The economics of recovery from natural disasters

Will Wilkinson has by far the best survey and treatment of this question.   Here is one excerpt:

By far the boldest claim, advanced in this 2002 paper by Mark Skidmore of the University of Wisconsin-Whitewater and Hideki Toya of Nagoya City University in Japan, is that some disasters can boost GDP by forcing upgrades in technology and infrastructure, and offering the opportunity for critical reappraisal of ingrained modes of economic activity, leading to a higher level of productivity and, eventually, to net gains in growth. They find that this holds for some weather-related disasters, but not for geological disasters. They find persistent, long-run negative effects for geological catastrophe, suggesting any upside from Japan’s earthquake and tsunami is unlikely. The argument of this paper, which is as strong as the disaster-bonus case gets, is a touchstone for a good deal of later research.


Makes me wonder what's so different between geological and weather disasters? Just magnitude?


Japan 2011: 9.0 earthquake -> 10,000 dead
Haiti 2010 : 7.0 earthquake -> 300,000 dead

Amazing what technology and preparedness can do to the survival probability!

PS. Japan has a population density [ ppl / sq km ] of 340 and Haiti 370. So not that different.
Also, the scale is a log scale. So the Japan earthquake was orders of magnitude more destructive than Haiti.

Orders of magnitude more, but location plays an important role too.
If that 9.0 hit directly under Tokyo or on the west side of Tokyo Bay, it would be a very different story there.

Yes, if you had an earthquake 100 times more powerful in an area with four times the population of the entire country of Haiti, there would have been a lot more deaths. So what?

Well, this article has a great point here. If Japan will focus on developing earthquake resistant buildings and tsunami defense things, then it might have a big possibility that the economy will rise.

Tyler, please you don't need to write that WW's post is the best survey because WW recognizes that his post is based on the best survey written for IADB. As a matter of fact, yesterday Ilan Noy --one of the authors of the IADB survey-- wrote a very good post for Econbrowser. If your readers want a good list of references they can check Noy's post in Econbrowser.
More important, most of the points made by Noy and all the other economists that have been studying the economics of disasters the past few years were made long ago, in the 1940s, by the great economist Ken Boulding in a series of papers and books about the economic consequences of war and peace during and after WWII. You can find in Boulding --unfortunately now I don't have access to his work and I couldn't find good references in the web (readers can check the entry on him in Wikipedia)-- the reasons why he expected a substantial increase in output after WWII (maybe he was the only Keynesian that was not worried at all about a recession after the war). I'm sure the late Boulding would be predicting a strong recovery of the Japanese economy in the next 10 years, one so strong that indeed would later lead to a slowdown in growth (as it happened to the world economy after 1970, when the recovery from WWII was complete).

Broken window fallacy

The Keynsian theory is that the Japanese just have not tried hard enough to stimulate the economy. An alternate theory is that Keynsianism just is not working.

Japan has had zero inflation over the past decade. If the problem was as diagnosed by Keynesians, one would expect to see the frequently announced, yet seldom observed, deflation.

So is the current disaster a geological one or (for Japan yet again) an atomic destruction one?

One killed 10,000 or so people, the other killed none. This is a really tough call

Among its many flaws, the most glaring inadequacy of GDP as a measure of economic health is that it does not take into account changes in a nation's wealth. A firm's net income (its true net income - not the "operating income" b.s.) is basically it's increase in net worth. But GDP only counts production. Thus, the true extent of the Great Recession is understated by the drop in GDP because it does not reflect the loss in household wealth that occurred.

Similarly with natural disasters, while the economic activity to recover from the disaster may well result in an increase in GDP over what it would have been otherwise, the country is still worse off due to the destruction of wealth the disaster caused.


Nobody is saying this is good for Japan. All we are saying is that - due to the nature of GDP - it will probably boost their income, at least in the short term. As ziel says, the problem lies with GDP, not with any sort of economics.

Is the claim really that bold?

Rebuilding in and out of itself could be enough to boost construction spending.

If my house burns down, you can sure bet than I (or my insurance company) will be spending more money and drawing down savings in the coming months. Given this disaster affects a relatively well-off group of people who likely have, on average, access to substantial capital, I'd expect spending due to rebuilding to go up, at least in the short run. At least that's my layman's analysis of this.

how does broken window fallacy apply at all? it's GDP. if national income was $100 before the quake, when people are spending on things they want to buy, and national income was $110 after the quake, when people are spending on reparations, then woohoo! whatever they wanted to spend on if the quake had not happened weren't the best things they should be buying.

Asides from growth an interesting question is the re-distributive effect of these large disasters. Does inequality increase or decrease. Is the aftermath of a disaster akin to progressive or regressive taxation? Any work on this?

"how does broken window fallacy apply at all? it’s GDP."

I think that's how it applies.

The broken window FALLACY is made when someone argues that (1) natural disasters increase GDP and (2) increases in GDP increase welfare so it follows that (3) natural disasters increase welfare. It is a fallacy because (1) is incomplete; while it is true that natural disasters increase GDP, they also decrease wealth. So the result in terms of welfare depends on the net effect of the natural disaster (i.e., to what extent the increase in income offsets the decrease in wealth).

As should be clear, one might argue that natural disasters could increase welfare without committing a fallacy. They could argue, for example, that the economy is below full employment. Or, as I believe is the case with Skidmore, they could argue that individuals do not upgrade technology at an optimal rate (perhaps because of a behavioral bias or policy barriers). Either way, it modifies the conclusion to (3') Natural disasters increase, decrease, or have no effect on welfare. We may debate whether these arguments are valid, but they are logically sound--so there is no reason to accuse Skidmore and others of committing a fallacy.

My sympathies go out for all who lost everything in this disaster.

message, this time about the threat of nuclear brinksmanship and the dangeres of atomic energy in all its forms. Drawing on public insecurity…the new Godzilla was intended as a cinematic wake-up call. “We wanted to show how easily a [nuclear] accident could occur today,” Tanaka remarked, “but vivid images of nuclear war are taboo…Gojira (1984) is not particularly subtle in its sermonizing, depicting the monster gutting a Japanese nuclear powe

Comments for this post are closed