Broken Windows

Is it possible for a broken window to actually increase or decrease GDP from the level that it would have been had that money been spent in a way the individual would have initially desired?

For example when an earthquake happens i often see economists quoted as saying that this may actually increase GDP in the economy. I don’t see how considering the quantity of money spent is exactly the same regardless of where it is spent. Are they possibly referring to the fact that the multiplier might be bigger (i.e. construction workers are less likely to save their money and thus the multiplier is larger?)

GDP might be larger but that’s because people are spending on things they wouldn’t otherwise have spent on. If we shoot someone then there’s a whole host of things that have to be done, but it does not increase people’s satisfaction.

After an earthquake, people may indeed spend more to rebuild (or work more to offset the costs), losing comfort and savings, which would also increase their uncertainty.

So the GDP may increase, while the life of those measured by it may become more miserable.

To be honest, there’s any number of ways economists and mathemeticians could build formulas to show almost anything. All you have to do is lump the right sectors of the economy together in the right fashion and give them variables, and assign the right values to them. It doesn’t matter an economy of 310 million people is more complex than that, and that there is a whole process of production and final goods don’t just appear out of nowhere.

So to answer your question, yes, you could show a GDP increase after just about anything if you worked the math a certain way. You literally could show that a decrease in GDP actually causes an increase…in GDP. Not kidding.

“GDP” is meaningless. [1] [2]

Have a look…GDP was increased here:

(For the record, this was addressed here in this thread.)

Under normal circumstances, income spent on repairing the broken window would have just been spent on something else of value. So GDP wouldn’t have been any different had the window not been broken, but now there’s a net loss of wealth.

There is a case where the broken window will actually increase GDP. That’s the case where the money spent on repairing the window wouldn’t actually have been spent on something of value otherwise. This is the case of money hoarding. Normally whatever money you don’t spend (ie, money you save) will be used to fund productive investment which you will get a return on. Money hoarding is when you hold on to liquid money as security against an uncertain future. You might lose your job due to a recession and might not be able to find a new one, so you want your savings to be highly liquid in case you need to access them soon. When everybody does this at once, total income falls and the economy falls (deeper) into recession. In this case, the money spent repairing a broken window is not money being diverted from productive consumption or investment. It’s money that was hoarded, sitting around doing nothing. But now this money gets spent, and while in the first round it creates no extra value (since it’s just replacing a window), that money is now income for somebody which gets multiplied.

Of course the solution to money hoarding isn’t to go around breaking windows. Normally what would happen is the central bank would inject money into the economy until people’s desired money balances are achieved, which would prevent total income from falling. (Note: this is not inflationary because the new money injected gets hoarded, not spent! Rather than causing inflation, it just prevents deflation). This is the kind of policy that Hayek recommended the Fed use:

"“The moment there is any sign that the total income stream may actually shrink [during a post-bust deflationary crash], I should certainly not only try everything in my power to prevent it from dwindling, but I should announce beforehand that I would do so in the event the problem arose.” ~ Hayek, American Enterprise Institute, 1975

http://hayekcenter.org/?p=3045

As for earthquakes: an earthquake destroys part of the capital stock which will be quickly re-accumulated (consistent with the Solow growth model). So the post-quake investment will drive strong GDP growth, but again, this is just to get back to the level of GDP it would have been at otherwise (so no value is added).

Now in the case of Japan, their monetary policy didn’t stop total income from shrinking due to money hoarding. In fact, they entered into the liquidity trap, which made things… harder. So the earthquake there, if payed for through hoarded money which otherwise wouldn’t have been used productively, may actually increase total income.

Of course, nobody would recommend breaking windows, earthquake or otherwise, to fix this problem. It’s just an interesting and bizarre paradox.

Of course they have to A) contain assumptions which are reasonable, and more importantly, B) stand up to rigorous empirical testing.

Really? Can you show me how, please?

That would increase GDP, but the point others have been making is that GDP should not be confused with real wealth, which is what the broken window fallacy is about. Also, your explanation here arbitrarily assigns no value to the money itself, even though the market actors apparently find value in increasing their current money holding. So, in the case of money hoarding, GDP going down is actually indicative of more, not less, satisfaction.

It is most certainly inflationary, and distorts the market and hampers the market’s corrective action.

