You have left your house for a walk. You intend to pass through some woods on your way to town to meet a friend. After about 25 minutes of walking, however, you stop, puzzled. The trail through the woods should have put you in town several minutes ago. What happened?
Suddenly, you remember the fork in the trail by the big chestnut tree. You always had trouble remembering whether the left or the right trail led to town. You must have turned the wrong way!
Clearly, you can still make it to town, but you'll have to backtrack. You arrive in town 20 minutes late. When you get to the café where you friend was to meet you, the waiter tells you he just left. You and he were going to see a movie, but you don't feel like going alone, so you walk back home.
Such errors are an inevitable part of human action. If the future were certain, we would not need to act. Given that it is uncertain, however, we can never be sure that our plans will come out as we expected.
It is also clear that any error is costly--otherwise, why would we consider it an error? Both you and your friend will feel that you wasted your trips to town. Given that you were not going to see the movie or even meet, there were probably other ways that each of you would have preferred spending your time. Furthermore, the unfortunate consequences were not even limited to the two of you: The theater owner lost two ticket sales for that showing as well.
All of the above is so obvious that it would hardly seem worth remarking upon. However, the prevalence of economic models that assume all economic actors possess perfect information, all plans are coordinated, and all adjustments to new data are made instantly and without cost, necessitates stressing that those features of economic models are without parallel in the real world.
Such models have a use, in that they portray a logical endpoint for the market process. But they must not be mistaken for descriptions of how the process plays out in time. As Mises says, "the mathematical economist does not contribute anything to the elucidation of the market process. He merely describes an auxiliary makeshift employed by the logical economists as a limiting notion, the definition of a state of affairs in which there is no longer any action and the market process has come to a standstill."
Because they often project the features of equilibrium models onto the real economy, mathematical economists can find many aspects of Austrian economic theory sorely puzzling. Austrian business cycle theory (ABCT), the idea that an artificially lowered interest rate leads to a distorted pattern of investment which eventually must be corrected, is particularly suspect in their eyes. For instance, economist Fiona Maclachlan writes:
[One] of the key difficulties with the Austrian theory of the business cycle... is... showing how malinvestment leads to general unemployment.
If one believes that prices work to allocate resources, there is always the possibility that a misguided, bankruptcy-inducing investment by one entrepreneur will represent a profit opportunity for another. If there's a market for used cement trucks, why should a cement truck purchased under incorrect price signals lie idle when its original owner has gone under?
The Austrians suggest that bankruptcy and liquidation lead to general unemployment when it seems far more natural to believe the causality runs the other way.
Maclachlan's comments are typical of detractors of ABCT. Paul Krugman raises a similar question: "[How can] bad investments in the past require the unemployment of good workers in the present?"
The world such economists envision in their queries is one in which information and resources flow instantly and without cost throughout the economy, and in which no capital goods are specific to a particular business. But in such a world, why would there ever be bankruptcies? The instant that the opportunity cost of owning a resource dropped infinitesimally below the marginal revenue expected from employing it, the resource would be sold, at a price equal to the opportunity cost, to another owner better able to employ it. Marginal revenue would always equal marginal cost, and no one would ever make capital gains or losses. As O'Driscoll and Rizzo ([1985] 1996: 54-55) point out, such models "must neglect both the learning and the accompanying processes… adjustments must have an infinite velocity and resources must be infinitely mobile for a process to take place at a mere instant." Maclachlan suggests that general unemployment leads to bankruptcy and liquidation, but her view of the price system precludes even the possibility of general unemployment!
For such instantaneous adjustment to occur, capital must be an entirely homogeneous blob that is always "on call." It is only under conditions of general equilibrium that the idea of capital goods as a uniform, self-replacing pool of capital makes sense. In general equilibrium, all plans are coordinated, all prices are equilibrium prices, all productive resources are deployed where they should be, and all necessary knowledge of changing conditions is instantly transmitted to everywhere it needs to be.
General equilibrium does not describe the conditions acting man faces in the real world. In the real world, mistakes are not discovered instantaneously and corrections are not costless. It is those facts that lead to the period of correction (the bust) following an artificial boom.
While driving down a local highway the other day, I passed a "For Lease" sign. The property had recently been occupied by a landscaping business. A couple of years ago, at the height of the boom, the business had built an extensive series of stone walls and fences in front of the building they occupied. My amateur eye estimated that they had put in from $50,000 to $100,000 worth of landscaping. No doubt, the business in question considered the effort to be an investment, as they were able to advertise the quality of their work on a busy highway.
Misled by the large demand for landscaping services at the height of the boom, they had not counted on a serious fall in demand as the economy slipped into recession and homeowners saw the value of their stock portfolios shrink. The owners of the landscaping business had invested as if there were more resources available in the economy than was actually the case.
