The Broken Clunker Fallacy

Shikha Dalmia has an article in Forbes detailing the hidden costs of the Cash for Clunkers program.

Some excerpts:
One, even if one accepts LaHood’s numbers, the fuel savings add up to only 72 million fewer gallons of gasoline every year–about what Americans consume in four and a half hours. This translates into 700,000 tons fewer carbon dioxide emissions annually–about what Americans emit every 57 minutes. …

the program might severely disrupt the ability of the used-car market to recycle parts, producing all kinds of negative unintended consequences for the environment. (Where is the green obsession with recycling when you need it?) The engine, combined with the drive train, accounts for about 35% of the value of the used car. But with this destroyed, it will make far less sense for recyclers to incur the cost of cleaning up mercury and other toxins to mine the remaining parts from the discarded vehicle. The upshot is that the car is more likely to land in scrappage with many valuable parts–engine, pistons, brakes–still intact. …

So, to recap, the Cash for Clunkers plan involves restoring the economy by destroying wealth and healing the environment by destroying resources. By this logic, we should use the stimulus money to fund a new Godzilla brigade to mow down the country and rebuild it in a more environmentally friendly way. Imagine how much richer and cleaner the planet would be.

This is about as classic an example of the broken window fallacy as you can imagine.

It’s a program built around a theory of economic stimulation that was debunked in 1850.

Seen and Unseen

In 1850, Frederic Bastiat wrote a little pamphlet Ce qu’on voit et ce qu’on ne voit pas (That Which Is Seen and That Which Is Unseen). In his introduction to the FEE edition, Friedrich Hayek wrote, “No one has ever stated more clearly in a single phrase the central difficulty of a rational economic policy and, I would like to add, the decisive argument for economic freedom.” Unfortunately, Bastiat is not often read in intro economics classes. (Neither, to my eternal dismay, is Henry Haslitt’s brilliant Economics in One Lesson, which is essentially a modern retelling of Bastiat.)

Bastiat begins his pamphlet with a simple story, the Parable of the Broken Window.

Have you ever been witness to the fury of that solid citizen, James Goodfellow, when his incorrigible son has happened to break a pane of glass? If you have been present at this spectacle, certainly you must also have observed that the onlookers, even if there are as many as thirty of them, seem with one accord to offer the unfortunate owner the selfsame consolation: “It’s an ill wind that blows nobody some good. Such accidents keep industry going. Everybody has to make a living. What would become of the glaziers if no one ever broke a window?”

The idea is simple. If the window cost six francs to replace, then the glazier makes six francs he other wise wouldn’t have made. As Bastiat says, That is what is seen.

What is not seen is what Mr. Goodfellow would have done with those six francs had he not needed to replace that window. He could have bought some new shoes, maybe a jacket for the boy, a dinner out for he and his wife, or he could keep the money in the bank (where it becomes investment capital). The glazier’s gain is the cobbler’s loss.

“Ah Ha!” The attentive reader jumps up. “The glazier will also spend those six francs in the economy! Perhaps he requires new shoes as well! The cobbler will benefit just the same!”

Yes. Money circulates, but the circulation of money, by itself, does not create anything. In this parable Mr. Goodfellow has lost value. Specifically, he’s lost the value that his unbroken window provided. The net loss in this parable is the window itself. Replacing the window has not added anything new to the economy. The six francs would have circulated regardless, but now they circulate and the town is down one window, and a new pair of shoes that the cobbler would have made.

Imagine, for a moment, that Mr. Goodfellow’s window was broken not by his own son, but by a local hoodlum under the employ of a corrupt glazier. The glazier, hoping to make six francs replacing Mr. Goodfellow’s window, pays the hoodlum a single franc to break the window. Such an act of vandalism is easily identifiable as a wrong. It’s a wrong because it’s a kind of theft. The vandal and the glazier have conspired to steal Mr. Goodfellow’s wealth.

We implicitly recognize that theft is wrong, and we rightly condemn it. But why? The money is still circulating in the economy, right? Even the loss of the window could be recovered if the glazier, aside from being a scoundrel, was also a master craftsman capable of producing windows of sublime beauty and function.

Leaving aside the problem that the character that leads a man to thieve is unlikely to also lead him to skilled, productive work, there is the more fundamental issue of property. Mr. Goodfellow is necessarily a man of limited means. When his window is broken he is deprived of the use of his wealth. To be sure, it is possible that he may enjoy the pretty new window more than the musty old one, but he is unlikely to enjoy that window as much as he would have enjoyed the new shoes he was eying. Or the medicine he planned on purchasing, or the jacket for his son, or the return he would have realized had he invested his money.

Bystanders who gather round Mr. Goodfellow’s home may decide for themselves that he needs a window and many of them may be adamant in their view. Some may be so sure of their belief in Mr. Goodfellow’s need for a new window that they would be willing to force him to purchase a new window. Or rather, they would be willing to pay a hired gun to force Mr. Goodfellow to purchase a new window. They may believe that the work of the glaziers is too important to be left to mere chance and decide to subsidize the glaziers directly.

