June 2022

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30    
Blog powered by Typepad

« A Terminological Suggestion | Main | Creative Destruction - Two Tales in Four Pictures »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Before I got into economics I spent 15 years in public relations. PR operates under the assumption that people adopt ideas for emotional reasons and then search for logical backup. Some research has backed this up. That's why data changes the minds of very few in fields like economics. Cowen's offer of data as proof against the ABCT is a good example. Everyone interprets the data from their own theoretical background.

It's good to respond to critics of the ABCT, but don't expect to change any minds. The responses are necessary to help those who follow the ABCT to better understand it.

A lot of people pointed out the logical flaws in Keynes, but they didn't stop it sweeping the world.

A little clarification about my post above. We owe it to ourselves to discover the truth. Most people choice their ideas for emotional reasons but we don't have to be among those. Truth should be the most important thing to us. If Cowen is right and the data contradict the ABCT then we should abandon it as he did.

"And keep in mind that in the standard textbook presentation of the Mises-Hayek story we do not see comovement, but the distortion of the structure of production, which is then corrected during the bust phase."

I thought the Mises-Hayek story said that credit induced boom will cause both investment and consumption to rise at the same time. And I thought Garrison and Murphy had good responses. What did I miss?

I don't get it either. The comovement argument that he raised in his book over ten years ago has already been answered several times to my knowledge. I would understand the whole fuzz if it was actually a very recent criticism, but it is out there for years. And that ignorant people like Krugman and DeLong prefer to rehash again and again the same arguments against ABCT (even though their arguments have been tackled as well) and accuse everyone who is not convinced by their rhetorics of being right-wing barbarious economists, etc. is not actually news either imho.

I picked up on Cowen's post here this morning: http://factsandotherstubbornthings.blogspot.com/2011/01/cowen-murphy-and-hayek.html

I'm not sure if I'm following the arguments as closely as I could, but it's a substantive attempt.

Jonathan Catalan has made similar arguments to Garrison in the past about how these comovements occur as well, but he could probably point you to his best expositions.

One thing about Garrison to keep in mind is that his PPFs are not like most people's PPFs. He talks about them in terms of sustainable production frontiers, not technological production frontiers. I know you don't want to veer into terminology, but this is one little piece of terminology that makes a big difference in what kind of answer Garrison can provide.

There's another really obvious way to have ABCT and co-movement, and that's to assume that we have an effective demand problem and that we are not at full employment. That doesn't invalidate ABCT at all, and in fact it can provide the answer to why we see co-movement in the data. I think a lot of Austrians are loathe to use this explanation, because as soon as they admit we have an effective demand problem they have (1.) capitulated substantial theoretical ground to Keynesians, and (2.) introduced a problem that is potentially considerably more serious than distortions of the capital structure.

This is roughly where I fall out with ABCT. I'm pretty confident distortions of the capital structure do exist - not in an overinvestment sense, but in the sense that artificially low interest rates could certainly make production more roundabout than would be natural or sustainable. This would, certainly, open the door to a bust. It just doesn't seem to me to offer the most pressing problem to be concerned with. A rebalancing of the capital structure doesn't strike me as being as substantial a problem as persitent underemployment of resources.

The first time Coordination Problem took notice of me (that I'm aware of) was precisely when I said this before and proposed an "Austrian-Keynesian synthesis". There is absolutely nothing about "full employment by accident" and a liquidity preference theory of interest rates that eliminates the Austrian point about the relationship between the interest rate and the roundaboutness of production. There is absolutely no reason why these two phenomena cannot coexist. I think the issue is, a lot of people don't want them to coexist.

Another obstacle - as Peter has alluded to - is that non-chalance towards empirical work that emerges out of the Austrian epistemological disposition is troubling for a lot of non-Austrians (its troubling for me). Logically I have no reason to think ABCT dynamics don't go on in a "full employment by accident" world. I'm a little skeptical still about how substantial those dynamics are.

Daniel, I see Austrians taking data with a grain of salt while others swear full allegiance to the danger and then ignore it. In other words, I don't see any more reverence for the data among other economics. I hear a lot of lip service, but when the data contradicts them they ignore it or explain it away. I think non-Austrians are very hypocritical in their allegiance to data. As I wrote before, I have never seen data settle an argument between economists of any stripe.

I agree whole-heartedly with Pete's post. What are the empirical implications of the theory? We have some idea already but we need more precision.

One reason I like the Cowen "revision" of the ABCT is that it brings risk into the picture more explicitly than the traditional theory.

But we must not forget Garrison's point that the ABCT is a theory about the boom and the upper turning point only. And I would add it does not (or should not) make the claim that it is the only explanmation of recessions.

The task is: (1) Make the empirical implications of the theory more precise -- or as precise as the theory warrants and (2) Do not oversell the theory's power.

