Arnold Kling  

Why I am an Austro-Keynesian

Behavioral Economics Ready for... Best Advice I've Heard in Mont...

A commenter on an earlier post points me to Steven Horwitz:

As excess supplies of money work their way through the market, they cause differential effects on prices. Some go up by a lot, some only by a little. These price effects divorce prices from the underlying preferences of producers and consumers and in so doing undermine all three informational roles of prices discussed above. When the informational role of prices is damaged, economic coordination is more difficult and economic growth suffers as a result. The real effects of a macroeconomic disturbance like inflation are the ways in which it undermines the microeconomic coordination process by disrupting price signals. If the analyst begins by assuming this coordination has already occurred, as do equilibrium models, then these effects of macroeconomic disturbances will be overlooked.

I agree with the Austrian view that many important economic phenomena, particularly entrepreneurship and economic dynamics, can best be understood outside of the general equilibrium framework. I strongly suspect that a slump in output and employment is one such phenomenon. That is, it is the dynamic process of adjustment that needs to be understood, not some equilibrium rest point. General equilibrium means, nearly by definition, a situation of full employment and maximum potential output.

As an aside, I would add that the process of adjustment seems to be even more difficult when there is general deflation than when there is general inflation.

Where I part with the Austrians is in their tendency to view monetary policy errors as if they were the only source of disequilibrium large enough to cause a major reduction in employment and output. Here, my views are Keynesian, with a subtle difference.

Keynes believed that saving was determined by a hoarding propensity and that investment was determined by "animal spirits." When investment demand falls short of the desire for saving, a slump results.

I also believe that capital market psychology plays a role in macroeconomic fluctuations. However, I see the main feature as the market risk premium. When financial intermediaries are trusted, the risk premium falls. However, this can lead to a bubble, which may suddenly pop. At that point, the market sends very different signals about where opportunities lie.

For example, over the past year the shocks to financial markets in the United States have sent signals to entrepreneurs and workers to leave the housing construction industry and instead to get into, say, export industries or import-competing industries. This is easier said than done, so in the meantime unemployment rises and output falls short of potential.

I believe that shocks to the financial system often are market-generated. In contrast, an Austrian would insist that the Fed is responsible for all bad things, such as the subprime mortgage market boom and bust.

Attributing every financial distortion to Fed behavior can be almost tautological if one is not careful. Here, the Austrian bias against empiricism gives me trouble. I would like the hypothesis that all economy-wide shocks are caused by the Fed to be falsifiable.

To the extent that Austrians make predictions that sound falsifiable, they tend to be like Paul Krugman (who is not an Austrian), repeating a mantra "bad times are coming, bad times are coming" every year. Then, when bad times come they can say, "See, I told you so." It would be more interesting if every once in a while they predicted good times.

Comments and Sharing

COMMENTS (16 to date)
Matt writes:

The Fed is pro-cyclical, but so are the federal and state legislatures, as well as Microsoft.

Microsoft's reaction to the web some 15 years back caused a similar disruption, but was limited to the information software sector. Microsoft was late to the Internet party, but its arrival helped push the tech boom into exuberance.

The Fed, being a state mandated monopoly simply does to money technology what a Microsoft does to software technology. Their disruptions spread to the extent that other sectors are invested in the technology.

aje writes:

Arnold - you seem to be overlooking the entire field of Transition Economics. Austrian's predict that if countries move away from budget deficits that are funded by credit creation to more stable monetary policy, then economic growth will rise due to the coordination effects outlined by Horwitz. This empirical claim has been overwhelmingly verified following the economic transformations in Eastern Europe. If you look purely at the USA, you'll find much pessimism. But I think it's unfair to ignore the role of Austrian prescriptions elsewhere.

People often ask me to categorize my positions on macroeconomics and I have a difficult time doing so because I do not easily fit into the mold of any particular group. Just by reading the title of your post, however, I instantly knew that this was how best to clarify my position. Although, I would rather refer to myself as an Austro-Leijonhufvudian.

Too often, people are apt to recall Keynes' applied theory of aggregate demand and simultaneously ignore his general theory (yes, that should be lower case). If one reads the Treatise on Money and The General Theory back-to-back, there is much more coherence and linear transition than some would have you believe. I would, however, do without Keynes' vision of the liquidity preference in favor of the Wicksellian foundations of the Treatise. In this sense, I must meet Keynes half-way and therefore rely on what Axel Leijonhufvud calls the Z-theory.

Nevertheless, I think that the failure of the Austrians to account for real shocks is what prevents their analysis from being a general theory of fluctuations. Steve Horwitz has done some work on creating a synthesis between the monetary disequilibrium theory of the monetarists (which is somewhat similar to the Z-theory, but still focused on nominal shocks) that I think serves to better transition the Austrian theory to a more general theory.

So while I consider myself an Austro-Leijonhufvudian, I think that the Austro-Keynesian label should suffice.

