Wednesday, August 24, 2011

One more quick thought

Why is Broken Window Fallacy thing such a big issue? It's not as if anyone has called for wanton destruction to stimulate the economy, after all. Even destruction that a lot of people genuinely support on its own merits - like killing Nazis - isn't something we go around looking for for macroeconomic purposes. So what's the argument over?

Crowding out.

And that is important even if these examples are mostly abstract.

Unfortunately, a lot of people would rather argue that "according to standard Keynesian theory, if all vestiges of civilization were destroyed, this would be a good thing" or "Krugman is a big fan of war" or "Krugman said Japan was going to see good economic times" or that "In Krugman’s version of Orwell’s Newspeak, destruction creates wealth, and war, though not ideal, is morally acceptable because it produces economic growth." And considerably nastier and dumber stuff than that.

How can one respond to this? First you point out the obvious - you don't think those things. But you can't entirely agree with these people on the impact of destruction either because you don't think crowding out holds right now. So to be fair and honest and forthright you've gotta make that still-important, controversial point.

And then when you do that the cycle of stupidity keeps turning. Some Austro-libertarian types do get it. But unfortunately these guys seem to be in the minority and aren't making a dent in the people that say we don't understand opportunity cost or that we actually want destruction.


  1. I hope you follow up to my reply in the "do get it" link. Bob Murphy wasn't correcting me on the major point of my post: that yes, you're right, a broken window can make a statistic go up, a statistic people equate with growth, but which no longer represents "growth" in any useful sense here. (Several others are making the same point, and you're ignoring all of them as well.)

    I would classify that as you "not getting" that any reasonable definition of growth would *ignore* "breaking stuff to give people work".

    An economist with genuine understanding would say something like this:

    "Yes, in the wake of an earthquake, you will see GDP numbers go up, but relative to the pre-quate situation this does not actually represent growth in the sense most of us care about -- it's just going to replace lost wealth. Certainly it's better to rebuild, *given* the earthquake, but the fact that we have to marshal resources this way is not some kind of blessing, nor would it have been an improvement for the earthquake to have been worse."

    Now, have you ever heard any Keynesian economist say anything like this? Have _you_ ever said anything like this, or even mentally instantiated that chain of reasoning?

  2. Silas I don't know how many times I have to say this is why we don't advocate destruction as a solution. If you're getting activity by doing destructive activity what's the point? GDP growth is a decent measure of growth because usually there's not this sort of self-inflicted destruction going on. But look, shit happens. When my car breaks that's destructive - but I still get value from improving my circumstances when I get it fixed. Dealing with the vagaries of life is valuable economic activity. What would you do - subtract off investment going towards capital depreciation or obsolescence from GDP growth? Should we subtract off the entire health care sector? After all, that's value creation in response to bad things that happen to us. I'm not sure what's so hard about all this.

  3. Silas I don't know how many times I have to say this is why we don't advocate destruction as a solution.

    What is this a response to? Because I don't see how someone who just read my post would just this as responsive.

    GDP growth is a decent measure of growth because usually there's not this sort of self-inflicted destruction going on.

    Right, Daniel_Kuehn, and economists who have that "genuine understanding" I alluded to are capable of identifying when a measure stops measuring what we think it's measuring, and therefore when the normal inferences based on that measure no longer apply.

    Generally when an economy has higher GDP than before, it is better off.

    But, if the GDP gain was simply due to fixing a natural disaster, it is _not_ necessarily better off than before the disaster.

    If you can't follow that (or, say, you change the topic to whether one should deal with the vagaries of life), and you're representative of Keynesians, then you richly deserve the ridicule sent your way on this topic.

    Yes, it's good to deal with the vagaries of life, but people with that "genuine understanding" can see that (except for "learning experiences" and whatnot) it is better *not to have* these hits to your utility at all, then to get the hits and then deal with them.

    If you believe that an earthquake+rebuilding effort is better (on any metric you purport to be useful) than an earthquake never happening at all, then you are guilty of precisely everything the Bastiat invokers accuse you of, even if you can justify your position (which would, in the process, refute Bastiat's argument, much as you think it wouldn't).

