Michael Bloomberg, the American billionaire, has his own TV channel and news services like e-mails, devoted mostly to economic issues. Some of the information they put out, including opinion pieces, can be very revealing on the nature of international capitalism. We are reprinting here in its entirety, an opinion piece by one of Bloomberg’s regular writers, Matt Levine:
In 2017 the World Bank issued some pandemic bonds. Investors who bought these bonds got a high interest rate, but they could lose all their money if there was a pandemic: The bonds would “trigger” if a pandemic occurred, and then, instead of paying back the bondholders, the money would go to the World Bank to fund relief efforts. The bonds “were originally conceived as a sort of public-private partnership to get insurance investors to assume some of the risk of the Ebola epidemic.”
Obviously, the current coronavirus outbreak is of interest to pandemic-bond investors, and John Dizard has a fascinating column in the Financial Times about the virus and the bonds. There are a lot of classic financial-engineering lessons in these bonds.
Dumping Ebola bodies in Rwanda
For instance, I just said that the bonds trigger if a pandemic occurs, but what exactly does that mean? Well, there’s a long document with specific provisions describing what does and doesn’t qualify as a pandemic, and—as is so often the case in financial-contract triggers—there is room for debate and gamesmanship. So it’s not a pandemic unless there are deaths in multiple countries—a devastating epidemic in one country would not trigger the bonds—and so you get this gruesome speculation:
As one London underwriter comments: “They only needed 20 bodies on the other side of the Congo border to get to the trigger. What if someone loaded up a truck and dumped those in Rwanda?”
To be fair, I asked much the same question – “Do you think that anyone at the World Bank was a little tempted to smuggle one Ebola patient over an international border to trigger the Ebola cat bonds?”—last year. So that is sort of standard financial thinking applied to global pandemics. But then there is this:
Insurance people, though, are remarkably chipper about the future of pandemic deals. After all, as that London underwriter says: “Nothing is uninsurable; you just need more data. And life risk gives you more data than earthquakes.”
The real problem with underwriting pandemic risk, he thinks, is that it tends to be correlated with financial markets. “You don’t get the diversification offered by hurricanes and quakes.”
See, the problem with a global pandemic, for financial investors, is not that it’s bad, it’s that it’s correlated. If there’s a global pandemic then all the stocks and bonds will go down; if you own pandemic bonds, you will lose money on the pandemic bonds at the exact same time you are losing money on your other stocks and bonds.
You lose your money if there is a catastrophe
Hurricanes and earthquakes, meanwhile—yes of course you can invest in hurricanes and earthquakes, those are called “catastrophe bonds”—tend to be less correlated to global financial assets. If you buy a catastrophe bond and there are a lot of hurricanes, your cat bonds will trigger and you’ll lose money, but you might be making money on all your other stocks and bonds because the global economy is not primarily determined by the frequency of hurricanes.
Conversely, if there’s a recession and you’re losing money on all your other assets, the cat bonds might still perform well, because the frequency of hurricanes certainly isn’t determined by the global economy.
If you are an investor, the hurricane profile is desirable; it gives you a diversification benefit. As with every financial asset, you do well in some states of the world (no hurricanes) and poorly in other states of the world (lots of hurricanes), but the states of the world are different from the ones that determine the performance of other financial assets (recession vs. no recession, etc.).
…buying bonds for the wrong disease
You can improve your risk-adjusted return by combining uncorrelated assets. Giving investors an uncorrelated asset, one whose performance has nothing to do with the performance of normal financial assets, makes them better off. I think often about issuing a bond that pays a high interest rate if I flip a coin and it lands on heads, and defaults if it lands on tails. “Ooh uncorrelated risk,” people will say, and rush to buy it.
One way to read the pandemic bond story is that the risk of an Ebola outbreak spreading from Congo to Rwanda does not seem likely to be highly correlated with global financial markets, but the risk of a deadly coronavirus spreading from China to the rest of the world probably is pretty correlated with global financial markets: The actual coronavirus is shutting down factories, disrupting trade, and generally causing large economic impacts, even as it is also risking triggering the bonds. People may have bought these bonds with the wrong disease, and the wrong correlation model, in mind.
Bloomberg e-mail, February 10, 2020
