Blaming Nostradamus
by Kara Lilly, Mawer Investment Management
In the past week, the scientific community has been up in arms about a ruling in Italy that will send six scientists to jail for their role in the 2009 L’Aquila earthquake. One of the deadliest earthquakes to hit Italy in recent decades, L’Aquila caused the deaths of roughly 300 people and was felt throughout central Italy. Now, the sentencing of six scientists and one ex-government official to six years in prison and a $1 million fine each for the crime of manslaughter has outraged many members of the international scientific community. They claim the punishment is hardly representative of the crime and indicates a deep misunderstanding of even the most fundamental tenets of science.
It’s a noteworthy debate. Not just because it impacts the lives of the six scientists, but because it highlights an important issue that rarely receives attention: the manner in which we as a society deal with randomness. It highlights an obsession in our society when it comes to making predictions and a failure to grasp even the basics rules around probability.
So what happened? Most major news sources cite the following tale: a team of six scientists were sent to L’Aquila in the weeks before the big one struck. The scientists had been dispatched to the city to assess the level of danger given the hundreds of small shocks in the area that had taken place. They judged the risk as minor. Speaking to a press conference, government official De Barnardinis infamously told reporters that "the scientific community tells us there is no danger, because there is an ongoing discharge of energy. The situation looks favorable"; a statement that was both foolish and markedly more decisive in its conclusion than the consensus that is implied from the committee’s notes. Six days later, a magnitude 6.3 earthquake struck L’Aquila, killing an estimated 308 people, leaving 1,600 injured and more than 65,000 homeless, and three years later, six scientists have been found guilty of manslaughter.
It might feel like justice has been served, but many find the conviction unsettling. Convicting individual scientists for failing to predict an unpredictable event is highly problematic for a number of reasons.
To begin, the scientists are being blamed for being wrong in a prediction they never truly made. With the exception of the rather careless statement made at the press conference (where the majority of the scientific team was absent), the team of scientists appear to have been providing their view on the probabilities of a large earthquake occurring. This is different from saying what will outright happen. The problem is that the Italian courts cannot seem to differentiate between the two, and are holding the scientists accountable for a prediction that they believe was made and was wrong. This is absurd. It is reasonable to estimate the odds of an earthquake happening but totally unreasonable to try to predict an earthquake. Like prices in the stock market, there is no earthly reason to suspect that seismologists possess the ability to systematically predict earthquakes with any reliable degree of accuracy.
Of course, the Italian courts are claiming that they are not jailing the scientists for a faulty prediction. They claim that communication was handled poorly and that better communication would have allowed people to flee their homes. This may be partly true but it does not seem to be the whole story. One need only examine the logic of the court to see that underpinning their argument is anger at what the public views was a failure to predict. “Why weren’t the people warned that an earthquake was coming?!” the public appears to demand. Yet to ask this question is to demand a prediction, which is an approach we know to be misguided. It seems to us that much of public ire against the scientists is deeply misplaced. Since when have scientists been responsible for public policy? The current Hurricane Sandy storm, in comparison, turned out to be much worse than predicted but public officials appear to have made prudent emergency plans just in case.
And even if the scientists in L’Aquila had made a prediction, there is still considerable hypocrisy in any system that hires citizens to make predictions, knows that predictions will inevitably be wrong, and then punishes them outlandishly for their failures anyway. Rather than hiring people to make predictions, could we not do a better job building up systems that are more resilient to extreme events? This we believe would be a more fruitful use of taxpayer funds. We suspect, for example, that many lives might have been saved had the buildings in L’Aquila been monitored for their capacity to sustain a major shock and stricter building codes enforced. Something the committee of future jailed scientists had recommended.
The lesson here is twofold. First, there is a disparagingly low level of understanding when it comes to probability in our society. Is it the scientists fault for making a bad prediction, or is it our fault for poorly understanding their answers and for expecting predictions of them in the first place? Second, we need to stop playing Nostradamus and start trying to build resiliency against extreme events instead. This is true not just within the world of earthquakes, but in every aspect of society that is potentially fragile (banking and investment management come to mind). We suspect this is likely more effective than replacing forecasters every time they are wrong. Not only do we believe this approach is the right one to take in managing investment portfolios, but it strikes us as the most prudent way for nations to manage the risks they face.
Kara Lilly
Copyright © Mawer Investment Management
