Arnold Kling  

ABC vs. Recalculation

Murphy on the Stock Market... My Immigration Podcast...

A reader asks me to explain the difference. Answer below the fold.

I think of Austrian Business Cycle theory as based on the notion that the structure of production can be long or short. Or, if you will, more roundabout or less roundabout. Should I build a car the less roundabout way, using labor on the assembly line, or should I build it the more roundabout way, which consists of first designing and building robots, and then using those robots on the assembly line? Workers are now divided up between working on the auto assembly line and working on building robots. This production process results in more capital per worker and more cars produced per worker.

If the interest rate is really low, that is a signal to use robots. If that signal is a misleading one created by the central bank, then I over-invest in robots. When the interest rate reverts to normal, I realize that I am using too much capital, and I go back to the process of using workers on the assembly line. However, the change-over involves unemployment.

The way I think of Recalculation, all production is roundabout. Take David Friedman's example (in The Economics of Everyday Life), in which we grow automobiles on the wheat fields of Iowa. That is, we grow wheat, put it on ships, send it to Japan, and the ships come back with cars. That is another way of conducting roundabout production.

The signal to grow automobiles in the wheat fields of Iowa may have nothing at all to do with the interest rate. The signal involves opportunity cost, leading to PSST--patterns of sustainable specialization and trade.

Another sense in which all production is roundabout is that we are in a Garett Jones economy. Workers are not producing final product. Instead, they are producing organizational capital.

I am more with Keynes and Minsky than with the Austrians on what induces firms to invest heavily in either physical or organizational capital. That is, I think that the interest rate may not be the primary factor. I see a role for "animal spirits." I also see a role for a Minsky cycle, in which there are periods where firms are very conservative, followed by periods where they are more speculative, followed by periods where they create Ponzi schemes.

A problem with both ABC and Recalculation is that you still need to explain unemployment. In the ABC story, why is it that we get unemployment only when firms are shifting from more roundabout production to less roundabout production, rather than the other way around? In the Recalculation story, why is it that sometimes the patterns of production evolve along paths that involve low unemployment, while at other times they involve high unemployment?

I have attempted to answer this question before, perhaps not satisfactorily. My thinking is that when new, more efficient patterns of production emerge, people are "pulled" into them gradually. That does not involve unemployment. However, when existing patterns of production break down because of sudden changes, they are "pushed" out of their jobs and they do not know where to go.

I use an immigration analogy. If people come to this country because employers lure them with good jobs, then the immigrants are employed. Instead, if people come as refugees fleeing war and starvation, they may very well not be employed when they arrive. Eventually, entrepreneurs will figure out what to do with them.

The way I see it, the end of the housing bubble, the death of the mortgage securities market, and the ongoing restructuring of the economy caused by the Internet and globalization all have combined to dump a large number of refugees into the labor market. The process of entrepreneurial trial-and-error to find PSST using these refugees is gradual. Maybe government policies are speeding up the process. Maybe they are slowing it down. I do not claim to know.

Anyway, the differences that I would emphasize with ABC are: I attach less importance to the interest rate, and less importance to the central bank; I emphasize that labor acts like capital; and I think that there are many dimensions along which production can be restructured, not just purely "longer" or "shorter."

Comments and Sharing

COMMENTS (11 to date)
Bill writes:

Some Austrians will probably answer, and I look forward to the debate, but it seems you have a fairly narrow view of Austrians.

I don't see much difference, other than the Austrian view of central bank intervention. You wrote: "Maybe government policies are speeding up the process. Maybe they are slowing it down. I do not claim to know."

I think Austrians would say that central bank interventions are designed "specifically" to retard the flow of recalculation; but not necessarily that the central bank is the only thing that causes cyclical behavior in the economy.

crossofcrimson writes:

A couple of years ago Robert Murphy provided one of the best articles on the ABCT point of view that I've seen to date. It might be worth a read for anyone who wants to compare and contrast views, or beg more questions possibly.

RPLong writes:

I'm not an expert on ABC theory, but it seems to me that malinvestment takes more time than market corrections.

It's not as if one bad central banking policy is enough to induce a huge and widespread surge of malinvestment with a corresponding surge in structural unemployment. Instead, the steadily increasing money supply induces a gradual migration of capital investment and employment.

But the market correction isn't gradual at all. It's induced by the sudden realization of investors that their investments are unsound. First the investors pull their money, and instantly business valuations change, thus determining the course of events for each of those individual business firms.