Gee. That’s interesting. You know, considering a most recent Nobel Prize winning economist just got through saying how “[economic models] are abstractions that make very simple assumptions that aren’t realistic.”

I’ll leave it up to the forum readers to decide which of you is a better authority on that subject.

I have better things to do with my time than create a “proves too much” mathematical absurdity argument just to show you it can be done. If you want a perfect example of this feel free to read this article, or this one.

If that’s not enough for you, pick any three times when GDP has decreased, then look at the governmental policy responses, and track them along with the subsequent GDP rise, accounting for how the intervention directly caused the increase. Determine patterns of the size of the policy response, and the size of the GDP change, assign variables to each of those details, then construct a model to plug them into which assigns a “measured” value based on the historical data to exactly how much policy response results from how much of a GDP decline, and then follow the math to show how x decrease in GDP, directly leads to y policy response, which directly leads to z increase in GDP. Decrease X causes policy response Y causes increase Z. Then follow it up with obligatory “QED.”

That’s true for a given income. The point of this is that when everybody decides to start hoarding money at once to achieve a certain level of money balances, total income falls as do total savings. It’s fallacy of composition stuff. When one person does it that person is better off. When everybody does it everybody is worse off.

Tell it to Hayek! It’s not inflationary at all since the demand for money has risen.

Yeah he’s saying economic models make simplifying assumptions. He’s not saying you can just use any old absurd assumptions. If you keep listening, he goes on to talk about rational expectations. He says, sometimes the assumption holds reasonably well, sometimes it doesn’t. When it doesn’t, we have to teach our second year class how to deal with that. He didn’t say “we make any assumption we like, no matter how absurd, and consider any results drawn from that assumption an undeniable proof”.

Hahah. You seem to be trying to argue against the use of mathematics by saying “oh, people who use maths make arguments which are mathematically consistent but don’t apply to the real world”. And this is ridiculous. Firstly, maths is logic. Pure and simple. What maths gives you is a way to start at things which are true, and find out other things which are necessarily also true. If your assumption is false, or one of your logical steps is incorrect, you will not end up with true conclusions. It therefore cannot be used to “prove” things which are clearly absurd. If you end up at an absurd conclusion, you’ve made either a logical error or you’ve started with a false assumption. In fact, there is a kind of proof in maths called “proof by contradiction” which uses this. Your “proves too much” thing is just an example of proof by contradiction (A implies B, B is absurd, therefore A is incorrect). Second, the very article you use as evidence for this shortcoming in mathematical argument gives an example of a sign with some utility analysis on it, and goes on to show that the argument containts a mathematical error. The person’s sign is not wrong because maths and reality are disjoint and incompatible, the sign is wrong because the person fucked up their maths.

I don’t know how you think maths works, but all the errors you made don’t just get ignored by people who model this stuff.

He’s saying they make unrealistic assumptions. And that (at least) some of the time, those assumptions don’t even hold up in their unrealistic mathematical world. That kind of sounds like the exact opposite of “reasonable” and “stand up to rigorous empirical testing.”

I’m sorry you believe that. Because I’m not arguing that.

The sign is wrong because it implicitly purports to show that a government can make people better off by taxing a good and then compensating the consumers of that good. And MMT is wrong because it confuses accounting identities with behavioral laws. This is not a flaw in the math. It is a flaw in using a formula that proves too much…as in, it reaches a desired conclusion for that specific case, but ignores the implications it makes in general…as in, it ignores reality in favor of making the desired claim.

And the reason people are able to get away with such nonsense more often than they otherwise would is because it is buried in mathematical jargon that generally requires a high level of training and intellectual effort to understand…when, if the arguments were made in plain English, they would easily be seen for the absurdity they are.

I don’t know how you think models work, but what I described is not far off base from what can pass as a legitmate formula. Again, GDP itself makes unrealistic assumptions that render it essentially meaningless.

And even if there wasn’t an economist who would make such assumptions and purport such a work that I described (which, I wouldn’t bet on it if I were you), but even none would, that’s irrelevant. It’s still a way to show that a decrease in GDP actually causes an increase…in GDP.

Money that is ‘hoarded’ is removed from the money supply, meaning a reduction in the money supply and so an increase in the value of money (deflation). This continues until the value of the hoarders’ money is sufficient for them to achieve their “desired money balance”, and they start spending again, with no ‘aid’ from a Central Bank required. This could still result in real GDP growth, if the deflation was greater than the decrease in nominal GDP. Falling total income needn’t mean falling real income, if prices fall faster .

