One useful insight from Austrian business cycle theory

I am not an Austrian, but...

Three decades ago (ouch!) I made a valiant effort to understand Austrian business cycle theory. I failed. I wanted to believe it, because I found the existing theories (Old Keynesian or New Classical) unsatisfactory. But I couldn't get it to make sense.

Despite that (and never mind who's to blame for my failing to make sense of it), I did get one useful insight out of reading that stuff. It's an image which has stuck in my mind ever since, and which I was reminded of a few weeks ago when I saw a photo of a construction site, where work had stopped on an unfinished building (I'm sorry, but I can't find the blog where I saw that photo). And I was reminded of it again when I read Stackelberg Follower's post on potential output. And again, during the CD Howe Monetary Policy Council meeting, when a participant (non-attribution rules) said that relative demand for different types of labour had changed.

Here's the Austrian metaphor (where did I read this?): an architect has blueprints for a new building, and a stack of bricks. For some reason (that's where they lost me) the blueprints get accidentally magnified. They start the construction, but run out of bricks halfway through. They can't get more bricks for some reason (they lost me again there) and so the building remains half-finished. During the boom, the investment seemed twice as big and twice as valuable than it should have been. When reality dawns in the credit crunch, the half-finished building is worthless. It's a parable of malinvestment. It only makes sense when you disaggregate investment.

Watch for signs of this parable. Sometimes they will be literal, when we see partly-finished construction projects on which work has stopped. Other times the signs will be less literal, when we see there has been too much investment in some capital goods and too little investment in other capital goods, so that the total value of all the capital goods, in terms of producing things that people want to buy, is lower than it would have been. And sometimes the malinvestment will be in human capital, not physical capital. Some people won't complete their training. Others will realise they have training in the wrong skills.

During the boom, the value of output and income appeared higher than it now appears to have been, because some investment was not as valuable as people thought it would be.

The validity of this Austrian insight does not mean we should follow the rest of the Austrian prescription, and just let the recession take its course. But neither should we ignore it.

We should recognise that it would be unwise to estimate potential output by extrapolating from past actual output. With hindsight, past actual output was less valuable than we thought it was at the time. (This was Stackelberg Follower's point, if I understand him correctly). More importantly in my view (since I am always leery of GDP, and even more leery of estimates of potential GDP), we need to recognise that the natural rate of unemployment will be temporarily higher than it once was as workers reallocate.

This does NOT mean that all increases in unemployment are due to an increased natural rate. Rapidly-falling inflation should convince us otherwise, as should widespread declines in employment across many sectors.

It does NOT mean we should not use monetary and/or fiscal policy to increase aggregate demand. It does mean we should recognise the limits on aggregate demand policy. It does mean we should also focus on improving labour mobility, both occupationally and geographically, using fiscal and/or other policies. (Geographic mobility will also be harmed by home-owners with negative equity being unable to switch houses, and that is an important area for public policy on which I really wish I had some good ideas, because I'm sure something useful could be done).

Pick and choose. We don't have to agree with everything someone says to be able to learn something useful.

(And Austrians: yes, I realise that the blueprints got magnified because interest rates were too low, but I couldn't make sense of the idea that monetary policy could set (real) interest rates too low in a world of perfectly flexible prices. So I just like to add some Austrian sauce to a New-Keynesian/monetarist/whatever meal.)

Comments

[Patrick]

I think the logic works if you insert market failures to explain some of the places where they lose you. For example, the market failure of lenders lending money to people with no ability to pay it back inflated demand for housing, shifter the demand curve to the right, and pushed up prices. Isn't that more or less 'magnifying' the blueprints?

[Nick Rowe]

Patrick: that would indeed magnify the blueprints, and it does make sense (though you still need to explain why so many lenders made the same mistake at once). But it's VERY un-Austrian to invoke market failure!

[Don Lloyd]

Nick,

FWIW -

Pick any normal economy and record all the investments being made.

Add new credit to the economy. Is there anything we can say about the new investments made compared to the ones made in the original economy?

I would claim that the vast majority of the new investments being made are going to be inferior in risk, reward, prudence, and/or anything else that may have prevented them from them being made when less credit was available.

In addition to the lower quality of the new investments, they also have a large negative impact on the original investments as they bid away resources which they will often waste as they fail.

Regards, Don

[Greg Ransom]

Before you continue writing on a subject you don't know, Roger Garrison has some very easy to read introductions to the model here:

http://www.auburn.edu/~garriro/articles.htm

And note well, what you write here is:

"The validity of this Austrian insight does not mean we should follow the rest of the Austrian prescription, and just let the recession take its course."