MattS writes:

It's interesting that Horowitz departs from the standard Austrian model. Traditionally, it's the increase in the money supply that is the culprit for Austrians. For Horowitz, it's inflation, whether or not it was caused by an increase in the money supply. I think this allows him to avoid Caplan's substantial theoretical critiques of the Austrian theory of the business cycle. But your right about it needing to be integrated into a general theory of fluctuations, which takes into account both Keynes and Schumpeter.

Patrick writes:

Imagine a bank with $20,000 in demand deposits, $18,000 in loans, and $2,000 in reserves. The loans are spread out, so that $1,800 in loans mature each year.

A bad year occurs and 5% of the loans either default or get written down. Since loaning is more risky than it appeared, the bank must lower the interest rates it's paying to customers. This might cause customers to choose to put money under their mattress rather than in the bank. Imagine 5% of customers do that. The banks reserves drop by half, and so for the next six months it must not make anymore loans, so that its reserves can recover. This contracts the money supply and drives up real interest rates. Now, businesses who relied on low interest rates start to fail. More loan failures, more people withdraw money from the bank, and the entire system collapses.

If you look at economic history, every great bust has occurred where the economy has had a massively leveraged credit structure (like the one above). I simply don't think that Keynesian can explain economy wide busts with animal spirits.

Where the Austrians go wrong is in advocating liquidation if it's a depression-sized credit contraction. You're better off inflating the money supply, because liquidation has far too much collateral damage.

What the monetarists get wrong is that you never want to inflate during ordinary times, because that creates the credit and asset bubbles you are trying to avoid ( this is especially so when the mechanism of inflation is subsidizing borrowing, as we do now). Also, because you never allow liquidation, each time you're forced to inflate even more. The end result is either permanent economic stagnation ( as we've seen in real wages the past few decades) or a currency crisis ( which we are dangerously close to).

How do you prevent economy wide leveraged credit structures? Austrians claim you have to allow a completely free market with hard currency. In a completely free market, banks that tried risky credit strategies would always fail before they got big enough to crash the entire economy. Austrians also claim that any attempt to regulate will ultimately be counter productive because the market moves faster than government. If you regulate bank deposits, then they'll just create CDO's and SIV's. People will then call for a new round of regulation, and so on, until the financial sector is nationalized.

I don't know if the Austrians are right that the only solution is a completely free market. But I am pretty sure that we should not be using central bank policy to inflate the currency by subsidizing borrowing.

Matt writes:

If perfect markets do not exist, then equilibrium will not exit.

The question Arnbold puts forward is how do markets change, then, if they never equilibriate?

We know the mechanism, but dispute the cause. The mechanism, in all these investment shocks is sudden volatility causing investment decisions to be made over shorter estimation periods, we shorten outlook, and shorten our integration periods for estimating the discount rate, hence the general discount rate fluctuates.

Why we do this is simple, we can only handle so much volatility, so we integrate out to the point that out uncertainty quota is met.

But, the question remains, why are adjustments so rapid and volatile while seemly normal times are moderate and longer. Why do inefficient markets cause us to make sudden rapid adjustments interspersed with longer period of normalcy. Why does the volatility pile up and hit at a certain time?

Patrick writes:

Here's a great quote from Alan Greenspan:

"Central bank provision of a mechanism for converting highly illiquid portfolios into liquid ones in extraordinary circumstances has led to a greater degree of leverage in banking than market forces alone would support."

Talk about understatements! Matt - there is your answer - "leverage". Leverage is a way of creating short term profits at the risk of massive collapse. And the Fed is in the policy of subsidizing it.


Also, check out 2blowhards on the insane profits of the financial sector:

Matt writes:


I understand leverage, but I understand the same complaint is made about Microsoft. Microsoft gained exceptional advantage, or leverage, and the cost to programmers was unreal. Many times programmers were forced into complicated gymnastics because Microsoft would leverage about 20 lines of code with another 500 lines of code to disguise the original code. They did this under the facade of innovation.

I hear the same discussion, pro and con, about innovative financial instruments as I do about Microsoft.

One might argue that one monopoly was freely conceived, and the other done by state mandate. Well. actually, they say one was created because Bill Gates mom was on a charity board with a bunch of ignorant IBM executives, but that is another story.

Matt writes:

OK, I went back and re-read Brian's critique of the Austrians, and did a little bit of follow up reading.

First, about the cardinal, ordinal distinction, which says in summary, Austrians believe the marginal utility function is handled by rank, not by counting. That is, we rank the various goods, and and sequentially choose the current highest ranked good. As opposed to cardinality in which we treat utility as linear (this good is twice that good).

The distinction, I think, is incomplete markets, what our fed is dealing with now. In an incomplete market, transaction counts are too low, compared to transaction values (or vis versa?). Ben has this problem with the investment houses and banks. So, Ben is operating in Austrian mode, ranking his various customers, one by one, sequentially, according for their need for safe paper.

But, here is the catch. If the system is gaining instability, then Ben's queue of customers line up faster than he can rank them. At that point, Ben looks out his discount window and says, hey, look at this crowd! Now I can operate a more complete market, (higher transaction counts), but not a complete market (Ben is still a monopoly). Ben then opens public auction, and the months he spent ranking is resolved in days.