Unrealistically simple. Did you even watch the video, or did you stop at that sound bite? He gave the example of rational expectations. An unrealistically simple assumption. He say he teaches it to his first year students. Sometimes it holds reasonably well, sometimes it doesn’t. When it doesn’t, he has to teach his second year students how to deal with that. That is, when it fails to hold reasonably well, they use a different model.

What does that even mean?

Okay, I’m sorry. That’s what it sounded like. What exactly are you arguing?

The sign is wrong because of logical error. The step “an increase in utility in this market => consumers are better off” is logically incoherent. Because it’s looking at one market, but total utility is derived from many markets. Compensating the consumer in the market they’ve drawn means creating utility-destroying distortions through taxation in other markets. That’s a logical (mathematical) error. It’s got nothing to do with proving too much, it’s got nothing to do with “ignoring implications in general”.

Nobody is able to get away with it. Do you think people just go, “Well, I don’t understand what that sign says. It must be true”? People who don’t understand it ignore it, and people who do understand it instantly spot the bullshit.

That’s just plain untrue. Maybe actually go out of your way to read an academic paper once in a while.

That’s a different conversation.

And if you could actually show that, then good for you. But, you can’t just make up bullshit to do it. Because as soon as you publish a paper saying that, somebody is gonna say “Whoa whoa. Hang on. That’s inconsistent with Nelson and Plosser (1982) which couldn’t reject the hypothesis that GDP follows a random walk with a drift. Somebody has screwed up”. Your peers will go through your paper with a fine-toothed comb, and if you’ve used bullshit, you’ll be called on it and discredited.

While it’s true that left to itself, an economy will usually equilibrate, there are two important caveats to that. First, the short run non-neutrality of money will result in a contraction in output and an increase in unemployment (even if GDP growth isn’t negative, it’ll be far below potential and there will be unemployment). This is unnecessary suffering which can be alleviated by an increase in the supply of liquidity.

Second, under conditions of heavy debt burden, this kind of deflation will result in the real burden of debt increasing and the demand for liquidity increasing in kind, creating more deflation, creating more debt burden, and so on. The so called Deflationary Spiral. In this case it is necessary for the central bank to break deflationary expectations.

First, the short run non-neutrality of money will result in a contraction in output and an increase in unemployment

Can you please explain how you know this? And when it has ever happpened?

…suffering which can be alleviated by an increase in the supply of liquidity.

Can you please explain how printing more money doesn’t cause any suffering to those whose purchasing power has been reduced by inflation?

Do you grasp the argument why “increasing the supply of liquidity” causes unemployment, rather than alleviating it? Can you summarize it and then refute it?

Second, under conditions of heavy debt burden, this kind of deflation will result in the real burden of debt increasing and the demand for liquidity increasing in kind, creating more deflation, creating more debt burden, and so on. The so called Deflationary Spiral.

What evidence do you have for the existence of such a phenomenon?

Also, I’m not I understand. If there is a heavy debt burden, how can people hoard money? They have to pay their debts.

See: A Monetary History of the United States, 1867-1960, Friedman & Schwartz (1963)

The most prominent examples would be the Great Depression and the Volcker Disinflation.

Yes, you may have an Austrian explanation for the cause of the 1929 recession, but the cause of the length and severity of the depression was excessively tight monetary policy.

“I agree with Milton Friedman that once the Crash [of 1929] had occurred, the Federal Reserve System pursued a silly deflationary policy. I am not only against inflation but I am also against deflation. So, once again, a badly programmed monetary policy prolonged the depression.” ~ Hayek, 1979

The increase in the money supply is just to compensate for the change in the demand for money. It’s to prevent tightening of money and income from falling.

“The moment there is any sign that the total income stream may actually shrink [during a post-bust deflationary crash], I should certainly not only try everything in my power to prevent it from dwindling, but I should announce beforehand that I would do so in the event the problem arose.” ~ Hayek, American Enterprise Institute, 1975

The expansion of money does not increase inflation at all, it just prevents deflation. And like Hayek said before, he’s against deflation too.