Hayek discusses things that can be done to block a deflationary downward spiral, and things that can be done to soak up access labor ("public works"), etc.

Hayek's work is always at an advanced level -- none of it serves as a useful introduction to folks steeped in Keynes and neo-classical "capital" theory. Read Garrison. Especially his book, _Time and Money_.

The logic of the time structure of the economy isn't actually that hard -- the best introduction is actually Bohm-Bawerk. But I know that economists don't do "history of economic thought".

[Greg Ransom]

Let's do that again.

What you write here is false:

"The validity of this Austrian insight does not mean we should follow the rest of the Austrian prescription, and just let the recession take its course."

Hayek discusses things that can be done to block a deflationary downward spiral, and things that can be done to soak up access labor ("public works"), etc.

One thing people ignore is that in the very early 1930s Hayek was speaking strictly about a unique British problem that had been ongoing for a decade -- caused in large measure by the return to the gold standard in 1925 at the old pre-war parity, causing massive deflation running up against newly powerful unions, especially in the coal mines. Britain had priced itself out the international coal market. Hayek's "popular" remarks in the 1930s were specifically addressed to the British situation -- and Hayek fully recognized the validity of using inflation to solve the British problem. As Hayek noted, economic theory before Keynes full explained the process. Hayek's well considered view is that Keynes invented bad theory in order to "win" contemporary debates, and changed his "theory" as contemporary debate dictated. Hayek's view is that no Keynes was necessary to make the case for inflation to correct the "sticky wages" and deflation problem in Britain.

[Greg Ransom]

Look.

There's "real analysis" / "pure theory" / non-monetary economics where "prices" are "perfectly flexible". This is a purely logical / mathematical model. A "barter" economy.

Then there is the real world where prices are not "perfectly flexible". When you inject falsely money and credit into specific points in the economy, the time structure of the production economy distorts. As Hayek explained to everyone, there are knowledge problems blocking perfectly flexible prices.

What don't you understand about that?

You write:

"I couldn't make sense of the idea that monetary policy could set (real) interest rates too low in a world of perfectly flexible prices."

[Greg Ransom]

I don't know if it's my bad eyes or what, but sorry for the missing words. Make that:

"When you inject new money and falsely _priced_ credit into specific points in the economy, the time structure of the production economy distorts. As Hayek explained to everyone, there are knowledge problems blocking perfectly flexible prices."

[Ami Ganguli]

Don,

I don't follow. First, my personal (flawed?) understanding of the concept of "money" tells me that you can't really have money without credit. Whenever I accept money as payment I'm effectively granting a loan. Even if the money is backed by some good, I'm taking the money and allowing whoever issued the money continued use of the good - basically a loan.

But even leaving that aside - if we assume the creation of money through some magical mechanism that doesn't involve credit - what's the basis for assuming that new investments due to increased credit would be inferior?

... Ami.

[Edward J. Dodson]

The most insightful analysis of business cycles was provided by Henry George in the late 19th century. His analysis has been verified by economists who ignored their training in neoclassical theory and treated land markets as distinct. Absent the public collection of rent (i.e., location values), land markets do not respond to the price mechanism as do the markets for labor and capital goods. Rising land prices stimulate even more acquisition of land for speculation rather than development. Credit fuels this speculative instinct, driving up the cost of land beyond what businesses can pay and maintain acceptable profit margins. And, as we have seen in the residential markets, rising land prices make it increasingly difficult for households to make down payments, pressuring lenders to provide higher levels of financing (funding the purchase of more land and less housing) to borrowers whose creditworthiness is less and less stable. These stresses bring on the collapse in land markets every 18-20 years.

One economist who is making a major effort to bring this analysis into the public dialogue is British economist Fred Harrison. His books and videos on youtube are essential reading for anyone who seeks answers to our economic problems.

[Patrick]

Why did so many lenders make the same mistake at the same time? Oh man, that's way beyond me, but I'll take a stab at it anyway just to see if my own narrative makes any sense when I read it back to myself.

The story seems to be that securitization and the 'de-personalization' of banking created an agency problem and some very perverse incentives in mortgage lending. At the same time, the ratings agencies were essentially like a captured regulator, and they failed completely, which created a 'bag holder' class who could ultimately hold all the toxic waste created by the aforementioned pipeline of broken lending. Unfortunately, it looks like the ultimate bag holders were the lynch pins in the financial system. Specifically the investment banks and the worlds largest financial institutions like Citi, BoA, etc... Also, there was a massive failure by regulators for a variety of reasons. I think regulatory capture played a role, I think 'starve the beast' free market fundamentalism played a role. Policymakers explicitly eviscerated the regulators to make them incapable of doing anything.

