Great Short Video on the Broken Window Fallacy

Steven Horwitz

Tom Palmer and Austin Petersen from Atlas have put together a nice, short video looking at Bastiat's Broken Window Fallacy by asking why people think natural disasters are good for the economy.  Tom does an excellent job in explaining the fallacy with several useful examples (if 9/11 footage bothers you, be forewarned).  Along with Sean Malone's nice piece on profits and, of course, the now million views SAMMIE winning "Fear the Boom and Bust," we are getting quite the stable of video treatments of good economics, all of which are highly usable in the classroom.  These are exactly what is needed to reach younger people.  Great work by all.

Bonus points to anyone who emails a copy to DeLong.

Comments

[Jerry O'Driscoll]

Excellent video. Thanks for the post, Steve.

[Lee Kelly]

Are there any conditions in which a natural disaster (or similar destructive event) could be economically beneficial? I do not mean to defend the mistaken reasoning of those quoted in the video, but merely to ask whether it is possible for a similar claim to escape the fallacy of the broken window. It seems to me that such conditions could exist, even though I suspect they rarely do.

The primary condition is a monopolisation of the money supply. I suppose it needn't be a government monopoly, though it probably would be. The secondary conditions would a socialisation of the risk of owning money, and a failure to sufficiently expand the money supply to an increase in demand.

Anything that lowers the relative risk of owning money will increase the chances of a sharp increase in money demand, because owning money is perceived as a safe alternative to other assets. Government monopolies socialise the risk of owning money: legal tender laws, deposit insurance, bailout precedents, and tax payment each ensure that money retains an artificially high demand. In other words, the risk of owning money does not rise and fall proportionally to the risk of owning other assets. Thus, present government policies intensify swings in the demand for money during times of panic and uncertainty.

A more volatile demand for money requires more timely and accurate changes in the supply of money. However, central banking institutions popular with governments around the world are limited to extremely crude means of estimating the supply and demand for money. Even if central banks were capable of satisfying changes in money demand in a timely manner, none (so far as I am aware) make it their goal. In fact, most central banks ostensibly pursue modest inflation targets--a goal that can only be achieved by purposefully engineering monetary disequilibrium.

These two conditions produce a situation favourable to the emergence of Keynes's "paradox of thrift." Since this is quite a controversial claim, perhaps I should further explain its rationale.

What is true for a part of the economy may not be true for the economy as a whole. Although an individual may increase his savings by reducing expenditure below income, aggregate expenditure and income must be equal--one's expenditure is another's income, and vice versa. Thus, any change in aggregate expenditure must correspond to an equal change in aggregate income. Unlike an individual, it is impossible for the economy as a whole to reduce expenditure below income.

If the economy as a whole cannot save by increasing cash balances (currency + checking accounts), then how can it save? For the economy as a whole to save it must reallocate resources from producing consumer goods to capital goods. This can be achieved not by reducing aggregate expenditure below income (which is impossible), but altering the composition of expenditure/income, i.e. reducing expenditure on consumer goods while increasing expenditure on capital goods. Although sellers of consumer goods see business decline and workers laid off, sellers of capital goods goods see business rise and workers hired, and the composition of aggregate expenditure/income is altered without any change to its total.

It may be appropriate to distinguish between three economic processes by which expenditure (and resources) are reallocated between the production of consumption and capital goods. The first two are simple and uncontroversial, but the third may require more explanation. First, direct expenditure on capital goods instead of consumption goods (equity financing). Second, indirect expenditure on capital goods instead of consumption goods by purchasing bonds and the like (debt financing). In both cases, the composition of expenditure changes while the money supply remains constant.

The third process occurs with financial intermediaries that offer transactions accounts and/or currency, and their reaction to an increase in demand for their money. For the sake of simplicity, I will hereafter refer to these financial institutions as banks.

Although we call money placed in a bank account a "deposit," it is actually a loan to the bank. When one "deposits" into an account, one is relinquishing ownership of an asset in exchange for bank IOUs. An account balance is not a record of how much money a customer has "in the bank," but a record of a bank’s debt. Since bank customers do not frequently try to redeem all their IOUs at once, banks can operate successfully while holding only a fraction of total "deposits" on hand, and lending the remainder out to earn a profit. Although customers run the risk that they may be unable to redeem their IOUs if too many other customers decide likewise at once, they are compensated by the elimination of storage fees and perhaps the accrual of interest.

Since customers' accounts do not decline even as new accounts are opened for borrowers, banks can lend money into existence. However, banks' money creation is constrained by the rate at which customers attempt to redeem their IOUs, i.e. spend money. When customers' aggregate expenditure rises or declines, banks can operate with a larger or smaller fractions of "deposits" on hand to satisfy redemptions. Therefore, if customers save money by reducing aggregate expenditure, banks experience fewer redemption demands for their IOUs, and are empowered to create more money through credit expansion.

