Tuesday, March 15, 2011

Broken Window Fallacy?

Recent events in Japan have prompted discussions of the broken window fallacy. The typical chain of events goes something like this:
1. Natural disaster
2. Reporter (or, reporter citing an economist) notes GDP might increase because of disaster.
3. Economist (probably not the one cited by the reporter) yells "BROKEN WINDOW FALLACY!" And, presumably, scores points for being reasonable.
Today, I will venture into the land of heresy by suggesting those who point out that natural disasters might make people better off are not necessarily committing a fallacy.

A fallacy is a chain of reasoning whereby the premises do not support the conclusion. But consider the following chain of reasoning.
1. Individuals may have a bias whereby they do not upgrade appliances when doing so would make them better off by their own assessment.
2. Natural disasters force them to buy new appliances.
3. The net effect of natural disasters is ambiguous (i.e., it might be positive, negative, or zero)
Where's the fallacy? Note: I am not saying that natural disasters make us better off. I am saying they might make us better off if the behavioral problem identified is significant. I think it is unlikely that the benefits from upgrading our refrigerators would be so great that they would offset the buildings destroyed by the natural disaster. (Of course, if the underlying behavioral problem is significant, it might apply to more than just household appliances.) But this is an empirical question.

My suggestion: stop calling such claims a logical fallacy. Point out that an increase in measured GDP does not necessarily imply an increase in welfare and then move on. There is no need to paint those we disagree with as unreasonable, illogical, or stupid.


Eli said...

Will, your last paragraph is exactly correct. I happily join you in your heresy.

Unknown said...

Most reasonable thing I've read lately.

Greg Finley said...

But suggesting that I weigh less than a feather is also an empirical question, but it doesn't make me any less of an idiot to say it.

Will Luther said...

Some people are stupid--or say stupid things. There is no denying that. But assuming people are stupid from the outset when they say something that *appears* to be stupid seems to me a poor way for an academic to approach the problem. You are unlikely to learn something new when you assume the person you might otherwise learn from is stupid. And, if it turns out that your assumption is justified, you are unlikely to persuade them of their stupidity.

KipEsquire said...

Even Bastiat conceded that broken windows are great for glaziers.

The whole point of the BWF is that individual circumstances (i.e.., exceptions and outliers) should not be substituted for aggregations (also, secondarily, that growth for growth's sake, without any frame of reference, is not a wise objective function to maximize).

jeremy h. said...

Will, I don't think most of those economists finding silver-linings in disasters are making behavioral claims about the optimal replacement rate of capital. Typically, I believe, they are making an aggregate demand argument. It seems reasonable to call this a fallacy, especially if they do not stress that even in the best case the benefits are short-run with a negative long-run effect.

Will Luther said...

Kip: My claim is that broken windows might be best for those who get their windows broken by requiring that they replace windows sooner than they otherwise would, if they are currently employ suboptimal replacement rates.

Jeremy: I used the behavioral example because I thought it would be an easier case to make considering my (macro hating) audience. You are correct in that most top-tier academics making the claim are assuming insufficient aggregate demand. But one could make a logical case with that assumption as well.

1. The economy is below full employment.
2. The disaster brings idle resources back into use.
3. The net effect of the disaster (increase in welfare from employment of idle resources minus wealth destroyed by disaster) might be positive, negative, or zero.

We might agree that the argument is not valid description of the present state, but it is certainly sound--and therefore should not be called a logical fallacy.

jeremy h. said...

Will: I would argue that point #3 is the heart of the BWF, the claim that the net long-run effect can be positive.

Will Luther said...

Jeremy: I agree. And certainly the argument Bastiat was addressing was a fallacy, because it went something like this.

1. Natural disaster increases GDP.
2. Natural disasters make us wealthier.

This obviously doesn't follow. It is a fallacy. What I am trying to point out is that someone might construct a logically sound version that reaches a similar (though not as strong) conclusion.

Justin Ross said...

I'll back up Will on this, in fact I wrote something similar a few years ago: http://perfectsubstitute.blogspot.com/2008/10/re-destruction-is-creation.html

Seamus said...

If I voluntarily choose to upgrade an appliance that still works, the value of the replaced appliance will be greater than zero. The appliance may even have some value in the resale market. If a natural disaster forces the upgrade through destruction the value of the replaced appliance is zero.

Will Luther said...

Seamus Coffey: Wealth is destroyed by natural disasters. No argument there.

This just means the change in welfare (the difference in utility between old appliance and new appliance) must be positive, and so large that it entirely offsets the reduction in utility associated with the wealth destroyed, for one to conclude we are better off having experienced the natural disaster. There is no fallacy here.

D3 Gold said...

amazing post... may help us a great deal,that maybe what I used to be trying to find! Cheers. Diablo 3 Gold

Guild Wars 2 Gold

Anonymous said...

Oh, seems a little horrible, hope you well

GW 2 Gold

swtor credits said...

This is a great post, i like it very much. I am very much interested in these types of topics from childhood and it's my habit to read this. you can visit and find diablo 3 gold, i like it very much.

guildwars said...

The open beta of Dungeons and Dragons-inspired MMORPG Neverwinter will finally come to its end on June 20, when the game will be officially launched! Cryptic Studios is running special End of Neverwinter Open Beta Events to commemorate the game's transition from beta to release build! Want to see what’s on, buy neverwinter gold for sale on safewow and join the events to improve your game chances in Neverwinter!

zoyi330 said...

I like read, this is a good android phone cheap article

Unknown said...

I love love love this article! Its very good quality and looks amazingly cute! It came very fast in the mail and its a very nice fit too! No cheap phones for sale complaints at all! I would recommend this to anyone looking for a good quality article!