically governments, figuring out how to pay for providing a minimal level of health care, of which medicines and vaccines are one part, to all their citizens. Perhaps, governments or government-backed insurance programs could become fund investors and in return get guaranteed pricing for any products that result. And as I tried to show with my simplistic financial model in my post “Rolling the Dice”, if one assumes a modest price for a drug paid over time (like through insurance) by the many millions of people at risk, the possible returns justify a reasonable level of investment.
I buy into Lo et al.’s basic premise that “by creating financial incentives for solving social problems like cancer, society is able to tap into a much larger pool of assets” and solve those problems. I think it would be mega-fun if they applied their financial engineering wizardry and hedge-fund smarts to figure out a Megafund for Global Health that would be safe enough to be part of anyone’s 401K portfolio. As we say at MIT, the solution may be trivial.