John Carney: “We’re All Austrians Now”
I have been so busy I haven’t had time to lavish praise on John Carney, who has been doing his part to bring Austrian business cycle theory to the table at CNBC. For example, a few weeks ago he ran this interesting piece. Here’s the intriguing intro:
It’s no accident that Austrian economics is newly popular. It provides the best explanation for the business cycle we just lived through.
But the resurgent popularity of Austrian economics may actually be hampering the ability of the Federal Reserve to reflate the economy with low interest rate policies. Businesses, now aware of the dangers of a low inflation- sparked economic bubble, may simply be refusing to fall for the age-old boom-bust trap.
I’m very sympathetic to Carney’s position. In fact, I had the same intuition when I wrote up a formal model–which relied on entrepreneurs who updated their beliefs according to Bayes’ Law and obeyed rational expectations–where manipulation of interest rates could (a) cause more booms than would otherwise other and (b) lead to more sluggish growth than would otherwise occur.
However, in that model (as well as Carney’s exposition), business people wouldn’t necessarily fall for the bait of lower interest rates. In my model, the idea was that the interest rate gave a signal about the underlying state of the economy, and the Fed’s interventions introduced noise.
I don’t think that is the whole story, but I do think it is closer to the original spirit of the Mises-Hayek exposition, than the Carilli-Dempster attempt to frame ABCT as a prisoner’s dilemma. For one thing, in a genuine prisoner’s dilemma, nobody ever slaps his head and says, “I can’t believe I defected! What was I thinking?!” But in the standard exposition of Austrian business cycle theory, the entrepreneurs do say, “Oh no, there aren’t enough resources to get me through this project. I wish I hadn’t started it.”
I wrote the paper when I was a professor at Hillsdale. I sent it to the RAE, thinking they would surely like an exposition of ABCT that was consistent with rat ex and Bayesian updating. But the referees were fools, which just means they didn’t like my paper.
Here is the latest version of the paper. As I put in a note, I no longer necessarily endorse the first section where I critique Carilli-Dempster. But the rest of the paper is still worthwhile, I think.
Bob, why not rewrite the article (after reviewing the most recent literature) and resubmit it to RAE?
For one thing, I’m no longer in academia and so that’s not a good use of my time, at least for the foreseeable future. For another, one of the referees was pretty adamant that this was the wrong approach.
Bob,
This is going to sound like an attack, but I really don’t mean it that way. In 2007, you predicted that the U.S. was not about to go into a recession. In 2009, you predicted that we would have massive inflation that year (I believe you said that if this didn’t happen it proved you didn’t know what you were talking about).
Now obviously nobody is perfect and you are to be commended for at least making testable predictions. However, I think you have to admit that these were some pretty major errors. If it’s true, as Carney says, that Austrian economics provides the best explanation of our current troubles, then how is it you have been so wrong in your own assessment of the situation? Is it that you don’t understand Mises? I doubt that. You’re a smart guy and you are second to none when it comes to explaining Austrian economic theory to a lay audience. No, I would submit that the problem is with Austrian theory.
Well, he is not the only Austrian economist in the word. And other Austrians have, in fact, predicted the crisis and many other things that happened. If you look at the economist who did make accurate forecasts you’ll see that the share of Austrians is actually quite large, given the small relative number of Austrians around.
As regards inflation, I think it’s just a matter of time. Austrians tend to overestimate the impact of economic actions they oppose. But they are not wrong regarding the general direction in which the economy is developing. There has been considerable devaluation of the dollar. Just have a look at the dollar index (and think of what it would look like if the Euro wasn’t in a crisis as well). The CPI is basically a worthless joke to measure inflation since they removed everything with some volatility out of it.
So I think Austrians are spot-on. They just overestimate the pace.
