HOME | NEWSROOM | ARTICLES

Best of a bad situation: choosing which disasters to prevent

In a world of finite resources, which worst-case scenarios should we prepare for? And which should we let go?

October 27, 2015

approaching storm

New economic research treats global disasters as interdependent events

For millennia, policymakers hoping to prevent the worst of catastrophes have had to go with their gut. Metrics for deciding whether it’s wisest to protect one’s citizens by building a levee, a sewer system, a wall, or an army have been rudimentary at best.

Until now. A new article in the American Economic Review from MIT Sloan economics professor Robert Pindyck and London School of Economics co-author Ian Martin shows how to decide which among seven contemporary catastrophes is worth preventing—and demonstrates in the process that major catastrophes as interdependent events and cannot be analyzed independently of each other. In fact, Pindyck writes, some catastrophes should not be prevented at all.

A new approach to prevention analysis

Mega-viruses. Climate change. Nuclear terrorism. Bioterrorism. Floods. Storms. Earthquakes.

Pindyck and Martin consider the benefits and costs of preventing each of these.

The research centers on a society’s willingness to pay in order to prevent a catastrophe—or at the very least mitigate its effect—compared to the cost of prevention, expressed as a permanent tax on consumption. Most previous examinations of catastrophe prevention have used this type of cost-benefit analysis.

The approach works for considering a single event in isolation. But in a world of major potential catastrophes, prudent government policies must consider them in the aggregate, not independently. Once this is done, however, simple cost-benefit analysis no longer works.

Suppose that society faces five major potential catastrophes. If the benefit of averting each one exceeds the cost, straightforward cost-benefit analysis would say we should avert all five. The authors show, however, that it may be optimal to avert only, say, three of the five, and not necessarily the three with the highest benefit/cost ratios. This result might at first seem strange, but the authors show how it follows from economic principles.

The authors’ results highlight a fundamental flaw in the way economists usually approach potential catastrophes. Consider the possibility of a climate change catastrophe—a climate outcome so severe in terms of higher temperatures and rising sea levels that it would sharply reduce economic output and consumption, broadly understood. A number of studies have tried to evaluate greenhouse gas abatement policies by combining greenhouse gas abatement cost estimates with estimates of the expected benefits to society from avoiding or reducing the likelihood of a bad outcome. However, all such studies look at climate change in isolation, which, as the authors show, is misleading.

In addition to cost and benefit, deciding which catastrophes to prevent depends on a third factor: a society’s overall level of risk aversion. It turns out that as risk aversion increases, the number of potential catastrophes that society should prevent goes down.

Catastrophes don’t occur in isolation

As a numerical example, Martin and Pindyck consider seven potential catastrophes: nuclear terrorism, bioterrorism, a mega-virus, climate change, floods, storms, and earthquakes. They show that even if conventional cost-benefit analysis indicates that all seven should be prevented, in fact only four or five should be prevented. Which four or five depends on the society’s level of risk aversion, as well as other factors such as gross domestic product growth. For large projects such as averting a major catastrophe, conventional cost-benefit analysis breaks down. Projects cannot be considered in isolation. In this case, the seven different catastrophes are inextricably intertwined, and must be analyzed together.

“Can’t we look at these things one by one, the way we’re looking at climate change? Get a social cost of whatever—nuclear terrorism—and then figure out what to do?” Pindyck asked at an MIT Sloan conference in Santiago, Chile this year.

“What I’ve found is the answer is ‘No,’” he said. “That it turns out that social benefits of averting major catastrophic events are interdependent. And the decision as to what do you do when you have four, five, or six, or seven major catastrophes that society faces, it’s not simply ‘Go down and create the benefit-cost ratio and go sequentially.' It doesn’t work.”