But If we expand the scope of our little example and consider the implications for policy, we see something striking. If we formulate a policy a subsidy for the glaziers then we can see that such a subsidy would amount to nothing more than wanton destruction. Taking money from the people to give to the glaziers amounts to exactly the same thing as running through the streets and breaking windows. The tax is taken from Mr. Goodfellow and given to the glazier: just as if the window were broken.

“Aha!” Shouts the attentive critic! “The perennial gale of creative destruction! The engine of prosperity! How can destruction in general be good, but destruction in particular be inimical?”

Creative Destruction is the term coined by Joseph Schumpeter to describe the “churn’ of a modern economy. The argument is simple. Progress requires capital and since capital can’t be found roaming the plains, it must be removed from one use and applied to another. The resources that make up one business may not be used to create a new business until and unless the first business is destroyed.

In a modern market economy, competition and innovation are relentless destroyers. As companies and individuals innovate, they compete with each other to lower prices, improve quality, deliver faster service, more convenience, better flavor, or more value. As the innovation continues, competition becomes more and more fierce and those organizations that cannot keep up are driven out of business. As Schumpeter put it,

The fundamental impulse that sets and keeps the capitalist engine in motion comes from the new consumers’ goods, the new methods of production or transportation, the new markets, the new forms of industrial organization that capitalist enterprise creates. (Capitalism, Socialism and Democracy)

Of course, that destruction comes with a cost that Schumpeter recognized.

Schumpeter and the economists who adopt his succinct summary of the free market’s ceaseless churning echo capitalism’s critics in acknowledging that lost jobs, ruined companies, and vanishing industries are inherent parts of the growth system. The saving grace comes from recognizing the good that comes from the turmoil. Over time, societies that allow creative destruction to operate grow more productive and richer; their citizens see the benefits of new and better products, shorter work weeks, better jobs, and higher living standards.

Herein lies the paradox of progress. A society cannot reap the rewards of creative destruction without accepting that some individuals might be worse off, not just in the short term, but perhaps forever. At the same time, attempts to soften the harsher aspects of creative destruction by trying to preserve jobs or protect industries will lead to stagnation and decline, short-circuiting the march of progress. Schumpeter’s enduring term reminds us that capitalism’s pain and gain are inextricably linked. The process of creating new industries does not go forward without sweeping away the preexisting order.

Attempts to save jobs almost always backfire. Instead of going out of business, inefficient producers hang on, at a high cost to consumers or taxpayers. The tinkering shortcircuits market signals that shift resources to emerging industries. It saps the incentives to introduce new products and production methods, leading to stagnation, layoffs, and bankruptcies. The ironic point of Schumpeter’s iconic phrase is this: societies that try to reap the gain of creative destruction without the pain find themselves enduring the pain but not the gain. (“Creative Destruction“)

But then, if capitalism requires the creative destruction to provide capital and resources for innovation, what then are we to make of the poor Mr. Goodfellow’s broken window? Is the crowd correct? Is the shattered glass a blessing?

No. While the broken window frees resources and capital for the glaziers to ply their trade and even possibly improve the quality of Mr. Goodfellow’s window, the destruction of the window is inefficient because Mr. Goodfellow did not choose to destroy his window. The capital that flows from destruction is of a benefit to society only if that capital can be better used elsewhere. If Mr. Goodfellow was in fact about to order a new window from the glazier, then he may be ambivalent about the broken window: at most it represents a mess to swept up. But if, as is far more likely, Mr. Goodfellow was perfectly content with his window, then the destruction represents a loss to both Mr. Goodfellow and to the economy in general.

In the process of creative destruction, capital is freed by competition to be used in more productive enterprises. Creative destruction moves capital from inefficient uses to productive, efficient uses. That increase in production (faster service, better quality, lower price, more convenience, etc…) is wealth. But how do we know that one use is efficient and one inefficient? Who decides that new window is either less or more efficient than the broken one?

The inescapably evaluative nature of [efficiency] raises a fundamental question for every attempt to talk about the efficiency of any process or institution: Whose valuations do we use, and how shall they be weighted? Economic efficiency makes use of monetary evaluations. It refers to the relationship between the monetary value of ends and the monetary value of means. The valuations that count are, consequently, the valuations of those who are willing and able to support their preferences by offering money. (“Efficiency“)

The glazier may well be able to promise a newer window, but only Mr. Goodfellow is in a position to accurately assess whether or not he needs that new window because he is the one spending the money to buy the new window. Indeed, there is nothing that prevents him from hiring the glazier to replace the window at any time and he surely will hire the glazier as soon as he decides that he needs a new window. But if we attempt to force the destruction of the window because we believe the act will be “creative,” then we presume to know that Mr.Goodfellow really does need a new window more than he needs new shoes. Such a presumption would almost certainly be wrong; no matter how certain we think we may be, Mr. Goodfellow is in much, much better position to accurately asses his needs than we are.