  4. It's a response to this: "I would classify that as you "not getting" that any reasonable definition of growth would *ignore* "breaking stuff to give people work""

    I do get it. This is why breaking things is not advocated as a growth strategy, because as you say it's not a meaningful definition of growth.

    re: "But, if the GDP gain was simply due to fixing a natural disaster, it is _not_ necessarily better off than before the disaster."

    Right, but it is better off than not having fixed up the disaster. If you want to live in a utopia where disasters don't happen and measure things that way, I would respond that that's not very meaningful. Fixing things adds value with reference to the only meaningful counterfactual - not fixing things. Pretending bad stuff doesn't happen isn't a meaningful counterfactual, Silas. There's nothing wrong with recognizing that. That does not mean deliberately breaking things is a good policy.

    re: "it is better *not to have* these hits to your utility at all, then to get the hits and then deal with them."

    Right. You're not going to hear any arguments on this point.

    re: "If you believe that an earthquake+rebuilding effort is better (on any metric you purport to be useful) than an earthquake never happening at all, then you are guilty of precisely everything the Bastiat invokers accuse you of"

    Right, but nobody is saying that!

  5. Pretending bad stuff doesn't happen isn't a meaningful counterfactual, Silas.

    Which is a large part of why I don't use it. What I do say, is that you shouldn't pretend bad stuff is really good because it gooses a metric outside of its domain of applicability.

    Let me put my objection this way: what *aspect* (narrow though it may be) do you deem good about earthquakes / shoplifting / breaking windows, and why do you have such a hard time seeing why people would object to lumping that in with more "genuine" kinds of growth? Because under your and Keynesians normal, sloppy use of the concepts of "growth" and "employment", you quite avoidably give the impression that you deem all growth good, earthquakes an instance of growth, and therefore earthquakes good.

  6. OK Daniel, you're right that Horwitz was putting words in Krugman's mouth. HOWEVER (and here I suppose you can accuse me of hedging just like I was accusing you over at my blog), to be fair to Horwitz, let's quote from the Krugman piece that had set him off:

    And when we’re experiencing depression economics, by which I mean a situation in which it’s hard to create sufficient demand to achieve full employment — mainly because short-term interest rates are up against the zero lower bound — the essentially amoral nature of economics becomes even more acute. As I’ve said repeatedly, this is a situation in which virtue becomes vice and prudence is folly; what we need above all is for someone to spend more, even if the spending isn’t particularly wise.

    The trouble in practice is that conventional modes of thought tend to prevail even when they shouldn’t; in particular, public spending on the scale needed never seems to happen. That’s why Keynes facetiously proposed burying bottles full of cash in coal mines, so people could dig them up again: since any proposal to spend money on things we need got shot down on grounds of prudence and efficiency, he proposed completely pointless spending instead.

    And what actually ended up doing the trick was spending that was beyond pointless, it was actually destructive – a sort of cruel joke on the part of the gods of economics.

    The point is that it would have been much better if the Depression had been ended with massive spending on useful things, on roads and railroads and schools and parks. But the political consensus for spending on a sufficient scale never materialized; we needed Hitler and Hirohito instead.

    Now c'mon. Above you are telling Silas that "nobody is saying" that sometimes on net the country is better off because of a disaster. Well, that's not obvious from the Krugman quotation above. He is clearly saying that he prefers (A) Big Spending on roads, schools and bridges to (B) Big Spending on fighting the Axis. But it's not at all clear that Krugman prefers (C) Doing Nothing and staying in Depression for another 10 years to (B) Big Spending on fighting the Axis.

    He actually calls it a virtue (perhaps facetiously), and says we "needed" Hitler and Hirohito.