I see no reason why the initial malinvestment would come with the same kind of unemployment surges that the crack-up boom unleashes.

Charlie Deist writes:

I think I remember Roberty Murphy explaining why business cycles lead to a low-unemployment boom and high-unemployment bust in "push/pull" terms as well. During a boom, firms producing capital goods (for more roundabout production) bid up the price of labor and "pull" workers from consumption goods sectors.

Rick Stewart writes:

1) As a businessman, I didn't calculate NPVs based on the interest rate, but on our (much higher) discount rate.

2) In today's economic climate, I would probably insist on adding 1 or 2 percent to our discount rate (it was 19.5% in the '90s), based on the current uncertainty of future government action. This might not sound like a lot, but it would definitely give some previously worthwhile projects a negative NPV and thus they would not be started.

3) My sunk costs for robots would be irrelevant in my decision whether to continue using them, or to switch back to people. Intuitively it seems unlikely I would quit using my robots, although I might not buy more (based on NPVs, not the interest rate).

Troy Camplin writes:

My understanding of ABC is that it also involves the misallocation of capital -- including human capital. Here is my discussion of it:

I invite comments and criticism.

fundamentalist writes:

The fact that lower interest rates cause an expansion in capital goods production and not in consumer good production is one of the most well documented events in economics. It was first noticed by Cantillon in the early 1700's and affirmed by Smith, Ricardo and many others. Instead of "roundabout", it's better to think of investment becoming more capital intensive, that is, producers choose to use robots instead of labor. In order to understand the ABCT, we have to disaggregate production into at least capital and consumer good production. Otherwise, the ABCT doesn't make any sense at all. The ABCT hinges on the necessary coordination of scarce resources between capital goods production and consumer good production.

fundamentalist writes:

Rick, you would be right about continuing to use the robots if you had the inputs necessary, such as the metal and rubber. But in the ABCT those become scarce and expensive because they are more profitably used in the production of consumer goods. So while the price of inputs are rising due to scarcity, the demand for capital goods is falling because consumer goods makers are using more labor and not buying capital equipment. That's what makes the robots malinvested capital.

Hunter writes:

I think you're right about the number of "refugees" being the determinate as to how good or bad a recession is. Basic economics say that when prices rise due to inadequate supply, entrepreneurs/ suppliers will rise up to produce the needed supply. Since investors tend to be lemmings and the propensity for people's overconfidence in their ability to take advantage of an opportunity and add to that the distortions that caused in asset markets in general because the price of asset goods are based on the last unit sold and not the amount of the entire stock of goods you have a recipe for uneven overproduction and malinvestment of any given good and this is constant. Normally this isn't a problem but if the size of a recalculating sector is big enough and fast enough it will disrupt your PSST far outside normal and take a lot of other more or less sound businesses with it necessitating a more complex recalculation. Great post. I wish we knew how to speed up the recalculation though.

Noah Yetter writes:

Your characterization of ABCT isn't wrong, but it is woefully incomplete.

1. The structure of production is per-sector. In a single boom phase it can be distorted longer in some sectors and shorter in others. Misallocation of capital resulting in sectoral shifts is a central part of the story.

2. The influence of the central bank is not necessarily through interest rates. Interest rates decline as a symptom of rampant monetary expansion. The Fed gives new money to banks, and banks want to get rid of it, so they invest it. A key point is that this inflationary influence can come from elsewhere than the central bank (see TAL's The Giant Pool of Money).

3. The classic S/I model suggests that any marginal investments using these funds will be just that -- marginal. Given savers' true supply of loanable funds and investors' true demand for investments projects, these new investments would not be undertaken. Here the key point is that the banks who are the initial recipients of inflated funds have no incentive not to move them, even if they believe ABCT, and the investors who receive the funds can't tell the difference (nor would they have an incentive either).

4. By random chance some of the marginal investments will prove profitable (e.g. eBay, Amazon) but most will fail as the contradiction is eventually revealed. Unemployment results because rapid reallocation of resources across sectors and degrees-of-roundaboutneess is impossible.

I've been reading your Reallocation posts for quite some time. Aside from the Minsky Cycle and Garrett Jones angles, it seems to literally be ABCT. And with a little theorizing, I think Jones could be integrated in a productive way.

Jacob Hedegaard writes:

I think the trouble with Recalculation is, that it doesn't explain why the need for recalculation occurs.

A satisfactory answer to this would improve a lot...

Comments for this entry have been closed
Return to top