William J. Polley

Yes, this is the GDP counts "bads as well as goods," and is hopefully drummed into every principles student's head. I know I make this point in most classes I teach even at the upper division level. It's an important point, I agree many seem not to fully understand that GDP is not a perfect (or some would say even very good) measure of well-being. I covered most of these reasons in a post not too long ago called "Measuring Human Well-Being," I think I listed most of the usual reasons.

But from the Fed's perspective, in an inflation targetng environment, the important factor is what happens to pressure on prices. If people, say, draw down wealth levels as they rebuild then this will add a positive shock to AD, or so it seems to me. But there are negative shocks as well, one is simply expectations which at this point are hard to gauge - how will this affect collective views of the future, particularly as oil prices go up? But looking forward it does seem lto me as though this has the potential to increase the demand for inputs in key sectors such as housing putting upward pressure on prices. It is true that we only have what we started with, but to get back there requires the use of additional resources.

"But looking forward it does seem lto me as though this has the potential to increase the demand for inputs in key sectors such as housing putting upward pressure on prices."

Yes. Bastiat's glazier sees an increase in demand. Someone else (unseen and unknown perhaps) experiences a decrease in demand. It is the potentially unseen decrease in demand that people don't always acknowledge.

"It is true that we only have what we started with, but to get back there requires the use of additional resources."

Precisely. All the more reason that natural disasters are NOT beneficial to the economy. If people draw down wealth levels to rebuild (and they weren't planning to draw down their wealth), that's not good. If you're using up resources to get back where you started, that's not good. Hence, I think it is extremely ill advised to put a positive economic spin on a natural disaster. There are opportunity costs. What would the money have been spent on if the disaster had not occurred?

And once you start arguing that a disaster is a net economic benefit, where does that argument end?

Actually, I don't really believe that the analysts in the article really believe that the hurricane represents a NET economic benefit (at least I hope not). But they are so quick to point out the (questionable) gain in GDP, and they seldom acknowledge the opportunity cost.

Then when the story is retold by someone who has even less training in economics, it often morphs into a net benefit story.

It seems to me that Bastiat’s argument is flawed, and anyone who has studied economics after 1936 should be aware of the flaws:

“It is not seen that, since our citizen has spent six francs for one thing, he will not be able to spend them for another. It is not seen that if he had not had a windowpane to replace, he would have replaced, for example, his worn-out shoes or added another book to his library. In brief, he would have put his six francs to some use or other for which he will not now have them.”

This seems to rely on the assumption that the citizen has no savings (or borrowing capacity) on which to draw to replace the broken window. If this assumption bore some relation to reality at the time, it certainly does not apply in the US today. Of course, you may want to argue that, if the citizen optimizes over an infinite horizon, he will choose not to replace his shoes, but this doesn’t square with my impression of how people actually behave, and certainly not with how governments behave.

Suppose the US had been hit with a major natural disaster in 1932 or 1938. Do you and Bastiat contend that it could not possibly have improved the condition of the economy? Consider the stone that Hitler threw through Poland’s window in 1939. Are you contending that this did not increase economic growth in the world?

The costs of World War II may have exceeded the benefits, but that is not an inherent property of such events (and some might argue, not even true of WWII, considering what the effects of an ongoing depression might have been). I think it was quite clear to Keynes that, given the multiplier effect, inherently worthless expenditures can have welfare-enhancing effects. (Does you Principles course end after the first semester?)

Valid questions, knzn, and I will do my best to answer them.

I don't believe that savings or borrowing capacity makes any difference to the point at issue. Let us cast the question in terms of lifetime wealth. If someone throws a stone through your window and you replace it you are using some of your lifetime wealth that you would have otherwise spent on something else. If nothing else, it throws off the timing of your decision to buy a new window (maybe it was brand new and had 30 years of life left). Prior to some miscreant breaking your window, you were quite happy with the decisions you had made to maximize your lifetime utility subject to lifetime wealth. Now, you must reoptimize. How is that benefical (except to the glazier, of course)? If you draw down your savings (or borrow) to replace the window, you lose the use of those resources for whatever you had planned to use them for previously.

As it happens, the U.S. was hit by multiple natural disasters in the 1930s. The Labor Day Hurricane of 1935 contains a cruel irony in that it killed many people who were working on New Deal projects in the Florida Keys, precisely the sort of stimulus you are alluding to. Hard to argue that was a good thing. The road did end up getting built (as it would have anyway), but there was tremendous cost in life and property.

Now, as for WWII, I was actually hoping someone might bring that up because it is, as you know, often cited as an example of Keynesian stimulus where economic growth was spurred by seemingly destructive spending activities like paying people to make bombs and then blowing stuff up.

Of course, if there are millions of people unemployed there is a certain stimulative value in paying them to make bombs, paint murals in post offices, write poems, or do other things along those lines. In the event that you are stuck in a bad equilibrium like the Great Depression, Keynesian style spending might be just what the doctor ordered. But I do contend there is a difference between Keynesian style stimulus aimed at constructing something new (like a bridge or a road that wasn't there before like we did in the 1930s) as opposed to rebuilding properly functioning capital stock.

Your final statment about WWII I think properly acknowledges that there are many dimensions along which WWII affected the economy and society, many negative, but certainly some positives came out of it. Such is always true of war, and of disaster more generally. WWII spurred research into technologies like radar and microwave. The Korean War and Vietnam changed the way we think about treating wounded people in the field and revolutionized emergency medicine. But this isn't really the point. Every major event or shock, economic or social, leads to ends which no one would have been able to predict. We simply cannot say ex ante (or even ex post) whether those benefits are worth the costs. That gets into a calculus that I simply don't want to do.