To me, the really hard thing is explaining what got the ball rolling. Securitization etc was not new in 2003/4/5. The catalyst to start the ball rolling seems to have been the monetary policy that was used to try to recover from the tech bust. I think it was Greenspan's promise to keep rates low for a long time that really created the problem. I also think that the worlds savings glut, Chimerica, and other macroeconomic imbalances played a big role, but stitching the macro story together into a coherent narrative is beyond me. No doubt many a tome will be written by better minds than mine trying to piece it all together.

[Nick Rowe]

Don: What you say about marginal investments being of lower quality than infra-marginal investments makes sense to me (trying to stay clear of Cambridge-Cambridge debates). But if expectations were on average correct, an increase in investment due to an increase in voluntary savings, even if temporary, should not result in aggregate in an increase in malinvestment or half-finished buildings, or workers abandoning their training and regretting their career choices. I think we need some mechanism whereby expectations about the future returns to investment are systematically distorted (or at least wrong) to make the story work.

Greg: I will give Roger Garrison a try.

I'm fairly sure I read Bohm-Bawerk 30 years ago (and many others, so I'm not getting it all second- or third-hand), but I may have forgotten. The time-structure of production made sense to me (though of course it's not as easy as Y=F(K,L)!). Menger and Mises on money seemed to make sense as well. It was just putting the monetary theory together with the capital theory where they lost me.

Now, you say: "As Hayek explained to everyone, there are knowledge problems blocking perfectly flexible prices." Is this a reference to "The use of knowledge in society"? Because imperfect price flexibility is not what I got out of that essay (though I did get a lot out of it). Should I re-read that essay? Or something else?

I was interested to read what you said above about Hayek, Britain in the post-WW1 return to gold, not advocating doing nothing, etc. What's current Austrian thinking on the best policy for the current problems? (As opposed to how we got into the mess in the first place.)

[Nick Rowe]

Patrick: that sounds about as good as anything else I've heard!

[Stephen Gordon]

The thing is, I don't see how this would apply to Canada. The US, yes, with the housing bubble. But not here. The wave of investment we had was in the tar sands, and although it's fallen off, I don't see it as being left to rot. Those projects will be profitable when oil prices start going up again.

[Patrick]

What about car loans and consumer credit? I'd be interested to know more about where this kind of debt is concentrated. Anecdotal evidence suggests that there are an awful lot of 20-something year olds driving $60K+ pick-up trucks and watching monster flat screen TVs.

[reason]

I also don't understand why investments couldn't also turn out to be better than expected. It would seem to me that lower interest rates might actually encourage safer investments with lower rates of return. It seems Austrian theory assumes businessmen are myopic - but these are the heroes of the economy. God help the rest of us.

It also is (from the point of view of WHAT ACTUALLY HAPPENED) clear that the problem was not so much borrowing for investment (and most of that was secondary) as borrowing for consumption or speculation that was the problem. Read Steve Keen for a better story (and one that is much easier to understand):
http://www.debtdeflation.com/blogs/2008/11/02/debtwatch-no-28-november-2008-what-is-really-going-on/#comments

[reason]

Personally,
I think the problem that economic theorists of all stripes have in understand the business cycle is that don't distinguish strongly enough between the financial and the real economy and don't understand the importance of balance sheets and the non-linearity that arises because of bankrupcy.
Hence my comment the other day about risk of default - Nick - and I think I'm still waiting for an answer on that one.

[reason]

To take another track and this may help Nick out with his questions as to what has happened in Canada, the real problem is not so much what the monetary authorities in the US have done, but what China and the oil exporters have done, in pumping money back into the US despite low yields. The international financial system simply failed to find a way to correct a massive and sustained imbalance in the circular flow of money.

The micro problems that the Austrians like to point are small potatoes, the real (and massive) problem can only be seen at the macro level.

[reason]

(And yes I know, if the Fed had deliberated generated a recession earlier in order to adjust the balance of payments the situation wouldn't have developed as it did. But come now, what does the Fed see as its job?)

[reason]

I'm always amused as well when Austrian's when really pressed point out that Hayek was not totally against social security (of some sorts at least) or public works in the case of severe recessions. So is Hayek a closet Keynesian? Is Hayek a real Austrian?