Besides, all those predictions can’t know how policies are going to change. When Peter Schiff made his predictions about the aftermath of the housing bubble in 2006/2007, he didn’t know how the FED was going to respond to it. There is still a considerable risk (aspiration I should say) of deflation. It all depends on what the FED is going to do. If they finally come to their minds and tighten monetary policy and stop shoring up the financial industries, America will experience enormous deflation. But that doesn’t mean that the inflation predictions of Austrian economists will have proven wrong. It just means they didn’t know what the FED was going to do.
Christopher,
The only Austrian economist I’m aware of who can be said to have actually predicted the crisis is Peter Schiff (vague statements about inevitable doom at some unspecified point in the future don’t count as a prediction).
And of course Schiff got most of what happened in 2009/2010 wrong. It’s not much of an excuse to say he was wrong because he didn’t know how the Fed was going to respond. How the Fed responds to events is a major factor in determining what will happen, so if you don’t know that you shouldn’t have much confidence in your overall predictions. And saying “it’s just a matter of time” is not any better.
Blackadder,
Thank you for your reply. Peter is not the only one, but you are right that he was the most precise. Others are Mark Thornton and Sean Corrigan. (http://mises.org/media/1214). But there is no denying that other Austrians completely failed to predict anything.
I don’t really understand why actions of the FED don’t count as an excuse. How can anyone possibly know?
Are you aware of what Mises wrote about the empirical method and mathematic calcuation? This is not really what he does. And I think he is very right in that. When you are in a bubble, whether it’s going to burst next year or in two years is immaterial. And for one to know that he would have to know millions of variables. Even trying to make predictions which require such an effort is wasted time. And it’s not necessary.
The question is not “How much unemployment is the minimum wage law going to cause?”. The question is “Is the minimum wage law going to cause unemployment?”.
Knowing whether it’s 100,000 or 800,000 is not necessary when making the decission. And you will have other effect as well. So you might enact a minimum wage law and cause unemployment but at the same time have other economic variables changing that offset the effect of the minimum wage law. So you might end up with lower unemployment despite the law. Of course, everyone would be saying “Look, since we enacted the law unemployment has dropped. So we proved those wrong who claimed the law would cause unemployment.”
So the question really is “Will credit expansion leed to (or increase the likelihood of) asset bubbles and economic downturns. Is the economic growth created by inflation sustainable?”.
It’s not “How big exactly will the bubble be and on what date will it burst”.
Anyway, was there any other school of ecnomic thought that made better predictions? I would be more than happy to know them! So let me know if you have anything.
Christopher,
Saying that if you raise the minimum wage unemployment will be higher than it otherwise would be is not really a prediction, as there is nothing that could happen that would show it to be false. If unemployment falls after the minimum wage goes up, this just means that it would have gone down more if it weren’t for the increase.
To say that Austrian theory predicted the crisis requires more than the fact Austrian theory predicts there will be crises. Keynesian theory also predicts this. So do the Real Business Cycle people. So do the Marxists. To have really predicted the crisis, you would have to be able to use Austrian theory to show not only that bubbles happen, but that there was a bubble in housing in the U.S. during 2005-2008.
There are a number of people who did correctly predict the housing bubble. Peter Schiff is one. Nouriel Roubini would be another. Paul Krugman and Dean Baker might also count. The people who correctly predicted the bubble do not all adhere to a common economic school. And for each of the above, you can find others who subscribe to the same economic theory who did not think there was a bubble. So, for example, Bob famously argued that Schiff was wrong, and that we weren’t about to go into a recession in 2007. I don’t think the problem was that Bob was misapplying Austrian economics while Schiff was applying it correctly. It’s just that Austrian theory, by itself, didn’t settle the question of whether there was a housing bubble. And since it didn’t settle the question, the fact that there was a bubble is not strong evidence in favor of the theory.
I absolutely agree with your last statement. The fact that there was a bubble in the housing market does not prove that Austrian Theory is correct. I think that Mises was right when he wrote that empirical observation is not an evidence at all. Have you read why he uses the a priori method? What do you think about it?