And take two aspirin (or a xanax). The world won't collapse if Hayek or Mises made errors with regard to business cycle theory.

And read THE ECONOMICS OF TIME AND IGNORANCE so that you can see the bigger picture.

PSS, honest defense of the ABCT is for young economists who can still change their minds and who care about the truth. There is nothing that anyone can do, and no amount of evidence that anyone can offer to change Krugman, Caplan or Cowen's minds. After all, Keynes didn't change the minds of many older economists; young economists created the Keynesian revolution.

Roger - really?

- What about early Keynesians who revised their multiplier expectations considerably downward when the data came in?

- What about the flurry of activity to revise theories incorporating expectations when the data on inflation/unemployment trade-offs came in (this work started earlier, of course, but was truly embraced when the data came in).

- What about the subsequent dusting off of old ideas and the abandonment of exclusive devotion to sticky wages when the data on the Japanese crisis came in?

- What about the JTPA evaluations that resulted in a huge change in the way we think about job training?

- What about the total sea-change in the response of liberal economists to welfare reform after numerous empirical studies on that came in? Fifteen years ago they were crying bloody murder. Now they're fine with it - and just interested in patching up a few things and addressing needs of some targeted populations. Why? Because the doomsday they were predicting didn't pan out empirically.

Your comment strikes me as far too strenuous and far too short on examples. Certainly we do not have many controlled experiments which means that its much harder to have conclusive statements from data. And in the absence of completely conclusive statements, a couple competing theories can be corroborated by the same set of data. And when that happens, sure, disagreements will persist. But that's not the same as being hypocritical in the use of data.

Hayek is pretty clear in Prices & Production on how there can be co-movements, IMO. Both Mises and Hayek assumed there was at least no change in consumption. There's no reason why consumption couldn't increase during an artificial boom, especially with the advent of consumer credit. It just means a heightening of the Hayekian triangle, and a lesser lengthening (or, it could just as well lengthen to the same degree, but the degree at which part of that investment [see Hayek's graph: http://images.mises.org/4924/Figure1.png] represents malinvestment, due to the scarcity of capital goods, is greater).

Comovement of consumption and investment is not an "embarrassing" fact that lessens the credibility of the ABCT. It's a possible way that the structure of production might change.

It might cause greater widening, rather than lengthening, since consumer good prices would go up, and therefore so would investment in second order goods, so and and so forth. There is more of an incentive to invest more new money into each stage of production, rather than a tendency for the rate of profit in each stage to fall sooner, and therefore cause non-specific factors of production to move to earlier stages of production.

Roger - another example - you don't think empirical findings have contributed to the change in the way development economists view foreign aid? Again - the perspective is completely different from what it was several decades ago, and a big reason for that is because the data came in.

More of a popular post that isn't only about ABCT, but maybe some people might find it interesting: http://mises.org/daily/4924

Like Mario, "I agree whole-heartedly with Pete's post." Pete is pointing out that we need to be serious about an *empirical* research program in macroeconomics. We have a few papers and some great historical work. We do not have a serious empirical research program, period. Forget about whether DeLong and Krugman are cloven-hooved demons. ABCT is not in the list of currently serious options to DSGE. It will not make the list until we start delivering fact-based research on cycles and crises. It is fine and dandy to point out the gap between your theoretical categories and the available data. But then you gotta close that gap in a sensible way. You gotta close that gap in a way that lets us have a serious conversation about whether your theory or mine fits the facts better.

This. is. the. moment. The Great Recession dealt a powerful blow to DSGE. Richard Caballero cites Hayek to say that DSGE neglects complexity (“Macroeconomics after the Crisis: Time to Deal with the Pretense-of-Knowledge Syndrome,” Journal of Economic Perspectives, 2010, 24(4): 85-102). Radical uncertainty and the state of confidence are back on the radar screen of mainstream macroeconomists. And so on. The ground has been laid for an Austrian macroeconomics that the mainstream can understand and appreciate. Austrians can get in the game now, but only if they get serious about facts. That requires immersion in the mainstream literature on what measured variables do and do not move pro-cyclically. And it means re-crafting our theory to make it testable with the currently available numbers. Milton Friedman’s theory of the consumption function was great precisely because it showed us how to craft theory to facilitate empirics. Austrian macro still has not caught up with that change in economic method. It’s time to catch up and get serious.

Great post! Great challenge! Perfect attitude!

Roger -

re: "Radical uncertainty and the state of confidence are back on the radar screen of mainstream macroeconomists. And so on. The ground has been laid for an Austrian macroeconomics that the mainstream can understand and appreciate."

