« First Economics Department Blog? | Main | Protect this house »

August 30, 2005



I disagree. (See also my comments on William Polley’s post to which you link above.) If net spending is increased, and if there are slack resources (which, in a sense, there must be, in order for spending to increase in real terms), then somebody is almost certainly getting a net welfare benefit (specifically, the category of people who were unemployed before the disaster, were subsequently employed, and on whom the direct effect of the disaster happened to be relatively small). The article is only suggesting that the cloud may have a silver lining, even if it is a very large cloud with a very small silver lining.

It’s true that if you ignore distributional effects (and if the multiplier effect is small or the economy is near a resource constraint), then the growth that often follows a catastrophe is not true growth in a welfare sense. But surely you would not want to ignore distributional effects. The increase in welfare is quite large for a newly employed person. The decrease in welfare may be fairly small for those who must forego the results of the “more productive activities” you cite. Of course it would be wrong to suggest that hurricanes are a good thing, but I would argue, it is also wrong to deny that they can have some good effects.

John Whitehead

The good effects that you mention are almost certainly overwhelmed by the bad effects. The checks for the reconstruction come from the insurance companies. These firms lose money and premiums for everyone else may go up. The construction crews that will head down to New Orleans are not necessarily slack resources. They may be diverted from building homes and commercial buildings in other areas.

GDP may go up but this is only a reflection of the sometimes lousy way that GDP measures welfare. The Exxon Valdez oil spill also increased regional and national GDP.

I like the quote at Division of Labour: Destruction is bad.


There’s no question that the bad effects of the storm far outweigh the good effects. But that doesn’t mean we should be admonished to ignore the good effects.

GDP is not a measure of welfare (NNP, maybe, but that’s beside the point). GDP growth is related to welfare improvement in two specific ways. GDP growth consists essentially of employment growth and productivity growth. Employment growth is associated with welfare improvement because people who can get jobs are usually happier than people who can’t. Productivity growth is associated with welfare improvement because it expands our long-run set of consumption possibilities.

In the case of a disaster, the main issue with respect to GDP growth is employment. If Exxon Valdez increased employment, then it had some good effects, too (again, outweighed by the bad effects). It may also have been the occasion for technological advances which improved productivity.

Regarding Katrina, the question of whether there are slack resources to be employed in reconstruction is an open one. But by most accounts, we’re building too much in the rest of the country now, anyhow. Insurance premiums, I would suggest, are a minor issue compared to employment.

I do agree that destruction is bad, and that’s why I will vote against having a hurricane if it’s ever on the ballot.

John Whitehead

GDP is the most commonly used measure of aggregate welfare. How can you say that it isn't a measure of welfare?

This isn't an academic debate among people who sortof know what's what. It is a debate about how the media misinforms the public about the economic effects of a disaster.

I could agree with your excellent points but the media makes no effort to balance their statements that disasters lead to increased GDP and therefore disasters are good for the economy. Today, my students were parotting this logic. This is bad thinking but everyone thinks it is quite clever to point out that GDP will improve.

Plenty of economists have been willing to provide quotes in these "disasters increase GDP" articles. Outrageous.

Find me a news article that says that any increase in GDP following a disaster illustrates why GDP is not good measure of welfare and therefore the increased GDP should not be a focus of the discussion of the economic effects of disaster ... and I'll buy you a beer.

William Polley

John, you are exactly right. The point is not whether growth of GDP will follow the disaster. Of course it will! Faster than trend growth rates will be required to get the economy back to potential GDP. That's not the issue. Take a Solow growth model, wipe out half the capital stock, and the ensuing growth rate will be higher than it would have been. But the present value of lifetime income (and utility) is lower.

I say again, barring any Great Depression style outcomes, any time GDP falls below trend, there will follow a recovery period with faster than average growth. The same is true of any garden variety recession.

So how far would a president get with such platitudes during a recession? Would you go on camera saying, "growth will actually increase due to the recession"? Of course not. Yet somehow the media feels it's appropriate to say it after a natural disaster. I just don't get that.

And your reason for disgust over this, John, is exactly the same as mine. Students, as well as the folks out on the street, do parrot this logic, without the qualifications we all know are important.

At face value, such pronouncements just say that it's not all bad, there is some good. The same could be said of crime, riots, forest fires, etc. It's true, but at best it's not a very useful statement. At worst, it can lead to big misunderstandings if you're not careful. All you and I want people to do is be careful how they talk about it and avoid saying things that aren't useful.

Interesting blog, by the way.

The comments to this entry are closed.