Lessons from empirical studies of the costs of extreme weather events

by Fran Sussman

A critical gap in our knowledge about the economic costs of climate change is the cost of extreme events. Until earlier this week, the calmness of this Fall’s hurricane season had lulled  the public and policy makers into a sense that perhaps we have more time. But the devastation and loss in life caused when Typhoon Haiyan—among the worst in history—slammed into the Philippines has quickly dispelled any developing sense of security about extreme events.  As we make choices concerning the types and magnitude of investments and institutional changes we need to make for the future, the question arises, “Is there anything to be learned from studies of the costs of extreme events about how to prepare—how much to spend on prevention and preparedness, or where to target investments?”

The story of the “broken window” is often used to describe a fallacy about the positive effects of disaster, suggesting that it can actually help the economy:

This is a reference to Frédéric Bastiat who, around 1850, wrote about a shop owner whose window was broken. Some onlookers convinced everyone that it was actually better for the economy because now the window-fixer would be employed and he would pay others, and so on, creating ripple effects in the economy. Our intuition suggests that the simple destruction of capital should not be a net benefit, and the error in the fallacy is the neglect of the fact that had the shop owner not needed to repair a window, he would have used the funds elsewhere—the broken window did not create new economic activity, but just diverted funds from one use to another. Similarly, owners of homes destroyed by tornadoes or hurricanes would have spent money elsewhere that they instead have to use for rebuilding. 

The fallacy is indeed a fallacy. But what is missing from this story is the fact that a broken window can be replaced with a better, more energy efficient window. Even if the shop owner would not have chosen to replace the current window with a better one for some years (if ever), and would certainly not have broken it to make the replacement necessary (let’s ignore the incentives provided by insurance), nonetheless he ends up with a better window than he had, and society has one more energy efficient window in place. There are still net costs to society from the whole event (resources diverted that could otherwise be used elsewhere), but those costs are very slightly offset by the long term cost savings from less energy consumption.

Similarly, once a community affected by an extreme weather event moves past emergency aid and restoring services, it begins to rebuild and reconstruct, making decisions all along about how to do so. Such decisions—how strong to make the bridge, what type of glass to install–will affect the long term economic path it is on. Moreover, suppose it was a shoemaker’s shop, and that it was not a window but rather equipment that was destroyed or the assistant who was injured. How badly the equipment is damaged, and how the shoemaker responds could well have economic consequences.

The immediate effect of an extreme weather event on the economy is to destroy or damage infrastructure and other physical capital, disrupt services, and, in the worst case, result in death and illness—all of which should have a negative effect. As the community rebuilds and reconstructs, we would expect to see faster growth in some areas and sectors, but also would expect the resources contributing to that growth to be diverted from other uses, which would grow more slowly or decline. In the long term, we would expect to see the community return to a stable growth path—perhaps the one it was on before, or a higher or lower one. The question is how to ensure that the economy recovers quickly and returns to a strong growth path.

A recent thoughtful article, “Informing Climate Adaptation: A Review of the Economic Costs of Natural Disasters,“ by Carolyn Kousky of Resources for the Future,” undertakes the rather daunting task of assessing the empirical literature on the costs of natural disasters.

In the article (from which the quoted paragraph about the broken window is taken), Kousky reviews the existing studies—which are not that numerous—of the economic impacts of disasters. The range of results in these studies makes it perplexing to find a convincing storyline about the relationship between extreme events and the economy. In looking at short term (one to five year) impacts, some studies find immediate negative impacts on GDP that are eliminated by subsequent growth, while another finds negative impacts on GDP that persist, and another finds no impact. By contrast, several studies of longer term impacts find persistent and negative effects, at least for some types of impacts, such as those affecting a high proportion of the population, or permanently affecting agricultural yield.

Not surprisingly, the negative effects on GDP appear to be higher in response to more intense events or more frequent storms, and in developing countries and those with larger informal sectors that have less access to insurance and reconstruction aid.  But it also may be that the long term economic impacts of an extreme event will be highest when a storm irreparably damages productive capabilities—whether natural, human, or physical capital–or when repairing those capabilities is expensive or will take a long time.

Recognizing the pathways by which the economy is affected by extreme weather events (and how those effects persist over time) may help in devising policies that make the economies of countries, cities, and communities more elastic—more able to bounce back—after a weather-related disaster.  Reducing long-term economic impacts could be fostered by policies that:

  • Focus climate-related investment on protecting  productive capacity and inputs into production
  • Incorporate economic diversification into development strategies, to reduce the dependence of local economies on climate-vulnerable resources or processes
  • Use post-disaster recovery efforts to implement  long-term plans that reduce the vulnerability of economic systems and resources to extreme events, by moving resources out of harm’s way
  • Take advantage of the “broken window” to rebuild infrastructure and institutions that are less-climate-sensitive, as well as more productive for the economy
  • Recognize the importance of “tipping points” in economic and other human systems—the point at which migration begins, economic production collapses, or services break down

While Kousky’s article is written with an eye towards assembling the requisite understanding of the frameworks and underlying data from these studies, so that the natural next step—estimating the economic impacts of future climate-related natural disasters—can eventually be taken, these studies can tell us much more than simply the cost of disasters.  Understanding the underlying factors driving the impacts on economies may also—ultimately—help us to understand how impacts can be lessened over the long term by the actions we take to prepare, and the choices we make when we rebuild.

Leave a comment

So, what did you think?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: