Every venture exists to solve a problem. Since the dawn of the Covid-19 pandemic, pharmaceutical companies and investors have been scrambling to solve what has quickly become one of the biggest problems in modern history. But as the Financial Times observed earlier this month:
“For most biotechs, success will be elusive. One winner will take most of the market. A best-case scenario is being priced in. Getting closer to a cure is good news for patients. For investors, the gamble does not bear out the risks.”
Despite the benefits of diversification, the scale and risk of some problems is so vast that their solutions can’t be addressed by existing financial theories like the MPT or existing structures like VC funds. The problem facing us right now isn’t new though. John Hull of the University of Toronto along with Andrew Lo and Roger Stein of MIT discussed this very problem, prophetically, early in 2019 in a paper they called “Funding Long Shots.” In fact, regular attendees of the Global ARC conference in Boston may recall Lo discussing the trio’s idea back in 2016.
Hull, Lo and Stein argue that modern portfolio theory leads to market failure when the scale of investment, the probability of success and the duration of the payback exceed a certain threshold. In other words, VC investors can only handle so much. Initiatives beyond their capacity, such as finding a cure for cancer or solving climate change can best be tackled by government since the payback from such investments mainly come in the form of positive externalities (i.e., they are so disperse that they can’t be adequately captured by investors).
They propose that research into these problems should be securitized in the same way that we securitize debt in CDOs, or funds in CFOs. In a nutshell, they suggest that a “research-backed obligation” (RBO) could be created that would provide different risk tranches and payouts to different investor classes under different scenarios. This could appeal to a range of investors, not just those envisaged by the Financial Times (above) who aim to “gamble” on the discovery of a Covid-19 vaccine. After all, the US debt market is around $40 trillion while the US VC sector is under $500 billion.
You’d be excused for wondering why this idea is particularly well-suited to major society-spanning problems like the search for vaccines. After all, why not securitize every kind of new venture “long shot”? It turns out while traditional equity diversification strategies (i.e., VC or venture debt) work for smaller investments whose benefits can be captured by investors, they suffer from the whims of equity markets. The trio also cites this paper which finds that while start-up returns have a positive skew (like buying a call on the equity markets) VC funds actually have a negative skew (like selling a put on the equity markets). When you adjust for the fair value of that put, VC fund returns actually are negative. Thus, when markets go into a tailspin, the performance of most equity long -hot investments tends to correlate (witness the performance of VC funds in Q1 this year).
But according to Hull, Lo and Stein, the cross-sectional correlation of drug discovery research is more stable. This doesn’t mean that research initiatives are uncorrelated. In fact, different research initiatives based on the same underlying concepts would tend to be quite correlated. But at least it means that these cross-correlations will be more stable over time than those of traditional equity investments. This makes them well suited to securitization (unlike, say, mortgage-backed securities which clearly were not uncorrelated during the global financial crisis).
The incremental value of securitization (over traditional VC) for long-shot investments also comes from removing the proclivities of the VC manager from the equation according to the three researchers:
“The investors in VC and private equity funds may be well-diversified, but the funds themselves are not well diversified. A significant part of a fund manager’s compensation is related to the performance of the fund. Furthermore, the contract with investors is set in advance of deals being signed with entrepreneurs to acquire firms for the fund portfolio. This creates a principal-agent problem in which idiosyncratic risk is rationally priced by the fund manager even though it would not be priced by the fund’s investors if they were negotiating terms directly with entrepreneurs.”
Like any asset-backed security, CDO or CFO, a “CRO” would have a variety of unique features like the timing of cash flows, staged monetization, and of course, the timing, likelihood and scale of the drug development costs themselves. A CRO could be used for rare diseases (i.e., orphan drug development), Alzheimer’s, or cancer.
Which brings us back to our current mess. It may be that the positive economic externalities of a coronavirus vaccine are simply so vast that governments need to become significant co-investors in CROs. According to Hull, Lo and Stein, the math laid out in their paper could be used to distinguish CROs that can be funded entirely by the private sector from those where government can “bridge the financing gap.”