As I pointed out above, I don’t think that predictions about detailed happenings in the future make much sense at all. And honestly I think everyone who had it right, including Peter Schiff, just got lucky. I mean just think of what could have happened, had the government bailed out Lehman. Nobody would have put up “Peter Schiff was right Videos” on youtube. But that wouldn’t have changed the underlying true constitution of the US economy. Just change one variable and all his predictions would have been wrong.
So I really think, just like Mises, that we should look at macroeconomic policies just as you look at the minimum wage law. The question is “What happens if the money supply expands in general. What happens in a credit expansion environment in general. How does it affect investment decissions and so forth.”
Just like “What happens if we enact this law in general”.
I’d be delighted to continue this conversation via email, if you want to. I just don’t want to post my address here, so maybe Bob can help us out on that?
I’d be happy to continue the conversation privately if you’d like. I have a gmail account, username edwardblackadder25.
Just to use another amazing analogy I just thought of:
If you were about to drive your SUV into a buildung at 90 miles you could ask a physicist if it will destroy the buildung. But no reputable physicist would get into predictions as to what exact damage you would cause and which wall would break down first. To know that he would have to examine every tiny structure in the buildung, every pipe that might cause a fire. He would have to know the exact angle and speed of the car. The direction of the wind that might make fires spread into a certain direction. The likelihood of rain.
What he will tell you is that it will cause serious damage the exact extent of which is unpredictable.
QE will cause inflation. How it is going to play out exactly depends on too many aspects that noone can possibly know. No only does it depend on other economic aspects but also on a many political uncertainties. The direction is clear. But the process can easily be slowed down or even offset by other effects that are at work.
Blackadder,
You’re definitely right about the inflation call; in fact can you dig up where you got that quote from? I want to publicly lash myself.
But the early 2007 call was my fault, not the fault of Austrian economics. At the time, I was working at a financial firm. My boss was very bullish and I spent most of my time doing research along that line of thinking. I was responding to Peter Schiff when I made that horrible prediction, and he wasn’t saying (at least in the stuff I was critiquing) that, “Hey, Greenspan lowered interest rates, and that has set in motion an unsustainable boom.” No, Schiff was saying, “Trade deficits are bad, this is going to come back and bite us.”
After I left that job, and looked at the data with my own eyes, I realized we were in the midst of a massive boom a la ABCT. I wrote a forecast for a bank in July ’07 that had some very accurate predictions. (I summarized that stuff here.)
So it’s true, I personally botched things in early 2007. But it truly wasn’t a flaw in Austrian economics. My mistake was that I didn’t check the past 6 years of interest rates to see what was going on. (Back when Greenspan did that, I was just finishing grad school, getting married, and starting a job at Hillsdale College. So I was totally out of the loop back then, and didn’t follow financial news. I thought I was going to be a pure academic.)
So to repeat, I’m not saying, “Hey it’s not my fault that I screwed up.” Rather, I’m saying it’s not Mises’s or Hayek’s fault.
Here’s the post. The quote is from the penultimate paragraph:
“And to make my own position falsifiable, here you go: If the non-seasonally adjusted CPI rises at less than a 5% annualized rate in 4q 2009, I will admit I have been a fool for my warnings, and that I clearly don’t know what I am talking about.”
Btw, I understand that the main evidence Austrians point to in terms of the Fed blowing up bubbles is that interest rates were low. Suppose, though, that we had a 100% gold reserve system. Is it really the case that interest rates would never be low? If interest rates could be very low even under a free market system, then how can you tell just by looking at interest rates whether money is too loose or too tight?
When ‘Austrians’ say that the interest rate was too low, they mean lower than than the interest rate which reflects the real ratio of real savings to consumer spending.
Interest rates are ‘too low’ due to an expansion of credit, in the recent case encouraged by fed policy. The credit expansion causes interest rates to be lower than they otherwise would be given the real ratio of savings to consumer spending. This leads to the misallocation of resources and the subsequent bust.