Increasing the money supply artificially lowers the interest rate below the natural rate of interest (the rate that equilibrates supply and demand for savings). This drop in the interest rate is taken by firms as a signal that there are now more savings available to fund investment. This expansion of credit incentivises firms to (mal)invest in capital at the low end of the capital structure, investment which requires constant injections of cheap credit to remain viable. When the interest rate finally equilibrates back at the natural rate, these malinvestments are discovered to be unsustainable at the true, undistorted, market price. Due to imperfect substitutability of capital, this capital cannon be instantly put to work producing the goods that would have been produced had market prices not been distorted. Unemployment ensues as the capital devalues and the recession undoes the allocative distortions of the boom.

May have missed some stuff because I’m on the spot here.

My problems with that story are:

  • Why do firms take only the current interest rate as a signal of the profitability of investment? Why don’t they look back and think “it’s likely that in the near future the interest rate will rise, rendering this project unsustainable.” and either not invest or at least take out some interest rate futures? The rise in the future interest rate, especially since central banks nowadays have taken great strides to make their monetary policy transparent and predictable, should not come as a surprise to anybody.
  • Most people who tell this story never seem to acknowledege the fact that the natural rate of interest can change, even become negative.
  • Why isn’t the bust confined entirely to the interest sensitive sectors? Why does it affect the broader economy, even sectors which aren’t especially capital intensive?
  • Why is the unemployment so persistent? It should just be structural/frictional stuff lasting maybe six months to a year as workers relocate to the places where they’re most productive after the relative prices revert.
  • Why isn’t a boom in investment associated with a slump in consumption? The story is that relative price distortions reallocate workers to producing capital goods away from producing consumption goods. But consumption and investment are both procyclical. Both of them falling requires a fall in total income, and that’s the Keynesian story.
  • Central banks nowadays don’t just drop interest rates on a whim. They usually operate on some sort of Taylor rule where they only ease monetary conditions if output and inflation deviate from target rates. Australia for example kept the interest rate unchanged for a full year until this month. The Austrian story makes sense if central banks lower interest rates for no reason, and this excessive easing would be signalled by accelerating inflation. However, central banks only usually drop the interest rate in respons to falling inflation. So something demand side is clearly missing from this story.

The Depression, Japanese Lost Decade. See the works of Fisher, Minksy and Koo. You can see Koo talk about balance sheet recessions here: http://www.youtube.com/watch?v=HaNxAzLKegU

For every debtor there is a creditor. The creditors will hoard money, and the debtors will have to increase their demand for liquid money to pay down the debt. The further the debt burden increases, the more liquidity they’ll need.

“Yes, you may have an Austrian explanation for the cause of the 1929 recession, but the cause of the length and severity of the depression was excessively tight monetary policy.”

There were a whole host of factors which prevented the depression from being rectified, however, there is no inherent reason why a decrease in the money supply could not prolong this within Austrian theory if it helped to cause an increase in uncertainty. Furthermore a decrease in the money supply, especially one as swift and radical as the one which occured during the early years of the depression, would have to cause some sort of disruption no matter what model you are using.

“The increase in the money supply is just to compensate for the change in the demand for money. It’s to prevent tightening of money and income from falling.”

The very fact that money is not neutral implies that this cannot happen as such, there is no way to keep prices constant because the adjustment period. The closest way to do this would be to send everyone in the nation an equal check for the increase in the money supply, but even this could not prevent the changes in the price structure which would occur.

“Increasing the money supply artificially lowers the interest rate below the natural rate of interest (the rate that equilibrates supply and demand for savings). This drop in the interest rate is taken by firms as a signal that there are now more savings available to fund investment. This expansion of credit incentivises firms to (mal)invest in capital at the low end of the capital structure, investment which requires constant injections of cheap credit to remain viable. When the interest rate finally equilibrates back at the natural rate, these malinvestments are discovered to be unsustainable at the true, undistorted, market price. Due to imperfect substitutability of capital, this capital cannon be instantly put to work producing the goods that would have been produced had market prices not been distorted. Unemployment ensues as the capital devalues and the recession undoes the allocative distortions of the boom.”

I’d say that this is a fairly eloquent explanation of Austrian theory.

“Why do firms take only the current interest rate as a signal of the profitability of investment?”