[Nick Rowe]

Greg: I gave Roger Garrison a try. He writes well, and very clearly. He is easier to follow than Hayek or Mises. I understand what he is saying, but I still don't understand the ABCT. It's not what he says; it's what he doesn't say. Even leaving aside the truly Austrian elements (the Hayekian "triangle" for example), I do not understand even the canonical "British inter-war" (Robertson/Hawtrey etc.) loanable finds diagram with the central bank pushing down the rate of interest and creating forced savings. Or rather, I understand the diagram, but I don't understand what's going on *behind* the diagram. Are prices fixed or flexible? What is happening over time to the stocks of money and bonds? What is happening to output and employment? Prices and wages? Expectations of prices and wages? In short, what I need is a complete general (dis)equilibrium model to show me everything that's happening. That model can be as simple, crude, ad-hoc, unrealistic, etc. as you like, as long as it is complete, so that everything implicit is made explicit, and I can look at it from all sides, wind it up, and watch it cycle along the floor. (Once it runs, however badly, we can always improve it later). I have tried to build such a toy model of the ABCT myself, but have failed. If I assume flexible prices, I get no effect of monetary policy on real interest rates; if I assume sticky prices, I end up with a Keynesian model (where the *level* of the money supply affects real interest rate, not the rate of increase of the money supply).

Stephen: If I saw a half-finished tar-sands project, on which work had (temporarily) stopped, I would see that as a great example of the Austrian parable. Even if works slows down, it looks a bit the same. The typical neoclassical vision of capital has all the inputs going in at once, then a steady flow of output over time (diminishing slowly with depreciation) as you continue to add labour. Capital is like land, except you can make more of it. The tar-sands investments take many years to come to fruition, just like the simple Hayekian triangle, then you don't need any (or as much) labour when they start producing.

Patrick: I could be wrong on this, but I don't think consumer credit for consumer durables is as central to the Austrian vision. (Greg would know better). Not saying it's incompatible with Austrian capital theory, because it isn't, but I don't think it's central to their story of the business cycle.

reason: I haven't forgotten about default; I just can't think of anything new or interesting to say. I may just post something old and boring anyway. How about: "Default does not reduce aggregate wealth. (Well, maybe it does, but only in second-round effects)"? Would that work?

Yes, let me work on it. Then we'll see how it fits in with the rest of what you say.

[reason]

Nick
"default does not reduce aggregate wealth" sounds like you are ignoring my bit about the difference between the real and financial economies. Are you assuming instanteous price adjustments and bankrupcy procedures?

[reason]

i.e. Nick,
yes no real capital has disappeared, but some parts of the economy will be badly wounded and will change their behaviour drastically. It is like saying that theft has no impact on the economy - the ownership of goods has just changed hands.

[Nick Rowe]

I think the theft analogy is very good. See what you think down at the end of my new post.

[Greg Ransom]

Hayek's first essay on the division of knowledge theme "Economics and Knowledge" (1937) is actually better.

Hayek saw two tasks: explain how economic order is possible, and what coordination looks like, and explain why economic disorder occurs.

The pure theory of the "barter" economy (i.e. equilibrium theory, etc.) lets us "see" what it means for economics plans to be coordinated (see Hayek's discussion on this stuff in the first few chapters of _The Pure Theory of Capital_.) There's no knowledge problem when all "knowledge" is given, as it is in a pure logic like an equilibrium construction.

The money-capital connection was explained by Hume and Cantillon.

>>Now, you say: "As Hayek explained to everyone, there are knowledge problems blocking perfectly flexible prices." Is this a reference to "The use of knowledge in society"? Because imperfect price flexibility is not what I got out of that essay (though I did get a lot out of it). Should I re-read that essay? Or something else?<<

[Greg Ransom]

Austrians have different views depending on their understanding of the empirical facts on the ground.

Are we in a deflationary downward spiral or not? I've seen Austrian support for stopping the deflationary downward spiral, but the difference here seems to be how people understand the data on the ground. I don't think many have enough information to have "definitive" view on the matter.

As far as I can't tell, most don't support the bailout, because they see it as causing "regime uncertainty".

The work of Robert Higgs seems to be more influential than anything right now.

"What's current Austrian thinking on the best policy for the current problems? (As opposed to how we got into the mess in the first place.)"

The regulatory regime is a mess. I'd advocate fixing it.

Some "Austrians" are radical anarchists or hard core libertarians as much as they are economists, so I don't know how they compute the problems in regulation.

Others have said the issue isn't how much regulation, it's good regulation vs bad regulation.

