The Gnome Thought Experiment
I recently debated Karl Smith, a New Keynesian professor at UNC, on the possible merits of government spending (video here). Smith brought up what I consider to be the strongest argument against the Austrian position, namely that during a recession there appear to be "idle resources." According to the Keynesian view, if government spending could mobilize those resources (including unemployed workers), then it carries little opportunity cost from a society-wide perspective.
My answer to Karl (starting at the 44:00 mark) involved the Austrians' sophisticated view of the economy's capital structure. A cruder Keynesian model, which involves aggregates such as K and L, can't grasp the Austrian diagnosis of the typical boom-bust cycle, and consequently yields disastrous policy advice — such as massive deficit spending to prop up "aggregate demand."
"Idle" Resources and the Alleged Output Gap
Today's leading Keynesians tell us that the US economy is currently sitting on a host of idle resources — unemployed workers and businesses running at less than full capacity. Relying on estimates of output from the peak of the boom, they extrapolate forward and wonder why current output is so depressed. The following diagram (from Menzie Chinn) is typical:
Back in 2010, it was ironically Karl Smith himself who most succinctly made the (New) Keynesian point about the tragedy and uselessness of recessions:
We have very low capacity utilization (75%) and very high unemployment (10%).
That is, we have factories sitting idle for lack of workers — low capacity utilization. At the same time we have workers sitting idle for lack of factories — high unemployment.
There are machines waiting to be worked and people waiting to work them but they are not getting together. The labor market is failing to clear.…
This is a failure of our basic institutions of production. The job of the market is to bring together willing buyers with willing sellers in order to produce value. This is not happening and as a result literally trillions of dollars in value are not being produced.
Let me say that again because I think it fails to sink in — literally trillions of dollars in value are not being produced. Not misallocated. Not spent on programs you don't approve of or distributed in tax cuts you don't like. Trillions of dollars in value are not produced at all. Gone from the world entirely. Never to be had, by anyone, anywhere, at any time. Pure unadulterated loss.
Time and time again I see people speak about recessions as if they are a bad harvest — an unfortunate event wherein we have to figure out how to go with less. Some say we should all sacrifice — some say the sacrifice should be based on X or Y. Some say each family should take their lumps as they come.
However, they are all getting the basic idea wrong. This is not a bad harvest. The problem isn't that there is less to go around. The problem is that we are creating less, building less, making less.
We have people who would be working but are instead watching Judge Judy. We have machines that could be spinning but are literally rusting for lack of use. This is a coordination disaster.
The question is how do we end this thing as quickly as possible. How do we stop wasting our basic resources (men and machines), day-after-day, month-after-month, year-after-year.
Now to be fair to Karl, he is not suggesting in the above excerpt that the government go spend billions of dollars having people dig holes and then fill them up. His chief recommendation during the recession has been to cut payroll taxes on both employers and employees, preferably to zero. That's a proposal any libertarian Austrian would get behind, though perhaps for different reasons.
Even so, I excerpted Karl's remarks above because they epitomize what I think is wrong with the aggregative, Keynesian view of the economy. Working with such a framework, the Austrian theory of boom and bust is almost impossible to even conceive.
Attack of the Mischievous Gnomes
To illustrate what's wrong with the typical Keynesian view of the economy, in the debate I told the spectators to imagine that one night, mischievous gnomes decided to rearrange all of the capital goods and skilled laborers in the country. The next morning, brain surgeons who were supposed to report to a hospital in Albuquerque would wake up in Miami. Factory owners in Trenton would open their doors and see that their assembly lines were gone, replaced by defecating cows. Farmers in Iowa, for their part, would be baffled to see drill presses and computer servers sitting in their empty fields.
Suppose the Keynesian statisticians in this absurd thought experiment could make perfectly accurate measurements of their standard variables. What would they say? In the first place, they would be horrified to report that "real GDP" — the flow of newly created goods and services — suffered the worst collapse in world history. The annualized rate of real output in one day would have fallen perhaps 99 percent. Factories would be idle, and workers would be wandering the streets in a daze. Official unemployment statistics would be 90 percent or higher in the days immediately following the gnomes' evil prank.
The government and the public would demand that the elite economists explain this calamity. The Keynesians would check the various aggregates and consult their models. Was it a "supply-side" disturbance? It wouldn't appear to be so: the total stock of capital goods was the same as it had been the week before, and the same would be true of the supplies of various types of labor. It wasn't as if the economy had been struck by an earthquake or a plague; the same "productive capacity" was still there. Furthermore, there wasn't any negative "technology shock," as far as the Keynesians could ascertain. The recipes for turning inputs into outputs would all still work, just as they did the week before.
Depending on the specifics, the Keynesians might conclude that the tremendous fall in output was the result of flight to safety combined with "sticky prices" and nominal debt contracts. After all, when people awoke to find themselves plunged into near autarky overnight, they panicked and began to hoard their cash balances. Even though the output of goods and services plunged, it's conceivable that their unit prices quoted in dollars would fall too.