The "homogeneous blob" model of capital would suggest that those stone walls and fences will now smoothly "flow" into new uses called for by the new conditions. Perhaps they might turn themselves into briefs for trial lawyers ready to sue Enron, or airport security screening machines. Instead, we see that the investment was nearly a complete loss. A doctor or accountant will move into the office, and while he will surely regard his extensive stone and wood works as quite nice, he would never had invested in them if they had not been there already.
But perhaps I'm picking out an extremely special example, particularly unfair to the case for regarding capital as homogeneous. So instead, let's look at Maclachlan's example of a cement truck. We will imagine that a construction company in Silicon Valley, at the height of the dot-com building spree, purchased the cement truck. When the bust came, construction work there dried up, and the company went bankrupt. Whither now the truck?
It is very unlikely that another construction company in Silicon Valley wants it. They all overexpanded as well, and may be looking to sell some of their own equipment. They certainly aren't looking to purchase more. But let's assume that somewhere--say, Texas--there is a company that can use the truck. The first thing that has to happen is that the buyer and seller must find each other. Rather than occurring on the day the Silicon Valley company goes bust, such a process may take weeks or months.
The courts will have their say about which creditors are entitled to what portion of the remaining assets of the company. Then the truck is likely to move to a reseller of used construction equipment. He will advertise the truck's availability. The downturn probably will necessitate a drop in the price of cement trucks, but by how much? Unlike in neoclassical models, where supply and demand curves "shift" and everyone finds, much to their surprise, that they are trading at a new price* , in the real market, buyers and sellers must undertake a process of price discovery. The seller will be reluctant to lower the price too rapidly, and so he may be overly optimistic for a time, delaying a sale. When the potential buyer catches wind of the availability of the truck, he may want to inspect it before putting out the money to buy it. Once the transaction is complete, the truck will have to travel from California to Texas.
All of those steps take time, and all of them have associated costs. Furthermore, we might note that none of them involve employing any cement workers. Eventually, a worker laid off in Santa Clara may find work in Dallas, but that also will involve time and expense.
In the wake of the dot-com bust, many workers had to make similar treks. Douglas Caldwell notes: "Santa Clara County's population dropped by 14,276 people between April 1, 2000 [roughly the peak of the dot-com boom], and July 1, 2001, according to figures released by the U.S. Census Bureau. That is more than any other county in the state." We can assume that the moves cost all of those people time and money, and that they were not employed while moving. Many of the eventual émigrés may have hung on in Silicon Valley for some time, unemployed but looking, hoping to avoid the expense associated with moving.
J. David Kuo, a former vice president at Value America, has written an interesting book (Kuo, 2001) on the history of the company's dot-com flameout. At one point in the story, he and his wife persuade his sister-in-law and her husband to quit their jobs and move from Sterling, Kansas, to Charlottesville, Virginia, to work for Value America. Within a few months, the company was closing down. The hapless couple found themselves 1000 miles from home and from their families and their support network, in a somewhat remote area of rural Virginia, without jobs.
When we get a cluster of such stories occurring together, as we do when the misallocations of a boom are revealed, we find ourselves in a recession. The lack of instantaneous adjustment to the new conditions means that for a time, some useful resources will be unemployed, while they and the complementary capital goods they need to be productive are reallocated. Furthermore, as Roger Garrison points out:
The Austrian theory allows for the possibility that while malinvested capital is being liquidated and reabsorbed elsewhere in the economy's intertemporal capital structure, unemployment can increase dramatically as reduced incomes and reduced spending feed upon one another. The self-aggravating contraction of economic activity was designated as a "secondary deflation" by the Austrians to distinguish it from the structural maladjustment that, in their view, is the primary problem.
And bad policy, such as the U.S. government's attempt to keep wages high after the initial bust in 1929, can worsen the problem, as demonstrated by Vedder and Gallaway ([1993] 1997). To an extent, Maclachlan is correct in suspecting that Austrian business cycle theory does not explain "general unemployment." Instead, it explains the structural unemployment that, under certain conditions, can lead to general unemployment.
The mainstream view of capital as a uniform blob of "capital stuff" obscures the necessary but costly and time-consuming process of reallocation. The Austrian view of capital structure as a complex of interrelated plans brings the readjustment process to the forefront. As noted by O'Driscoll and Rizzo (ibid.):
Monetary shocks produce certain types of discoordination in markets. The process of adjusting to these shocks produces effects on aggregative variables. It is not possible, however, to analyze these effects in terms of a standard macrotheory. The level of aggregation necessarily obscures the economic phenomenon being studied. If one were, for instance, to aggregate heterogeneous capital goods into "capital," the complex relationships among capital goods would be lost.
- *Thanks to Robert Murphy for this formulation.