Assuming that people would prefer to lend out all their money which is not immediately being spent if the risk of not being paid back on demand were low enough, the expansion of credit by banks in response to an increase in demand for their services (checking accounts and/or currency) has equivalent consequences to increasing the expenditure on capital goods by purchasing bonds and the like. The money supply may expand in the former case and remain constant in the latter, but aggregate expenditure and its composition between consumer and capital goods will be the same. In other words, the only difference in relative prices between the two processes will be in the composition of expenditure on capital goods.

Continued below...

[Lee Kelly]

Back in the real world, banks are not in a good position to satisfy changes in the demand for money with changes in supply, because governments operate monopolies on the money supply and regulate banks' money creation. By stifling a process by which saving is transformed into the production of capital goods and increasing the volatility of money demand, government have intensified the peculiar conditions necessary for the paradox of thrift to emerge.

In an economy with an inflexible money supply and rapidly increasing money demand, total aggregate income declines faster than prices deflate; business profits shrink or disappear, inventories accumulate unsold goods, and workers are laid off. Although resources are "saved," i.e. released from their previous occupations, they are not reallocated to other ends, because nobody else as the enough money to purchase them at prevailing prices. A surplus of goods and labour develops. The seemingly rational response of tightening one’s belt during hard times actually exacerbates the problem by further reducing aggregate expenditure. If one’s increased cash balance is offset by another's increased expenditure, then there is no problem. However, the attempt to reduce aggregate expenditure below income is futile and actually impoverishes the future by creating recession in the present.

Although there are numerable ways to resolve the systemic disequilibrium depicted by the paradox of thrift, including deflation and credit expansion, another is to decrease the demand for money. This is the approach offered by those who advocate for fiscal stimulus. If aggregate expenditure has fallen because of uncertainty, then money must be considered a relatively safe asset. The government can increase the supply of an alternative safe asset by issuing bonds. Savers who are unwilling to part with their money for anything else may be persuaded to purchase government bonds, and the borrowed money can then be immediately spent to stimulate aggregate expenditure. Whether one agrees that this is a worthwhile trade-off given other problems with fiscal stimuli is another matter.

Finally, we now approach the thrust of my argument against the fallacy of the broken window. One way of reducing money demand is to destroy assets that people value highly, because while they may be unwilling to spend money on a new TV, they probably will be willing to spend money to replace a destroyed home or car. Although I would never advocate breaking windows as a means of resolving a disequilibrium in the money supply, it is possible that by destroying assets in the short run, money demand can be decreased and the economy more quickly brought out of recession. Thus, the long term benefits to "breaking windows" may exceed the short term costs.

Sorry that was so long, but when one is arguing for a possible exception to the broken window fallacy, one must explain themselves.

[Jerry O'Driscoll]

Printing pieces of paper cannot create wealth.

[Lee Kelly]

Jerry,

It seems to me that you are wrong. Printing pieces of paper really can "create wealth," at least in the sense that it may cause the creation of wealth by people. In normal times, when the economy is not in a state of severe monetary disequilibrium, you are surely right. However, a severe excess demand for money may bring about a situation where printing and distributing prices of paper can reallocate resources toward productive ends that would otherwise be neglected. Of course, I doubt such a situation would develop without a central bank and/or stifling financial regulation, since private enterprise would quickly capitalise on changes in money demand to make a profit. However, should such a situation develop given the present system, printing pieces of paper might just "create wealth" in the only sense in which it matters.

[Jerry O'Driscoll]

Lee,
I can't sensibly respond to an asserted "disequilibrium" without knowing what propogated it.

[Lee Kelly]

Jerry,

Why not? If there is a shortage of automobiles, cheese, houses, televisions, or whatever else, then is it important for entrepreneurs to know why there is a shortage? Isn't the beauty of the price system is that nobody needs to know such things, but merely responds to shifts in relative prices to satisfy consumers demands? Why is it important for a producer of money to know such things?

If a recession is preceded by a systemic misallocation of resources ala the Austrian story of the business cycle, then the composition of aggregate expenditure will emerge differently than otherwise. Certainly the government should not attempt to prop up genuine malinvestment. An increase in the money supply to satisfy new demand should be injected in a manner that spending is allocated by market forces and not government dictate.

Perhaps I do not understand your objection.

[Lee Kelly]

By the way, the intent of my really long comments was to say that while I do believe there may be an exception to the broken window fallacy, I do not believe the necessary circumstances would arise without government suppressing the ordinary equilibrating tendencies of markets.

[Jerry O'Driscoll]

Lee,

What entrepreneurs know operating in markets, and what economist observers know are completely different. One is concrete knowledge of circumstances of time and place, and the other is abstract knowledge of the workings of a theory or model.

If I don't know the cause, I can't predict endogenous responses. Did prices rise because of a demand shock, or a supply shock? Real or nominal shocks?

If you believe printing paper can create wealth, then you literally believe wealth can be created out of nothing. No matter how complex te story, that is the bottom line.