Interest rates can certainly be low on a free market with a 100% gold reserve system if the ratio of savings to consumer spending is high. Yet, these rates would not be ‘too low’; they would reflect the real ratio of savings to consumer spending, which does not lead to the misallocation of resources.
Richard,
My question is how you can know what the real ratio of saving to consumer spending would be absent Fed distortions?
BA,
Absent distortions created by credit expansion, the interest rate would reflect the ratio of actual savings to consumer spending.
Right, but those distortions do exist. To say that current rates are too low you have to have some idea of what they would be absent the distortions. My question is how you can know that?
I disagree. In fact that is part of the problem – no one knows exactly what the rates would be if the distortions did not exist. One can only say they would be higher (ceteris paribus).
If there is a subsidy to produce a good, say it’s milk, one could say that more milk is being produced because of the subsidy without having to have ‘some idea’ as to how much milk would be produced without the subsidy (again, ceteris paribus).
Richard,
If the government subsidizes milk production this will tend to increase the supply above what it would be without the subsidy. On the other hand, suppose that the government nationalized all the dairy farms and started producing all the milk itself. In that case it would not be clear whether the supply of milk being produced by the government would be more or less than if you had a free market in milk.
Similarly, if we had free banking and the government was somehow subsidizing the creation of new money, then the supply of money would tend to be larger than without the subsidy. But that’s not the current system. We have central banking with a fiat currency, and the money supply is ultimately controlled by the Fed (I’m simplifying here, but not in a way that matters for the analogy). So just as a nationalized dairy industry might produce either too much or too little milk (relative to how much would be produced in a free market), so the Fed can produce either too much or too little money relative to how much there would be in a free market system.
Well, ok.
My milk example was just an analogy to illustrate why one does not need too know what interest rates would be without the Fed encouraging credit expansion, only that they would be higher. I do realize the Fed has a monopoly over note issue and ‘manages’ it through its reserve policies. I don’t see how my response falls short of answering your question (if that is what you mean by your response).
Richard,
My question is how do you know that interest rates would have been higher in a free market? I don’t believe you’ve answered that.
In the mind of an Austrian the interest rate in a free market would constitute an equilibrium between savings and investment. If you have more investment than saving, the interest rate is lower than it would be without central bank manipulation.
Does that answer your question?
BA,
If you are asking how I know interest rates would have been higher in a completely ‘free market (i.e. 100% gold reserve and no gov’t intervention to encourage credit expansion), I don’t. I would say the interest rate under this scenario would reflect actual savings to consumption, but I have no idea whether the subsequent interest rates would have been lower or higher than today.
If you are asking how I know interest rates would have been higher in 2000-2007 if the Fed was not encouraging credit expansion, (increasing the AMB to lower int. rates, etc) it is because the Fed was increasing the money supply during that period, which encouraged credit expansion. Therefore, I am saying, that interest rates were lower than they would have been if the Fed was not increasing the money supply by encouraging credit expansion during that period. The interest rate, because of the credit expansion, did not reflect the ratio of savings to consumption.
Richard,
I agree that, when the Fed acts to lower interest rates, interest rates will be lower than if the Fed had not acted. ABCT, however, claims not simply that interest rates are lower than they might have been, but that they are “too low,” i.e. that they are lower than they would be in a free market. If you have no idea what rates would be in a free market, then you can’t say if rates were too low or not.
But Blackadder, when an Austrian says the interest rate is “too low,” he means relative to the free-market level.
And that’s not just an ideological tautology. The market interest rate (loosely) rations the amount of real savings among potential investments. So if the Fed pushes the rate below that, then people try to start too many investments. I.e. they set in motion projects, not all of which can be completed. Some have to stop halfway through and release the resources. Aka layoffs and plant closings.
Oh sorry, I see that you understood that part. Well, if you are a Rothbardian 100% reserve banker, then any Fed inflation necessarily pushes interest rates lower than they should be. In my articles on the housing bubble at Mises.org, I showed charts giving a pretty straightforward connection between increases in the monetary base, and falls in the fed funds rate.