    Again, I get the context and Krugman is surely not hoping for another Hitler. But c'mon, what he wrote above is ambiguous. In fact, I myself am not sure that Krugman wouldn't want, say, a mild alien invasion in which no humans are killed, if that would provoke world governments to triple expenditures for two years. Are you sure that he wouldn't want that? (Assume the expenditures don't create any lasting capital goods; we send up missiles that are destroyed when repelling the aliens.)

  7. I always hate WWII as an example because, while you may be a pacifist, I have no quibbles at all with saying that killing Nazis and destroying the Nazi war machine was a good use of resources. It doesn't mean everything was admirable. Obviously the best thing would have been for there to be no Nazis in the first place. But killing Nazis isn't all that bad of a thing to commit to.

    So let's put it in less... palatable... terms. Would Krugman have preferred invading a random country to prolonging the Depression for ten years? Do you think that's what's being implied here at all?

    If you do think that, I strongly disagree with you.

    If there wasn't a fascist threat I think a 1930s era Krugman (let's just call him John Maynard Krugman, for the sake of argument) would just be pulling his hair out over people doing nothing about the depression.

    Look at Japan. Krugman's been involved in that for a decade. Has he proposed any destruction over that period to cure what ails 'em? Of course not. Neither has Larry Summers. So when Summers makes a remark about the odd and perverse effect of natural disasters, let's not engage in this absurdity that he's somehow embracing it. If he embraced that sort of thing, why didn't suggest Japan blow up its nuclear power plants five years ago.

    Maybe... and it's incredible to me I even have to argue this... maybe these people are genuinely not as dumb or morally depraved as a lot of people seem to think they are.

  8. Silas is drawing out something here, Daniel, that I would like you to clarify. Up till now, I had thought Yglesias and you were saying:

    "Yes, an earthquake is bad because it destroys wealth. However, if it happens when we are in a liquidity trap, there are some offsetting benefits. It's true, on net the whole episode might still be bad, but the point is, it's not as bad as it would have been, had it occurred during a period of full employment."

    So are you NOT saying that? If there's an earthquake, are you saying its NET economic harms are the same, whether there were idle resources or not?

  9. I mean really Bob - if you really think it's not clear whether he prefers 10 years of depression to destruction, why hasn't Krugman been advocating imploding skyscrapers? Why hasn't he?

  10. Bob - that seems like an empirical question.

    I would imagine under almost any disaster scenario net benefits are negative. In a full employment economy the reconstruction effort shouldn't add to growth or employment. In a below full employment economy it could (ie - the shop keeper - insofar as he's a stand-in for all spenders - would get a new window and maybe 35% of the shoes he wanted). But even if that happens I'm still not sure why we would expect it to be a net economic benefit.

  11. Using war (or other destructive events) in your rhetoric is just a dumb idea whether you are right or wrong. Half of Krugman's problem is that (no matter how brilliant an economist he may be) he is a crappy salesmen for his viewpoint. Just like lots and lots of evolutionary scientists are crappy sales people for their viewpoint (which I happen to agree with).

    As for people saying things about Krugman you don't like, well, he says things about people that they don't like (some of those things he says are incredibly dumb too). Krugman stepped into the world of the public intellectual of his own will presumably; that means he gets to mix it up and give as much as he gives. I really don't cry for the guy or feel sorry for him.

  12. Let me make my second point a little stronger.

    Krugman isn't a victim; he is a well-paid, Nobel prize winning economist who gets a daily comment posted at the New York Times. He has to suffer through the harrowing experience of people making fun of him (which is what people ought to do to prominent people IMO) or telling him that he is wrong in sometimes unsavory ways (which is what people ought to do to prominent people IMO). Boo, fucking, hoo.

  13. Let's not overdramatize this, Gary. Krugman is certainly not a victim and that's not the claim. The claim is that there are a bunch of tiresome, incurious people out there who seem eager to snipe at this one guy for some reason, and who ultimately clog up the internets which could be mulling over more useful questions. Nobody's shedding a tear for Krugman. I'm pointing out that dumb arguments are dumb arguments, don't feel bad about doing that.