And it's also important to note that the fact that radical uncertainty and confidence are back IS NOT just an invitation to Austrian economics as well. It's also an invitation to Keynesians who think New Keynesianism is largely Pigovianism, and a few other camps. If Austrians don't answer the call, other people are HAPPY to answer it.

I'm not one that wants to see ABCT crowd out these other uncertainty and confidence based arguments. But I don't want to see ABCT crowded out either. The point is - nobody should make the mistake of saying "oh - they recognize the importance of radical uncertainty so Austrianism will come naturally". It won't come naturally. It has to be argued.

On empirical Austrian papers - does anyone know of a good survey article? I know of about a dozen of these papers, but I'm not aware of any good survey.

I sometimes raise some point I consider open about ABCT, namely the passage from monetary injection to long term interest rates to investment discoordination...

The possibility of overproduction is surely not one of the things that keep me awaken at night because of the urge to find a solution.

Let's consider that Hayek, although he explicitly originated the absurdity that consumption falls during the boom because of forced savings, provided some theory of overproduction. That Mises provided some by the theory of capital consumption, although merely in nominal instead of real "structural" terms. That Machlup wrote a paper touching these subject in his analysis of forced savings. And that one of the last words has been given by Garrison, making a great job in clarifying the issue. I'm not sure Cowen is aware of this, because he never discusses these papers.

Anyway, the standard Austrian view is that comovement is due to capital consumption. Capital consumption may arise because:

1. Labor is procyclical (Garrison)
2. Durable capital goods can be produced and contribute to production although it may be impossible in the long run to renew them because of excessive consumption.
3. Durable capital goods can be used in order to increase short-term production at the expense of durability (Hayek).
4. Durable capital goods can be produced and goods can be consumed at the expense of intermediate productive passages (pulling the triangle at both ends): I consider this not as a theory but as a metaphor, however.

I'm dissatisfied with the nominal view which sees capital consumption only as a nominal illusion due to illusory wealth, and unfortunately there's not much in Mises besides his intuition that the forced savings story of Hayek was wrong.

I do believe, in addition, that capital consumption requires to investigate the time dimension of durability instead of the time dimension of production time. It is not true that Hayekian triangles can be used to explain this issue, despite Garrison's opinion.

Months ago I proposed a model of overproduction based on a very simple capital structure, I'd call it an Austrian production function.

Let's assume that only labor produces capital goods and final goods. There's a durable capital good produced by labor and a final good produced by labor and the durable capital good. Investment (labor) in durable capital I call "X fund" and investment in final goods I call "Y fund". The economy needs to decide how to allocate the labor supply L between X and Y.

Durable capital is produced instantaneously by labor and lasts N periods. Final goods are produced instantaneously.

In time 1 a labor supply shock increases X. In time 1 the supply of capital goods is increased, up to time N. From time 1 to time N, the supply of final goods is increased. From N+1 on, it all depends whether X is maintained or not, but a transitory shock can increase productivity.

Let's now assume that investment in the X fund is nil. Capital goods will decrease by 1/Nth for period. Fund Y is maximized and the production of final goods can be greatly expanded. No one will notice until durable capital becomes too old.

In this economy it is possible to have a production overshooting which will turn out to be self-reverting once savings will become insufficient, and there is a built-in lag structure.

Daniel, I would classify your examples as trivial changes within a paradigm, not settling of major disputes such as the Keynes/Hayek debate. Yes, there have been major changes in development econ, but it took generations. And the numbers haven’t changed the minds of people like Jeffrey Sachs and many others. But data isn’t the deciding factor. PR research has demonstrated that data is far less important in changing people’s minds than something like an appeal to authority. Data reinforces, but it doesn’t cause people to change their minds.

Johnathan, good point! Here is a relevant quote: “Now, contrary to what we have found to be the case when similar processes are initiated by the investment of new savings, this application of the original means of production and non-specific intermediate products
to longer processes of production will be effected without any preceding reduction of consumption. Indeed, for a time, consumption may even go on at an unchanged rate after the more roundabout processes have actually started, because the goods which have already advanced to the lower stages of production, being of a highly specific character, will continue to come forward for some little time.” Page 88 PRICES AND PRODUCTION

I definitely agree that Austrians ought to provide more data. Even if it data doesn’t change people’s minds, it reinforces the logic. Unemployment rates in capital goods vs consumer goods industries is good data and supports the ABCT. Investment professionals fixate on cyclical and defensive stocks. What are those if not capital goods producers and consumer goods producers and retail, which is empirical evidence for the ABCT.

My take on other theories of business cycles is that they are all true at some point in the cycle. The ABCT provides the skeleton while others flesh it out.

let me just emphasize how deep Peter's challenge roots. after all, the major prediction of the ABCT is that a money-induced traverse always and by necessity reverses. at the root of this statement is the notion of forced saving. Now, forced saving is about increasing investment fed by falling consumption. So comovement of I and C is very hard to attune with forced saving which, in turn, is at the heart of the ABCT. I have no idea how to bring data into accord with this theory.