But I think your point is more along the lines of the renewal that comes along with the destruction (a sort of Schumpeterian notion). And I agree with that point as far as that goes. Recessions cause inefficient firms to go out of business and new ventures pop up in their place. A forest fire is sometimes necessary to create new growth. Such is the heartbeat of economies (and ecosystems). But none of this alters the fact that involuntary and unexpected destruction that is random and indiscriminant--taking out brand new capital along with the old and inefficient--is something that has significant wealth enhancing effects. Just think about carrying that argument to the next step.

As a final comment, Mark Thoma points out above, quite correctly, that prices will rise due to the additional demand. I think that is almost a certainty in the current situation in LA, AL, and MS. That would also tend to discourage me from thinking that there will be a lot of Keynesian style stimulus. We may not be precisely at full employment, but close enough that this will represent a significant disruption rather than putting to work a reserve army of the unemployed.

In the end, the effect is extremely redistributive, and by its nature, such extremely redistributive events can have growth consequences. But it's very suspect to claim an unambiguous aggregate gain--yet that's exactly what some will claim.

And to address your last question, when I talk about such things in class, I do mention the things I am saying in this comment and encourage discussion just like this. I've tried to be as thorough and precise as possible, and I thank you for taking time to write. I've seen your comments at other blogs and I hope you'll contribute here as well.

My post was more an abstract objection to Bastiat than anything else. I still hold to the point that disasters can be net welfare-enhancing (probably never in the sense of a Pareto improvement, but I’m willing to trade off James Goodfellow’s welfare for that of various others if the price is right). I don’t claim that outcome is typical, nor, indeed, am I sure that it has ever actually happened. But my general point was that resource constraints are not always present, and when they are not, even an unintentional stimulus with some obvious bad effects can still be more helpful than harmful.

Regarding Katrina in particular, I still have objections, though I doubt anyone would even try to make a credible argument that the effect was welfare-enhancing. Indeed, I would agree that the ultimate effects on growth, once we take energy prices into account, will almost certainly be negative. However, aside from energy prices, I think the effect on growth is an open question. There is room for a great deal of disagreement as to how close we are to a true resource constraint. The economists quoted, though they may turn out to be wrong, seem to me to be taking a reasonable position on this issue. The article as a whole is clearly not putting a positive spin on Katrina. (I guess, according to the right-hand rule, a cyclone always has a negative spin, but that’s another story.) While the article could have been a bit clearer about how all the parts fit together, I don’t think overall that it was misleading or in violation of economic principles.

Well at least we agree that disasters are not Pareto improving. I'm not quite with you on trading off James Goodfellow's welfare for others. But then I wasn't crazy about the Kelo decision either. Would I approve of it if it were a government destroying someone's law-abiding, well-functioning property in the hope that their rebuilding would stimulate the local economy? No. Of course with disasters, it's not a matter of choice--not something to approve of or disapprove of. We agree there is a cloud, and there's nothing we can do about it now. The question is how bright is the silver lining.

As a general point, redistribution of income and wealth (which any disaster will do) always has the potential for growth effects (positive and negative). When the redistribution is a policy decision, an argument could be made that those harmed now will be helped by future growth or compensated in some way. This is more problematic with wars and disasters (where, indeed, many will not live to see the future benefits).

We can disagree about whether we are up against the resource constraint. But the Great Depression and WWII are orders of magnitude larger both in terms of the departure from full employment and the size of the stimulus, so I would discourage that particular comparison applied to the present day. And even in that case, the nadir of business investment as a percenage of GNP was not in 1932, but during WWII (about 1943 by the chart I'm looking at).

But truly my biggest beef with the article (and the reason I chose to post about it) was that even if it would have addressed the things we have discussed (taking up more space than any newspaper would ever give it...let's hear it for the blogs!) it is the kind of article that lends itself to being misinterpreted by people who have not thought about it as deeply as you and I have. I don't think I said, and I didn't mean to imply that the article itself was in "violation" of economic principles. It did not do a very good job of fitting all the pieces together, however. I would demand more in a principles course.

When the story is told and retold, any notion of resource constraints falls away and it gets reduced to a lowest common denominator, Keynesian style stimulus story, which I don't think that you would be willing to accept at face value. The version I hear "on the street" is definitely worse than the version in the article, but I'm convinced that is because people have heard the Keynesian stimulus story so many times (including in newspaper articles much worse than this) that it becomes the focal point. For many people, media and otherwise, the Keynesian stimulus story is the beginning, middle, and end. The story on the street becomes one of "___(fill in your disaster)____ will grow the economy." Yes, I often hear the word "will." And yes, I did hear it about 9/11 and the Iraq war. When I teach principles, the Keynesian stimulus story is there (some have heard it before), but so is the other stuff (which most haven't heard before).

That's where I'm coming from. Again, glad to have you here commenting. Hope you'll stick around!

One more thing... on the redistributive aspect of the issue, which you raise (trading off one person's welfare for others):

That can be said about just about anything: crime, riots, etc. Yet, the Keynesian argument (at least as far as I have noticed) doesn't always surface in those contexts. Certainly such events are typically on a smaller scale and so there's not as much Keynesian style multipler impact on a national level. But how come we don't hear after a riot what a benefit this will be to the local economy with all the spending for repairing the damage? Or do we and I have missed it?

Any time spending is changed due to a shock, there will be winners and losers. There is a multiplier effect associated with the losers too. Random redistribution isn't something I like to see.