  14. Daniel,

    If you don't think that Krugman provides plenty of material for people to snipe at then you aren't paying attention. I don't want to get into psychological motivations here, but Krugman is getting exactly the sort of response an outside observer would expect him to get.

    You know, Milton Friedman also used to say things he probably knew would tweak a few noses or two (see his column in Newsweek or his comments on _Donahue_); and those tweaked noses responded (but not in the same sort of visible way that they can today).

    One can think of it as all part of the game.

  15. Gary I really don't care that people disagree with Paul Krugman.

    Your Friedman example is perfect. Yes - Friedman, like Krugman, knew he was saying things in a way that put himself out there. And people did critique him big time. But there are very intelligent ways to take issue with something Friedman says, and there are very dumb ways to take issue with what Friedman says.

    Some of the criticism of guys like Krugman is great. But a lot of it is comparable to Naomi Klein's attacks of Friedman: really dumb. And it's worth pointing out that it's really dumb.

    I don't think Milton Friedman should have expected not to be controversial for people, but that doesn't keep me from saying that Naomi Klein can sound like a real dumbass.

  16. So when is the third Hayek v. Keynes rap video due out?

  17. "Bob Murphy wasn't correcting me on the major point of my post..."

    He sure as hell was.

  18. Nice on the broken window fallacy hier:

  19. Ummm, no Gene. People still do a real shitty job criticizing Keynesians using Bastiat. As I said above - if you think this has all been a discussion of the extent of crowding out, you're deluded. That's why I gave all those examples. Do I need to dig up more?

    I suppose Bob might have thought he was criticizing my point, but again - I think he sees the whole blogosphere-wide debate a little differently than I do.

    It was a good post - don't get me wrong.

  20. I feel like this is going to be pointing out the obvious here, but then that's why I spent three hours ensuring that nobody else had previously said it.

    Taxes taken are a destruction. Six francs of money paid to the government and six francs of money paid to the glazier are, in either case, six francs of money not put to one's own end.

    The argument that we can tax money and then use it elsewhere to stimulate the economy is very nearly the logical equivalent to the argument that we can break windows and thereby stimulate the economy.

    This is the point of using Bastiat to criticize Keynes. People are really shitty at doing this, sadly. In three hours of reading about the Broken Window fallacy (and that's today only; I've read much more in the past), I found only one oblique reference equating tax takings and glazier takings.

    I hope it pleases you to forgive me for posting anonymously. I find it discomfiting to argue online under my own name. I lack confidence that academic disagreements will remain academic; too many people are too quick to invoke emotional content.

  21. Anonymous -
    Do you think this might have something to do with why Keynesians don't think tax increases are a good idea during a depression, and why tax decreases are in order?

    Now, even in a depression if the government spends all of a tax dollar where a consumer would only spend part of it and only invest part of the residual (keeping the rest liquid), spending out of tax dollars wouldn't be the world. But it's still weak fare and as you point out it's distortionary, which is precisely why you hear Keynesians arguing for increased deficit spending.

    Do some Keynesians have a chip on their shoulder about the Bush tax cuts? Sure. I don't. I didn't think any of the cuts should have expired. But nobody was arguing that letting them expire would be stimulative that I'm aware of.

    This is what was so goofy about the guy who set up the fake Krugman account - he was talking as if Keynesians wanted to tax and spend.

  22. Sorry Austrians, but you simply reveal yourselves to be more extreme and obtuse every time you open your mouths (or touch your keyboards). I have yet to encounter an Austrian who ACTUALLY UNDERSTANDS KEYNESIAN ECONOMICS (caps for angriness).

    The pinnacle of your contribution is either vacuous whining about bureaucrats, or simply completely missing the point theoretically (Hazlitt, Hunter Bryce). This is why you lost the debate (were annihilated) in the 1930s and why nobody takes yous seriously in academia.

    Civility is generally my forte but you guys just don't get it.

  23. Summing up the above argument:

    Keynesians good. Austrians bad.

    Did I leave anything out?


All anonymous comments will be deleted. Consistent pseudonyms are fine.