All in all, it would not be surprising in this fanciful thought experiment if most of the Keynesian economists concluded that the problem originated on the demand side. There was no physical reason for a bottleneck in production. For some inexplicable reason, workers and capital goods weren't getting together to crank out consumer goods and services desired by the public. This was clearly a gigantic market failure, demanding a massive burst of monetary inflation and large deficit spending to rescue the economy.
Unemployed Workers Not Necessarily "Idle"
Of course, the real problem in my thought experiment is that the production structure is a complex, interlocking collection of heterogeneous capital goods. It's not enough to say that the economy has the same number of workers and machines now as it did in 2007. That alone doesn't prove that the measured "real GDP" of 2007 was sustainable. (For those who prefer an illustration of this point with actual numbers, try my "sushi article.")
Steven Horwitz recently emphasized the centrality of capital theory to the Austrian paradigm:
Normally Austrians talk about the two big debates of the interwar period — the socialist calculation debate and the Hayek-Keynes debate. And those were big, particularly if "big" here means something like "had major, direct consequences on policy." It's a fun conversation to try to argue which one was more important.
But the more I think about it, and the more that the reaction to the events of the last few years plays out, especially the revival of Keynesian-style thinking, the more I think they come in second and third place.
The most important economics debate of the interwar period was the Hayek-Knight debate over capital theory.… [I]t is disagreements over the nature of capital and how decisions about its allocation are made that are at the root of the other two debates!
[L]ook at the debate we've been having here this week, as well as a bunch of earlier posts on the stimulus and Keynes. The centrality of capital to the calculation debate should be equally clear (after all, Ol' Karl knew what was at the center of matters when he titled his magnum opus).
Austrians have often treated capital theory as being an "advanced" topic that's "so hard," and one that we steer graduate students away from. I think that's been a mistake. I think it needs to be front and center in our vision of how social coordination and cooperation takes place, and it should be one of the first things, and not one of the last, we teach in Austrian courses.
The vision of a capital-using economy that starts with Menger and goes through Mises, Hayek, and Lachmann, and now our good friend Peter Lewin, is what marks the distinctiveness of the Austrian approach.
Once we understand the Austrian diagnosis of the cause of a recession — namely, that the preceding boom period allowed the economy to reach an unsustainable configuration — then we can also understand the "function" of so-called idle resources.
After the collapse of a boom, it takes time and genuine search for displaced workers and owners of capital goods to discover their best niches in light of the new information. In an unhampered market, the bulk of this readjustment would probably be over within six months. But with the government and central bank doing everything in their power to prevent this dreaded "liquidation" process, the alleged recovery churns along at an agonizing pace.
Go back to the silly gnome story for a moment. If the government and Fed sat back and let nature take its course, what would happen? It's true, the poor people in our story would suffer a dramatic collapse in their standard of living. Entrepreneurs would rush into the chaos, and try to salvage the situation by finding suitable matches between idle capital goods and unemployed workers.
The reader might at first think that it would simply take a month or so for all of the workers and capital goods to undo the damage of the gnomes, by moving back to their starting points. Yet this isn't correct. Remember that some very heavy pieces of equipment would have been moved thousands of miles; it might not make sense to ship everything back to where it started.
What's more, the transportation system itself would be paralyzed because of the gnomes' prank. Even if all the workers tried to make their way back home, for some of them it might take as much as half a year. After all, the airports, buses, and trains would all be inoperable at first, because the required spare parts, gasoline supplies, and trained personnel would have been dispersed across the country as well.
As Mises pointed out, the existence of malinvestments requires the entrepreneurs to do their best, letting bygones be bygones. In our hypothetical gnome scenario, it would be tragically incorrect for people to try to resume the pattern of production reigning on the eve of the collapse. If they did so, most people would starve to death before food output could be restored.
Instead, people would need to improvise. Brain surgeons might temporarily become manual laborers, and backhoes originally from Houston might be used to plow snow in Detroit.
The crucial point is that even the owners of "idle" capital goods and "unemployed" workers would be actively seeking out their new roles in the horrible adjustment period. Any attempt to short-circuit this process with a wave of fiat inflation or deficit spending would simply screw up price signals and make the adjustment process that much longer.
The Keynesian focus on aggregates such as the "stock" of capital and "supply of labor" leads to faulty policy recommendations. The Austrian School has always had a rich conception of the structure of production. Of the major schools of thought, only the Austrians can appreciate what happens to the economy in the wake of a gnome attack — or the collapse of an unsustainable inflationary boom.
 Purists might argue that the Keynesians would indeed correctly diagnose the calamity as a supply-side problem, because the "production function" had mysteriously changed overnight. That is, the same input of K and L would now yield a much lower real output than before. However, I could just as well argue the same about our current recession: we have the same amount of K and L as before, and yet real output is mysteriously lower than at the height of the boom. Because the Keynesians currently think we have a demand-side problem, I maintain that they might erroneously conclude the same in the gnome scenario.