They don’t, but the problem is that there is inflation which helps to add uncertainty and price increases to the entire scenario. A mere increase or decrease at the interest rate would not matter, the problem is that then input prices rise as well which means that the previous loan leves aren’t high enough.

“Most people who tell this story never seem to acknowledege the fact that the natural rate of interest can change, even become negative”

How do you believe that this affects the story? I think that they don’t seem to aknowledge it because it appears irrelevant.

“Why isn’t the bust confined entirely to the interest sensitive sectors?”

Practically the entire economy directly reacts to the interest rate in one way or other, but the entire point is that the structure of production is indeed a structure, all parts react to all other parts, so the entire economy is at least indirectly affected by this.

“Why is the unemployment so persistent?”

It shouldn’t be if prices are permitted to adjust. If they are not then the reasoning alligns fairly smoothly with the Keynesian story.

“Why isn’t a boom in investment associated with a slump in consumption?”

You lost me with this one, could you reiterate?

“Central banks nowadays don’t just drop interest rates on a whim.”

So? What matters is relative increases in the money supply which correlates to the actual savings rate and increases in the money supply, because as we all know more factors affect interest rates than just the direct actions of the fed in relation to the federal funds rate. It also really matters relatively little whether or not it is “on a whim”

Just dropping in to say that I’m really enjoying this conversation. I know Hayek had some interesting views on the idea of a basic income, but that quote on monetary policy–if actually said by Hayek–is quite fascinating.

Thanks for taking the time to reply Neodoxy. Much appreciated.

I don’t see any reason why a collapse of the money supply is inconsistent with Austrian theory either. Although, and this is just anecdotal, I find proponents of ABCT often object to the idea of any increase in the money supply at all, even if just to restore it after a contraction.

Well most inflation targeting central banks seem to do a reasonable job of it. While it may not be possible to achieve perfect price stability, you get more violent swings in the price level without accomodating the demand for money.

Thank you.

But inflation doesn’t occur until the new money gets spent. You have to actually go and buy things with new money in order for the price of them to adjust upwards. So inflation should only occur when the new capital is being (mal)invested in. I’m saying the malinvestment shouldn’t occur in the first place.

It affects the story because any time the central bank lowers the interest rate (again, this is anecdotal), I hear cries from ABCT proponents that it’ll create a recession in the next few years. It never seems to cross their mind that the natural rate of interest has dropped, and now the nominal rate has to fall also.

There’s also the point that when the natural rate of interest is negative (as there’s reasonable evidence that it is now), and nominal interest rates have fallen to the zero lower bound, the only way for savings and investment to equilibrate is for expected inflation to increase.

I still don’t understand how capital malinvestment causes, for example, a seller of retail consumption goods, to become unemployed. Why does good capital and labour get underutilised? It should just be the bad capital and its operators/builders that are exposed to the bust.

Exactly. So I don’t see why there’s so much animosity towards Keynesian theory. Rather than opposing it vehemently, the Austrian School should adopt it and just take a non-interventionist quasi-monetarist position.

Yep. So if nominal income falls, that’s the Keynesian story. The drop in aggregate demand causes income to fall, consumption and investment both fall. So if the Austrian school is rejecting Keynesian, demand-side reasoning, then nominal income will be the same and it’s just that relative prices are distorted (so production will change from making consumption goods to making capital goods). For a given level of income, you can either spend it (consumption) or you can save it where it will be invested productively (investment). So for a given level of income, stuff used for investment can’t be used for consumption and vice versa. So if there’s a boom in investment, income must be being diverted away from consumption (ie, consumption will slump). If total income remains constant, consumption and investment will perfectly mirror each other but move in opposite directions. If total income changes, that’s the Keynesian story.

So this goes back to my point earlier about how the natural rate of interest can change. If the nominal interest rate (set by the central bank) remains unchanged, but total income begins to fall, that tells you that the natural rate of interest will have gone down (and up if total income rises). The central bank will change the interest rate in accordance with, say, a Taylor rule, to make it equal to the natural rate of interest. If central banks are doing this, we shouldn’t experince Austrian business cycles since the interest rate never gets too far away from the natural rate.

Me too. I’m surprised to find so many reasonable people on here. Other Austrians I’ve run into have been very… dogmatic. I guess you’ll probably find that in all schools though.

I found that quote here if you want a source: http://hayekcenter.org/?p=3045