Moral hazard and the privatization of profits and the socialization of losses are big issues among Austrians. But I don't know may Austrian economists who are specialists in banking regulations and financial regulation.

[Greg Ransom]

Is there any way to put into mathematics "slow and distorted learning" and movements in time structure of production through the relative price structure?

These pure mathematical constructions with perfectly flexible prices and sticky prices falsify the phenomena of the coordination of relative prices through a capital using economy, with money and credit. It's a simple as that.

You wrote:

"Even leaving aside the truly Austrian elements (the Hayekian "triangle" for example), I do not understand even the canonical "British inter-war" (Robertson/Hawtrey etc.) loanable finds diagram with the central bank pushing down the rate of interest and creating forced savings. Or rather, I understand the diagram, but I don't understand what's going on *behind* the diagram. Are prices fixed or flexible? What is happening over time to the stocks of money and bonds? What is happening to output and employment? Prices and wages? Expectations of prices and wages? In short, what I need is a complete general (dis)equilibrium model to show me everything that's happening. That model can be as simple, crude, ad-hoc, unrealistic, etc. as you like, as long as it is complete, so that everything implicit is made explicit, and I can look at it from all sides, wind it up, and watch it cycle along the floor. (Once it runs, however badly, we can always improve it later). I have tried to build such a toy model of the ABCT myself, but have failed. If I assume flexible prices, I get no effect of monetary policy on real interest rates; if I assume sticky prices, I end up with a Keynesian model (where the *level* of the money supply affects real interest rate, not the rate of increase of the money supply)."

[jp]

The only problem with your description Patrick (a mishmash of securitization, bad regulation, low interest rates, market fundamentalism) is why this potent concoction primarily affected mortgage backed securities markets and not credit card backed securities markets. Credit card backed securities have been very stable through 2007 and 2008, yet are the same security type as mbs (the securitized type) and face similiar conditions to mbs (low rates, lack of regulation).

Because credit card backed securities have done fine, I think blaming securitization is largely a red herring. The real reason mbs became a bubble and not credit card backed securities is because the latter market was never subsidized by a massive government sponsored agency, whereas the former was dominated by 3 of them; Fannie, Ginnie, and Freddie.

JP Koning

[Nick Rowe]

JP: You raise an interesting puzzle there with the credit card-backed securities. But I'm not sure that Fannie Freddie and Ginnie explain the difference, since there were house price bubbles (and crashes) in many other countries. I'm wary of any US-based explanation of what seems to be a global phenomenon.

[reason]

JP should maybe wait a while. But just out of interest what is the relative size of the two markets?

[jp]

It's hard to find current data, but this paper (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=927489) from 2002 mentions:

"As of June 2002, the ABS market was made up of $6.6 trillion in tradable securities. About 70 percent of these assets, or $4.5 trillion, were mortgage-based, including
mortgage-backed securities, collateralized mortgage obligations, and real estate mortgage
investment conduits. Government-sponsored secondary market lenders, such as Fannie Mae,
Ginnie Mae, and Freddie Mac, issue the majority of MBS. Over $1.4 trillion, or about one-fifth of
the securitized asset market, is composed of asset-backed securities that are not collateralized by first mortgage assets. The underlying assets of these securities include student loans, mobile
home loans, vehicle loans, home equity loans, and credit cards. Approximately 30 percent, or
$400 billion, of this $1.4 trillion market is composed of credit card loans."

Most countries have housing agencies that subsidize mortgage finance (CMHC in Canada) or government created mortgage finance structures (German pfandbriefe). Fannie and Freddie were only the most aggressive. Also, many countries began allowing their central banks to accept MBS in open market operations in the 90s, encouraging market growth (1998 Spain, Canada and US both in 1999).

Very few if countries have agencies or laws that subsidized credit card financing and/or bought ABS, thereby engendering moral hazard. Canada and US, for instance, only allowed ABS to be accepted in central bank open market operations in 2008. That they were never coddled by the powers that be could explain their relative healthiness. Like any asset, in a recession credit card ABS will be affected but they aren't at the centre of the storm like housing MBS, nor is securitisation the evil it is often made out to be.

[reason]

jp
of course you may be right, but it could also be that 400 billion is small potatoes and the MBS were the first shoe to drop and cause a big splash. The pile of debt had to blow up somewhere. How much of credit card debt is securitised that way (i.e. what proportion of the potential have been palmed off on somebody else)?

[jp]

From http://www.sifma.org/research/pdf/ABS_Outstanding.pdf and http://www.federalreserve.gov/releases/g19/Current/ about 35% of the total credit card debt is securitized.