Hurricane Irene and the Economy

The economic costs of the last weekend's hurricane are still being tallied in the Carribean and along the eastern coast of the United States. That accounting takes place against the backdrop of the hurricane's human toll, including the loss of life and very real personal costs that will go unrecognized in the financial tally.

In the context of a fragile economic recovery, questions about the impact of the hurricane on near- and medium-term activity are prudent and necessary. If we sit precariously close to a relapse into recession, how will this setback influence the outcome, if at all?

As with any negative shock, the immediate impact of the hurricane will be a decline in economic activity and private capital flows in proportion to the severity of the event. In the short and medium term, however, activity in the affected states may rebound as capital and labor resources are quickly deployed in rebuilding.

In the aftermath of Hurricane Hugo, which struck the Caribbean and South Carolina at a cost of billions of dollars in September 1989, Paulo Guimaraes, Frank Hefner and Douglass Woodward described the reconstruction phenomenon as follows:

An irony of natural disasters is that although they destroy physical wealth, they often dramatically raise economic activity during reconstruction. Many urban and regional economies suffer immediate wealth losses from physical catastrophes. … However, some sectors of the economy experience temporary surges in income and employment that can be difficult to disentangle from other cyclical changes. ^[1]

This finding must be considered carefully. Any meaningfully positive impact from reconstruction spending must be disentangled from a broken-window fallacy, where activity increases simply to replace productive resources that have been lost.

The destruction and replacement of infrastructure following a disaster–whether that disaster is natural or a combination of natural and follow-on events–does not necessarily imply a process of creative destruction, even if capital goods are upgraded as a result. If it did, we would observe well-run governments tearing down those factories of their own accord.

In Professor Joseph Schumpeter's analysis, creative destruction describes how radical innovation can reduce or destroy the value of established technologies and business models. But a shock such as a natural disaster may deviate from the implicit assumptions of Dr Schumpeter's model, especially if valuable and highly productive assets are destroyed alongside the obsolete.

Debate on this question has not yet reached a conclusion. In fact, recent research suggests that climatic disasters, such as hurricanes, have beneficial long-run effects, while geologic events, such as earthquakes, have negative effects.

The research also shows that more developed economies and countries with larger landmasses tend to experience fewer human losses and recover more quickly from disasters. Decomposing individual disaster recoveries and controlling for variation in country development, increases in economic activity are neither even nor unambiguous, but tend to favor certain sectors of the economy.

The broad finding that more developed countries recover more quickly and more completely may result from their higher ex ante investments in disaster mitigation. In the case of the March 11 earthquake, for example, the infrastructure that alerted Japan's coastal residents to the imminent tsunami has been credited with saving an untold number of lives, even if seawalls failed under the enormous height of the waves.

All things being equal, the rebound in the United States will be easier than Japan's, in large part because the current dislocation has been on the margins. One finding of the research that seems to hold across otherwise contradictory analyses shows that scale matters for the economy’s capacity to rebound: "… If a natural disaster is large enough to dislocate economic activity, all the mechanisms that could potentially make it positive for growth would be weakened. Consequently, the data ought to show that severe disasters of any type lower economic growth."^[2] Another reason we can be thankful that Irene proved somewhat meeker in its bite than its bark.

1. Paulo Guimaraes, Frank Hefner and Douglas Woodward, "Wealth and Income Effects of Natural Disasters: An Econometric Analysis of Hurricane Hugo," The Review of Regional Studies, 1993.

2. Norman Loayza, Eduardo Olaberria, Jamele Rigolini and Luc Christiansen, "Natural Disasters and Medium-Term Economic Growth: The Contrasting Effects of Different Events on Disaggregated Output," Working Paper, 2009.