What is Richard Posner So Afraid Of?

The high cost of the falling sky


High-powered intellectual and federal judge Richard Posner spoke at the Washington, D.C., think tank Resources for the Future (RFF) on Wednesday about his new book, Catastrophe: Risk and Response. Posner aims to look at how policymakers and the public should respond to very low probability, very high cost events such as an asteroid hitting the earth, abrupt global warming, or a bioterrorist attack. Such disasters could just kill tens of millions or even wipe out the human race.

In his talk, Posner took a look at four different very low probability catastrophes. The first was a bioterrorist attack on the United States that unleashes a plague that kills 100 million Americans. He calculated the cost of such an attack at $1 quadrillion (100 million lives x $7 million per life + a $300 trillion pain and suffering factor). Since the Feds are spending $2 billion per year on research to avoid such an attack, he calculated that the implied annual probability is .000002 or 1 in 500,000. Posner then asserted that the probability of such an attack is much higher (how much higher he didn't say), therefore he concludes that Americans are underspending on preventing such an attack. But are we woefully underspending?

It's my intuition too that the probability of some bioterrorist attack is much higher than 1 in 500,000 annually, but what is the probability that I will be affected by such an attack? Remember that the anthrax attack in 2001 killed just 5 people and made 22 others sick. In that attack each of us faced about a 1 in a million chance of being exposed to the anthrax spores. Of course, future bioterrorist attacks could well involve infectious agents, which in a sense deliver themselves. Posner correctly noted that some day in the not too distant future, keeping smallpox securely locked up in two labs won't do much good because biotechnologists will be able to construct smallpox viruses with off-the-shelf biochemicals. Historically, smallpox has been 30 percent lethal in unvaccinated populations. Posner suggested that bioterrorists might be able to boost smallpox's or some other pathogen's lethality from 30 percent to nearly 100 percent.

But why does Posner limit his analysis to the $2 billion being spent directly on biodefense research and monitoring? Wouldn't a fair analysis also include the not-inconsiderable expenditures for intelligence and military activities that are currently disrupting terrorist infrastructure and planning worldwide? More broadly, the billions being spent on advances in biotechnology at universities and corporations aimed at curing and preventing natural diseases also provide spillover technologies that will enable us to counter bioterrorist pathogens. And the vast improvements in our communications systems like the Internet and broadcast facilities can alert people to an attack and provide them with the information needed to protect themselves from it. In addition, although it's possible that people will panic, I suspect that extensive social learning about the importance of maintaining quarantines will also aid us in preventing the spread of any supervirulent pathogens. When you add it all up, I would guess that our total anti-bioterrorism expenditures imply that we are protecting ourselves against the probability of an attack that kills 100 million Americans at somewhere between 1 in 1000 and 1 in 10,000. It's a lot harder to argue that that is not enough, especially since Posner did not offer an estimate of how probable such a bioterror plague really is.

Posner went through a similar analysis for asteroid strikes wiping out 1.5 billion people at an estimated loss of $3 quadrillion (foreign lives are cheaper than $7 million American lives). He said that astronomers estimate the annual probability of such a strike at 1 in 50 million to 1 in 100 million, whereas the implied probability of our current expenditures is 1 in 769 million. Posner also mentioned that scientists at Brookhaven National Laboratory who are running the relativistic heavy ion collider (RHIC) had initially estimated that there was a 1 in 500,000 chance that the RHIC could create strangelets, which could shrink the earth into a sphere 100 meters in diameter before causing it to explode. Strangely, Posner did not mention that subsequent calculations had shown that the probability of that mishap was far less than that.

Posner's final disaster was "abrupt global warming" of 14 degrees Fahrenheit. Posner reasonably noted that the scenarios for gradual warming over the next century actually did not imply the need for measures like the Kyoto Protocol, which would impose limits on the emissions of greenhouse gases like carbon dioxide. Why? Because considering the pace of technological progress, such long term problems might well be handled more cheaply and expeditiously by improvements in technology in 50 to 60 years. Nevertheless, Posner estimated the current expenditures on climate change to be $1.7 billion annually and the possible losses of "abrupt global warming" at $66.6 trillion yielding an implied probability of 1 in 388,000. Again, he asserted that the annual probability of abrupt global warming must be higher than that, and therefore we were once again underspending to protect ourselves against this threat.

Once more, Posner is looking solely at research expenditures aimed directly at studying climatology. He is apparently ignoring the vast sums spent on improving energy technologies and expenditures on basic research in areas like nanotechnology which are likely to yield solutions to energy production problems in the future.

In the final analysis, modern technological society is all about reducing risks—that is why we're living longer and healthier lives. Because of humanity's advancing technological, institutional prowess, we are vastly better positioned to handle plagues, asteroids and climate change than our ancestors even 50 years ago were. Given the trajectory of human progress, it's very unlikely that a Posnerian catastrophe will ever wipe out humanity.