Bob,
I think the problem is there is a tendency to equate the status quo ante interest rate with what interest rates would be in a pure free market scenario.
Suppose, for example, that because of prior Fed distortions, etc. the interest rate was above where it should be. The Fed then acts to bring the rate down. It’s true that the Fed’s actions are causing the rate to fall, but it doesn’t follow that the new rate is “too low.”
Yeah it’s an interesting point. I think if the Fed just stopped injecting new money altogether, that eventually things would settle down and you’d have the market interest rate (conditional on there still being a Fed etc.). When the Fed inflates, that new money specifically enters the credit markets first. If Bernanke ran the printing presses and literally dropped the money from a helicopter, that would just cause price inflation, not ABCT.
BA,
I would agree with you that if the Fed was actively decreasing the AMB you might have a point, but this in itself would probably lead to a recession. Besides, looking at the AMB, don’t see any point where the Fed decreases the monetary base; it is always increasing or leveling off (and where has been leveling off you usually see a recession shortly after).
Back to your original point, I still don’t see how ABCT implies you have know that the interest rate would be higher if there is no central bank and a purely free market money system before you can surmise that the interest rate is ‘too low’. ABCT, as far as I understand it, states that credit expansions, lower interest rates below what they would be whether you start with a completely free market monetary system, or one that is not completely free.
Richard,
My understanding is that ABCT does not claim all credit expansion/lowering of interest rates is bad. It’s only when government interference pushes rates too low (i.e. below what they would be in a free market) that you get problems. Even if you had a 100% pure Rothbarian system you would still have credit expansion, increases in the AMB, etc., and interest rates would still sometimes fall. So the mere fact that credit has expanded, the AMB has increased, etc., doesn’t prove that rates are too low.
Or with trying to use the theory as a predictive tool.
We can’t know exactly what the interest rates would’ve been in a free market, which is part of the reason why the rational expectations criticism of ABC is flawed imo. But there is strong evidence suggesting that the free market interest rates would’ve been much higher in the early half of the decade given that the Federal Funds rate dropped off dramatically:
http://www.marketoracle.co.uk/images/2009/Jan/fed-funds.png
at the same time the American savings rate dropped:
http://www.p2p-loans.com/uploaded_images/SavingsRateChart-715075.gif
which only makes sense, at least to me, if the Fed was printing tons of money to keep the interest rate low. The Fed printing money (through open market operations) necessarily implies that the interest rate will be lower than it would be absent Fed intervention.
Eli,
I agree with you. Yet I believe I was correct when I responded to Blackadder that I don’t know what interest rates would be today if we lived in a free market 100% gold reserve system. If we had such a system over the last 100 years i don’t know how anyone could know what interest rates would be today. If we had all voluntarily adopted such a system 5 years ago, again, I don’t know how anyone could know what the interest rate would be today. People’s desire to save under such systems might be very different – who knows – the interest rate under such systems might be even lower than it was in 2005 even with Fed credit expansions.
Slightly OT: I remember when you posted this a while back. I suggested that you mention that your model of nature (where there’s a 1-p chance of nature’s signal of profitability is inverted) is known in the information theory literature as a binary symmetric channel, and your latest version doesn’t mention this.
That makes me a sad panda.
Also, the claim that entrepreneurs can adapt to nature and distinguish the signal from the noise given enough time is equivalent to the Noisy Channel Coding Theorem.
Silas, this isn’t “the latest version.” This was the same version; you might note the “last updated in xxx/2005” at the top of it. All I did was slap “I NO LONGER ENDORSE THIS SECTION” on the part where I criticize Carilli and Dempster.
I only posted this because it was so relevant to Carney’s post.
Oh. Sorry.
Note: This doesn’t mean you’re wrong that entrepreneurs can adapt; it means they need an arbitrarily long time to do so, and can never recover the full per-period signal.
(Also, the parenthetical above should be “… 1-p chance that nature’s signal …”