Natural Disasters Don’t Increase Economic Growth
Hurricane season is nearly upon us, and every time a hurricane strikes, television and radio commentators and would-be economists are quick to proclaim the growth-boosting consequences of the vicissitudes of nature. Of course, if this were true, why wait for the next calamity? Let’s create one by bulldozing New York City and marvel at the growth-boosting activity engendered. Destroying homes, buildings, and capital equipment will undoubtedly help parts of the construction industry and possibly regional economies, but it is a mistake to conclude it will boost overall growth.
Every year, this popular misconception is trotted out although Frédéric Bastiat in 1848 clearly put it to rest with his parable of the broken window. Suppose we break a window. We will call up the window repairman, and pay him $100 for the repair. People watching will say this is a good thing. What would happen to the repairman if no windows were broken? Also, the $100 will allow the repairman to buy other goods and services creating income for others. This is “what is seen.”
If instead, the window had not been broken, the $100 may have purchased a new pair of shoes. The shoemaker would have made a sale and spent the money differently. This is “what is not seen.”
Society (in this case these three members) is better off if the window had not been broken, since we are left with an intact window and a pair of shoes, instead of just a window. Destruction does not lead to more goods and services or growth. This is what should be foreseen.
One of the first attempts to quantify the economic impact of a catastrophe was a 1969 book, The Economics of Natural Disasters. The authors, Howard Kunreuther and Douglas Dacy, largely did a case study on the Alaskan earthquake of 1964, the most powerful ever recorded in North America. They, unsurprisingly, concluded that Alaskans were better off after the quake, since money flooded in from private sources and generous grants and loans from the government. Again, this was “what is seen.”
While construction companies benefit from the rebuilding after a disaster, we must always ask, where does the money come from? If the funds come from FEMA or the National Flood Insurance Program (NFIP), the government had to tax, borrow, or print the money. Taxpayers are left with less money to spend elsewhere.
The economics of disasters remains a small field of study. There have been a limited number of empirical studies examining the link between growth and natural disasters. They can be divided into studies examining the short-term and long-term impact of disasters. The short-term studies, in general, found a negative relationship between disasters and growth while a lesser number of long-run studies have had mixed results.
The most cited long-eun study is “Do Natural disasters Promote Long-run Growth?” by Mark Skidmore and Hideki Toya who examined the frequency of disasters in 89 countries against their economic growth rates over a 30-year period. They tried to control for a variety of factors that might skew the findings, including country size, size of government, distance from the equator and openness to trade. They found a positive relationship between climate disasters (e.g., hurricanes and cyclones), and growth. The authors explain this finding by invoking what might be called Mother Nature’s contribution to what economist Joseph Schumpeter famously called capitalism’s "creative destruction.” By destroying old factories and roads, airports, and bridges, disasters allow new and more efficient infrastructure to be rebuilt, forcing the transition to a sleeker, more productive economy. Disasters perform the economic service of clearing out outdated infrastructure to make way for more efficient replacements.
There are three major problems with these empirical studies. The first is counterfactual. We cannot measure what growth would have been had the disaster never occurred. The second is association versus causation. We cannot say whether the disaster caused the growth or was simply associated with it.
The third problem is what economists call “ceteris paribus.” It is impossible to hold other factors constant and measure the exclusive impact of a disaster on growth. There are no laboratories to test macroeconomics concepts. This is the same limitation to Rogoff’s and Reinhart’s work on debt and growth, and many other bilateral relationships in economics. Using historical data from the early 1900s, researchers found that as the price of wheat increased, the consumption of wheat also increased. They triumphantly proclaimed that the demand curve was upward sloping. Of course, this relationship is not a demand curve, but the intersection points between supply and demand. The “holding everything else constant” assumption had been violated. In economics, empirical data can support a theoretical argument, but it cannot prove or disprove it.
So what do we do if the empirical studies have serious limitations? We go back to theory. We know a demand curve is downward sloping because of substitution and income effects. Wal-Mart does not run a clearance to sell less output! Theory also holds that natural disasters reduce growth (i.e., the more capital destroyed, the greater the negative impact on growth).
More capital means more growth. Robinson Crusoe will catch more fish if he sacrifices time fishing with his hands to build a net. Now, suppose a hurricane hits the island and destroys all of his nets. Robinson could go back to fishing with his bare hands and his output would have been permanently reduced. He could suffer an even greater decline in output by taking time to make new nets. The Skidmore-Toya explanation is to have him apply new methods and technologies to build even better nets, allowing him to catch more fish than before the hurricane. Of course, we may ask, if he had this knowledge, why didn’t Robinson build those better nets before the hurricane? This is where the Skidmore-Toya logic falls apart. Robinson did not build better nets before the hurricane because it was not optimal for him to do so.
If a company decides to replace an old machine with a new one, among the primary considerations are the initial price of the new machine, the applicable interest rate, and the reduced yearly costs of operation of the new machine. Using net present value analysis, the company determines the optimum time to make the switch (a real option). A hurricane forces a switch to occur earlier than would have been optimal under a price and profit motive. The hurricane therefore created a different path for growth. The creative destruction would have occurred, but on a different, more optimal, timeline.
The same conclusions can also be drawn from manmade disasters. Contrary to what many Keynesian economists would have you believe, WWII did not grow the US out of the great depression. Capitalism did!