Financial engineering for funding drug research
Concept
A group of people from MIT’s Sloan School of Management have put together a proposal for using financial engineering to get across the valley of death in drug development. The pitch is, approximately, to securitize a batch of drugs, and then re-securitize them after each stage of FDA trials. By using these securities, it becomes possible to build a fund which will finance a slew of drugs via both debt and equity, with rates of return comparable to things regular financial markets invest in. This makes it possible to put the financial might of mutual funds, retirement accounts, and people interested in hedging like medical insurers, behind drug development.
Megafunds
This takes a lot of money. A lot of money. In the original paper they modeled drug development as an investment of $200M, which after 10 years had a 5% success rate, with an average return of $12.3B on success (which number is the expected present value of $2B a year for 10 years, which itself is the expected length of monopoly guarantee). This is a heavy-duty risk, and most investors are unwilling or unable to take it. While it is clearly still worth it for drug companies, there is only so much money they can invest, and so the total investment is constrained. In the basic model they use 150 independent drugs, which means a megafund of 150 * $200M = $30B. The less independent the drugs, the more capital is required. So they need a way to distribute the risks such that they have access to this huge amount of capital.
Securities
The tool that they have developed for this is the Research-Backed Obligation, or RBO. Quoting from the original paper linked above:
A common form of securitization involves “cashflow” transactions in which a portfolio of assets—typically mortgages, auto loans, student loans, or credit-card receivables—is acquired using money raised by issuing equity and bonds of different seniorities. These assets and the cashflow they generate are pledged as collateral for the debt. In our proposed application, the assets include the initial capital raised from investors, all the subsequent biomedical R&D and licenses acquired, and all the profits generated by these activities or through sales of these assets in later periods. The application of securitization to early-stage clinical and preclinical biomedical has not been described previously, and we shall refer to debt that is collateralized by such assets as ‘research-backed obligations’.
For most of their simulations they have the security broken down into senior-tranche (the debt which gets paid first) junior-tranche (the debt which gets paid last) and equity-tranche (the people who own equity, and get everything that doesn’t go to debt).
Benefits
According to their short paper Can Financial Engineering Cure Cancer?, which simulated using these tools, the megafund generated average annualized returns of 5%, 8%, and 9.1% for senior, junior, and equity respectively. These averages are within the range for institutional investors, like mutual funds and pensions, and the risks are low enough to finance with credit. They compared a megafund which was all equity to two different leveraged ones, and via leverage managed to increase the number of drugs funded from 63 to 103, which is the real payoff for the rest of us.
Here is a Ted Talk from Roger Stein, and a TedX Talk from Andrew Lo, if you prefer a video summary (though they are short on details). Aside from the original paper they have proposed the cancer fund in the short paper above, as well as how it might work for orphan diseases and hedging medical insurance risk.
Is this very different from founding a pharmaceutical company?
Yes—this fund requires pharmaceutical companies to generate the IP in the first place, and also to sell the successful drugs. A new pharmaceutical company will face the same risk profile as existing pharmaceutical companies; I would be very surprised if one could suddenly start investing according to the opposite pattern the others use.
On the other hand, I don’t see any reason why an existing pharmaceutical conglomerate could not employ this strategy or a similar one. They already have a huge amount of IP laying around undeveloped (it is from them a fund like this would acquire it) and other huge companies like General Electric have deliberately explored financial engineering as a corporate strategy. It failed in that case, but in this one we are just talking about supplementing the core strategy rather than replacing it.
Andrew Lo’s website is here.
Roger Stein’s website is here.
It seems that most of the code and data for the simulations is available from one of these two places, but I haven’t verified any of it myself. In the original paper they explicitly use very simplified models to demonstrate the concept, which makes sense to me because there are a lot of different layers to the problem; but I feel like the details of how the risk is simulated are very important to the results going forward, and I don’t have a clue what the conventions are for that. Or if the conventions are good.
That being said, it seems like a good case for application of risk distribution in general. A huge amount of problems seem like they would be reduced if we followed the dictum of insuring inevitable costs and securitizing necessary assets.