Wreckonomics: Hurricane Sandy and Real Estate Risk Management

For all its human costs, the economics of a hurricane are uncomplicated. During the event and in its immediate aftermath, the stock of wealth is reduced and productive activity comes to a standstill. As activity resumes, resources are quickly deployed to return wealth to its initial level. Recovery spending rises in proportion to the destructiveness of the initial event. The local purse is rarely adequate to the task. An external agent socializes the costs.

In duration and magnitude, the stimulus exceeds the relatively briefer interruption of activity. The higher flow is captured in the GDP accounting, but the initial decline in the stock is not, giving rise to the glazier’s fallacy. The economy registers a bounce in the medium term, masking the fact that wealth is reduced. If the outcome of natural disasters were a net increase in wealth, we would engineer them ourselves. We do not, because creative destruction is the exception and not the rule.

Among last week’s most surprising turns, Mayor Bloomberg coming out as a single-issue voter adds a powerful voice to calls for action on climate change. The city’s first citizen made a point to acknowledge Sandy’s ambiguous relationship to the larger issue. For many Americans, the veracity of climate change is in doubt. The mayor’s point still stands: the stakes are high enough that the possibility of a linkage demands a concerted policy response.

Even if this ostensibly once-in-a-generation storm does not evince a causal relationship with climate change, the real question for New York is whether the frequency of catastrophic weather events is rising. Through the lens of risk management, the tails of the distribution may be getting fatter. Two data points are not a trend, but the back-to-back threats of Irene and Sandy raise suspicions. Our perspective bears on every aspect of decision-making in the built environment. First and foremost, how do we prepare for next time? For lenders, how should event risk influence our measure of credit risk, if at all?

Assume for a moment the tails are getting fatter. In an efficient market, property owners’ insurance premiums should rise in proportion to their risk-taking behaviors. In the crudest example, we should see higher costs for insuring properties in flood-prone areas. An imperfect proxy for flood risk, it then becomes more expensive to insure residential and commercial assets in evacuation Zone A. Holding the aesthetic appeal of waterfront property unchanged, asset values decline in step with their higher costs and lower expected productivity.

Investors who believe that there will be no long-term impact on value assume that Bayesian updates (my term of art, not theirs) will be more favorable to low-lying areas as time passes; they are betting there is no pattern and that the skies will remain clear in 2013. If a hurricane threatens the city again next year, that position will become tenuous. For the sufficiently risk-averse, three times is a trend if two is not.

From an analysis of historical behaviors, we can infer that investors believe the cost of flood protection will be borne across all property owners, or that government will offset the cost of serious events. Individual actors and investors are also myopic. When hyperbolic discounting of a presumably rare event leaves them underwater a second or third time, even detractors of big government will seek out disaster assistance. Premiums will not reflect risk-taking and the value of risky assets will be propped up by moral hazard.
dsc@chandan.com

Sam Chandan, Ph.D., is president and chief economist of Chandan Economics and an adjunct professor at the Wharton School.