If this is happening, why? We have to propose a theory that incomplete markets have an energy threshold before which they can reconfigure.

Todd writes:

Wouldn't an Austrian actually contend that Fed was only indirectly responsible for housing boom/bust in that it facilitated the undue trust in financial intermediaries that you have identified as the primary cause? I don't know what the historical record says, but since you've indicated empiricism as the appropriate standard, I'd be interested to know of the relative frequency and size of "market-generated" financial shocks prior to and after the creation of the Fed and/or the switch to a pure fiat monetary system. It may be hard to do given all the different variables that could also have an impact, but that's what economists are supposed to be good at, right, isolating variables?

fundamentalist writes:

I have a problem with the use of shocks as explanations in economics. It doesn’t seem very scientific because shocks don’t explain anything; they merely state that something happened. Are shocks supposed to be the random clustering of events that have no explanation? If business successes and failures are randomly distributed, it would make sense that you would get occasional clusters of success that we would call booms and clusters of failures that we might call busts. But does the data suggest such randomness? I don’t think so. I’ve studied Keynes’s “animal spirits”, Schumpeter’s productivity shocks, the neo-Keynesian sticky-prices/sticky-wages, and the neo-classical expectations. Each appears to be true, but not comprehensive. Each seems to be describing a small aspect of the business cycle, whereas the Austrian theory encompasses all of the above and then some. The situation appears to be very much like the old story of blind men trying to describe an elephant. Each focuses on one attribute of the elephant, such as the trunk, legs, ears or tail. Each one is correct in what he describes, but he doesn’t describe the whole elephant. In the same way, I think the other business cycle theories are true as far as they go, but describe small aspects of the whole cycle. The Austrian theory sees the whole picture.

One of the great scandals of econ has been the divorce between macro and micro. The two still aren’t on speaking terms in mainstream econ. I would think the econ profession would show more gratitude for Austrian econ because it never had that problem and successfully marries macro and micro, and thereby solves a problem that econ professors should be very embarrassed about, but don’t seem to be.

The “muddle that is macro” describes mainstream econ, not Austrian. Kling’s rejection of Austrian econ strikes me as someone who rejects a proposal from Miss USA because she’s too thin.

fundamentalist writes:

Todd: “Wouldn't an Austrian actually contend that Fed was only indirectly responsible for housing boom/bust…”

When Austrians gripe about the Fed, they use the Fed as shorthand for the fractional reserve banking system. Like when Austrians say the Fed prints money, no Austrian believes that the Fed is cranking up the printing presses; it’s shorthand from monetary expansion via credit expansion. The Fed becomes the bull’s eye because it leads to and facilitates greater credit expansion than could exist without the Fed. Austrians demonstrate that credit expansion causes unsustainable booms. The late, great economic historian Charles Poor Kindleberger came to the same conclusion at the end of his book “Manias Panics And Crashes : A History Of Financial Crises,” and I don’t think he was an Austrian. When the Fed lowers interest rates too low, then all of the aspects of the business cycle that the neo-Keynesian and neo-classical economists describe take place, but they won’t happen without that credit expansion first.

An interesting aspect of the econ debate is that the practical, applied fields, such as investment advisors, completely ignore mainstream econ and fly by the seat of their pants (that is, rely on experience). I noticed this a few weeks ago when I was very sick with the flu and watched CNBC all day for a week. Although they have never heard of Austrian econ, their analyses and recommendations follow the Austrian theory very closely. Austrian econ Mark Skousen makes this same point in his book “Structure of Production.”

fundamentalist writes:

MattS: "I think this allows him to avoid Caplan's substantial theoretical critiques of the Austrian theory of the business cycle."

I found Caplan's critique of Austrian econ very shallow. I think he does battle with straw men of his own creation, not real Austrian econ.

fundamentalist writes:

aje: " seem to be overlooking the entire field of Transition Economics."

You're right, aje. In addition, mainstream econ ignore development economics and the new institutional school. Both contradict mainstream econ and support Austrian econ. In fact, my favorite course in my masters economics program dealt with the theory and practice of development in the third world. People like Peter Bauer made me very dissatisfied with mainstream econ theory. Later I read Douglass North's work in institutions. Bauer and North prepared me to break with mainstream and adopt Austrian econ.

Mike Sproul writes:


Two things bother me about your view of money:

1) What happens to the value of the dollar when new forms of money, like credit cards, are introduced? I expect you'd say that the value of the dollar would fall, but this means the credit card company is like a counterfeiter--which it isn't.
2) What happens when the Fed prints $100 and buys a $100 bond? I expect you'd say that is inflationary, but as the $100 cash is added to the aggregate spending power of the economy, the $100 bond is subtracted from it. Thus there is no tendency to inflation.

Arnold, I think you are great!

"There are three kinds of lies: lies, damned lies, and statistics."

I have seen enough of Krugman (although intelligent) to know that every economy is somehow a bad economy. The bulls or pessimists can always find a cloud in the sky. And if all else fails the news media will be out to interview the last person on earth that does not have a job.

Comments for this entry have been closed
Return to top