This is known as the “knowledge problem.” In “The Use of Knowledge in Society,” Hayek defined the problem:

The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. The economic problem of society is thus not merely a problem of how to allocate “given” resources—if “given” is taken to mean given to a single mind which deliberately solves the problem set by these “data.” It is rather a problem of how to secure the best use of resources known to any of the members of society, for ends whose relative importance only these individuals know. Or, to put it briefly, it is a problem of the utilization of knowledge which is not given to anyone in its totality.

It remains demonstrably true that individuals closest to the problem have access to the best information and can make the best use of their knowledge in solving problems or determining relative value. That does not mean that the determinations made by individuals are always and everywhere perfectly accurate or optimal, but it does mean that in aggregate the accumulated decisions of free, independent actors tend to maximize efficiency. We can see this if we move outside our little parable and look beyond poor Mr.Goodfellow. While it might be true that the mob is right and Mr. Goodfellow should spend his money on a window, when we apply that judgment as policy, we immediately see its folly.

The question of whether society needs more glaziers (or teachers, or bakers, or lawyers, or auto workers) rather than more cobblers (or bankers, or tailors, or farmers, or scientists) is a question that no single person can ever answer with any degree of accuracy or honesty. The problem is that it is simply impossible to compare the relative value of a glazier with the relative value of a farmer.

It is impossible to make that comparison because it is impossible to decide the value of the glazier’s work for every possible customer. It may trivially simple to decide that I value the farmer’s produce more highly than I value a new window right now, or it may be trivially simple for you to decide that you value a new window in your car more dearly than a bushel of fresh-picked apples right now. But it is decidedly not a simple matter for me to decide what it is that you value more, or for you to ascertain what it is that I value more.

So economic planning is necessarily inefficient. But perhaps that error is relatively small? After all, it is not the mob to which we turn for economic planning and analysis, but rather our elected representatives. Our representatives in the legislature confer with each other, their economic advisers and interested lobbyists. How wrong could their judgment be?

Whenever we use tax money to subsidize a particular business or individual we rearrange the priorities of an economy.If we consider the problem of planning to essentially be a problem of selection–choosing one order of priorities over another–we can get an idea of the enormity of the challenge by looking at the number of possible states from which we must choose. This Stanford University economist Paul Romer illustrated the knowledge problem using a deck of cards. Imagine the cards as economic preferences; our job is to order them.

Imagine a set of only three cards. We can order these cards as we wish, in any of six possible combinations. The mathematics for this is 3! (3 factorial) or 3 x 2 x 1 = 6. If we have four cards, then we have 24 possible combinations (4 x 3 x 2 x 1). The problem is that the number of combinations rise quickly. With only 26 cards, there are 403,291,461,126,605,635,584,000,000 (403 heptillion) combinations. In a standard deck of cards there are more than 80 unvigintillion (8 x 10^67) combinations. If we return to our little town and imagine that Mr. Goodfellow is one of 60 households, and we simplify even so modest an economy to consider each household as one, unitary factor in the economy, then there are 60! combinations in this small town. That’s 8 x 10^81 combinations, or more combinations than there are atoms in the entire universe. Even if planning was extraordinarily wise and accurate–say 99.9999999999% accurate, the margin for error would still be impossibly, inconceivably, monstrously large (8 x 10^71).

Now, even if we don’t attempt to actually order or arrange all economic activity, even presuming to re-arrange a small portion of a large, complex economy (where the transactions number in the hundreds of millions) entails a huge degree of error. That error translates into decreased productivity, waste, and inefficiency. When we speak of subsidies, taxation and a transfer of wealth, the value calculus is simply absurd.

When the transfers are small, this margin of error may not trouble us very much. We do, after all, naturally expect that some subsidies are necessary: courts, police, defense, roads, law. However, as these subsidies grow and the share of our income that is being taken from us increases, it becomes increasingly implausible to justify that appropriation.

Let’s return to Mr. Goodfellow. If we take a third of Mr. Goodfellow’s income, it is highly likely that we will force Mr. Goodfellow to radically alter his economic priorities. He will undoubtedly save less and spend his money differently. It is simply impossible to imagine that such an alteration, when it affects the whole of a society, will not have serious and lasting effects–effects that will remain unseen.

In his introduction to Bastiat, Hayek says,

This is simply that if we judge measures of economic policy solely by their immediate and concretely foreseeable effects, we shall not only not achieve a viable order but shall be certain progressively to extinguish freedom and thereby prevent more good than our measures will produce. Freedom is important in order that all the different individuals can make full use of the particular circumstances of which only they know. We therefore never know what beneficial actions we prevent if we restrict their freedom to serve their fellows in whatever manner they wish. All acts of interference, however, amount to such restrictions. They are, of course, always undertaken to achieve some definite objective. Against the foreseen direct results of such actions of government we shall in each individual case be able to balance only the mere probability that some unknown but beneficial actions by some individuals will be prevented. In consequence, if such decisions are made from case to case and not governed by an attachment to freedom as a general principle, freedom is bound to lose in almost every case. Bastiat was indeed right in treating freedom of choice as a moral principle that must never be sacrificed to considerations of expediency; because there is perhaps no aspect of freedom that would not be abolished if it were to be respected only where the concrete damage caused by its abolition can be pointed out.