[Bill Woolsey]

To the degree the video is supposed to show that there is always adequate nominal expenditure, then it is wrong. However, it is the case that breaking someone's window reduces the victim's real demand for something else and results in lower well being.

If the shopkeeper ends up holding lower money balances because the window is broken, he is worse off.

In my view, the real point is not about nominal expenditure but rather scarcity. There aren't enough resources to produce all the goods and services that people can use, and so the key social problem is allocating resources to produce what people want most. If we break something and replace it, the resources could have been used to produce something else that is valuable. And so, the unbroken item and that other good or service could have been had. But the sacrificed good is never seen.

Monetary disequilibrium and the associated general glut of goods is a situation where the market system is not effectively achieving the social purpose of allocating resources to produce what people want most.

Depending on monetary institutions, the market process that avoids monetary disequilibrium (of the shortage variety) is either an increase in the nominal quantity of money, or an increase in the real quantity of money through lower prices and wages. Like everything else, such processes are imperfect.

I continue to find Jerry's statements puzzling. That pieces of paper can create wealth is no more surprising than that pure arbitrage can create wealth. Or, how about "paper" packaging that signifies the quality of the packaged good?

Monetary institutions that avoid shortages of money allow microeconomic processes to work better. If the market system does a better job of coordinating, then more wealth is created.

Excess supplies of money, do no good, and rather do harm. An increase in the quantity of money that matches an increase in the demand to hold money enhances economic coordination. And this improves well being.

Retail shops don't produce more physical stuff. They enhance well being.

Producing more cars when the cars are more valuable than the alternative uses of the resources, enhances well being. Producing more cars with the goods that could have been produced with the resources used are more valuable, reducing well being.

More cars are always better? No. More pieces of paper are always better? No. More cars are sometimes better? Yes. More pieces of paper are sometimes better? Yes.

I can't believe that Jerry is unable to think about increases in the quantity of money other than with the certeris paribus assumptions of given the demand to hold money and given the natural interest rate. While understanding that sort of process is important, surely one must go forward an analyze the scenarios of, say, the demand for money rising or the natural interest rate falling.

[Lee Kelly]

"If you believe printing paper can create wealth, then you literally believe wealth can be created out of nothing. No matter how complex te story, that is the bottom line." - Jerry

If that is the bottom line, then it seems to me a page is missing. Money serves a purpose, and there can be too little (or too much) of it *relative* to the needs of an economy.

[Jerry O'Driscoll]

I am reminded of an observation that my friend and former colleague Mario Rizzo has made. There is a prevelant view in macro that the microeconomic laws are suspended in "monetary disequilibrium." He and I think not.

Beyond that, I've made my point and will move on.

[Current]

I think Lee Kelly and Bill Woolsey make a good case. I think Jerry O'Driscoll is wrong about printing money, it can both create and destroy wealth in different circumstances. Like Steve Horwitz and the others around here I think Rothbard was wrong in his "hard money" theory.

The problem here though is in the "microeconomic" sense Jerry mentions.

There's a difference between a monetary disequilibrium causing a resources to become unemployed and it causing resources to lose their scarcity value. Recently Greg Ransom discussed some houses in California that were abandoned completely because of the recession. Those houses may have ceased to be economic goods. But, that case is by far the exception. In general the demand for and the prices of various types of goods simply change.

So, I don't think it's possible that a natural disaster could increase wealth. Yes, it may prevent the need for some price adjustment costs elsewhere, it may prevent the wastage of some resources. But its doubtful that that will make up for the waste of the original disaster.

[Bill Stepp]

Surely it depends on who is doing the printing. A central bank is a socialistic institution, and isn't subject to the same market checks that a free bank is. Central banks have a built-in inflationary bias, and tend to over issue their monopolized currencies. And let's not even get into its "historic role" of being a handmaiden to the State's mass murder machine.

[Lee Kelly]

"So, I don't think it's possible that a natural disaster could increase wealth. Yes, it may prevent the need for some price adjustment costs elsewhere, it may prevent the wastage of some resources. But its doubtful that that will make up for the waste of the original disaster." -Current

I disagree that it's "impossible" for a natral disaster (or some other waste of resources, like building a pyramid) to increase wealth. However, I agree that it's doubtful that its benefits would exceed the original waste.

I merely intend to point out that, *in theory,* we can imagine a situation where the destruction or waste of resources can lead to a long term gain. Furthermore, I mean to say that this situation is not only very rare, but most likely the product of governmen intervention.

[Lee Kelly]

Jerry,

Microeconomic laws are not suspended in monetary disequlibrium. The problem is that a shortage (or surplus) of money has more pervasive consequences than a ahortage (or surplus) of almost any other good, because one half of almost every transaction involves money. The laws of microeconomics aren't suspended, but rather noise (to borrow Horwitz's term) is introduced into the network of prices.

[topills.com review]

It may be appropriate to distinguish between three economic processes by which expenditure (and resources) are reallocated between the production of consumption and capital goods.