Roger -
RE: "Unemployment rates in capital goods vs consumer goods industries is good data and supports the ABCT."

But this is an example of my point that you're misreading responses to data. This supports standard Keynesian theories too. It corroborates both. You can't use it to arbitrate, so you can't be surprised when nobody changes their position on this basis.

In college I had an econ prof who had a PhD in physics and a PhD in economics. He developed a new macro model and went to DC to sell it to someone in government during one summer break. That fall he told us about his failure and said that economists need to know more about marketing than one can get by reading a book on it during the flight.

I have often thought about that story. Economist don’t know anything about selling or persuading others to change their minds about things. PR, marketing and advertising research has a lot of really good material on how to do that. Yes, it’s important to respond to critics and provide data, but don’t expect it to change minds.

"One thing about Garrison to keep in mind is that his PPFs are not like most people's PPFs. He talks about them in terms of sustainable production frontiers, not technological production frontiers."

Important point. In our dot-com boom-and-bust paper, Roger and I spent time giving examples of how temporary movement outside the PPF actually described real phenomena in that episode quite nicely, e.g., the Java programmer who worked 100-hour weeks for a startup dot-com because he planned on becoming a millionaire when the company went public. Then the crash came, and his plans were revealed as vain. The Garrisonian boom involves both malinvestment and overproduction, in the sense Malthus originally formulated the idea, that ex post, the 100-hour weeks will be judged to have been not worthwhile.

Isn't the co-movement issue only relevant at the peak? We should expect consumption to grow more quickly, pulling resources from investment. And that doesn't seem to be happening in real world turning points? Investment just falls, and then consumption slows. Something like that.

Catlans, I don't get it. Consumer credit? Yes, but what about scarcity? If you can produce more of everything, then you are saying there is no resource constraint.

Of course, just about every theory of "boom" is that the resource constraint is loose, and rapid growth in spending can temporarily raise output. That was my understanding of Garrison.

The orthodox position is that monopolistic compeition with preset prices makes expanding output when demand rises more than expected profitable. But the effect won't persist, because resetting prices at a higher level and producing less is even more profitable. There is excess capacity always.

Tie that into this catillion effect, and more of everything can be produced, but goods with more interest elastic demands would rise more than in proportion than goods with less interest elastic demands. But the problem really isn't with the extra output being unbalanced. All it does is explain why some areas of the economy expand more in the boom and contract more as the boom ends.

And, of course, in a growing economy, consumption and investment both can and usually do increase at the same time. The theory would be that in the boom investment grows too quickly. The scarcity constraint would say that consumption grows too slowly in the boom and that the investment ends up unprofitable because consumption demand will start to grow more quickly at the peak and move beyond the growing capacity to produce consumer goods because the investment was the production of the wrong capital goods that will only be able to expand the production of consumer goods even more in the more distant future. (This repeats my first statement about the turning point.)

Also, those of us who have known Cowen for years, know that he could spout off with conviction everything Rothbard said about the Business Cycle (and maybe Mises and Hayek too) at 16.

He went from understanding the theory better than most reading this (and I count myself) and accepting it, to rejecting it. I think Caplan had as good a grasp of the theory as most Austrian economists, and then rejected it.

Pete may disagree, but I would count him in too as someone who can't really claim that Cowen has a less thorough understanding than himself. It has to do with specialization. Garrison, Selgin, White, Murphy, Salerno, maybe not. Some of them may say that Cowen's understanding (like mine or Pete's) was not sufficiently deep. But when I read most people explaining the Austrian theory in comments on this blog, I am like--they are repeating stuff that I read a long time ago. Things that I used to treat as gospel. And now, not so much.

Daniel, I don't expect it to arbitrate between the two, but it does support ABCT. However, ABCT being accurate doesn't require that everything in the Keynesian model be wrong. There is overlap.

Bill, I just don't agree. I'm nowhere near as well-read or as intelligent as Cowen or Caplan, but I have read their critiques and was dumbfounded at how wrong they got the ABCT. In my opinion they don't get it.

Roger -
You're preaching to the choir - I agree that there is overlap. My only caution is that you hesitate on the efficacy of empirical work because you think people take it hypocritically. Rather than accuse them of being hypocrites, it's worth distinguishing between "empirical work that corroborates ABCT but won't convince a skeptic" and "empirical work that proves ABCT and will convince a skeptic".

Part of the difficulty, may I suggest, is that Hayek's particular version of the theory in "Prices and Production" is taken as the only benchmark for the theory.

Indeed, in my opinion, it is this version that Roger Garrison considered himself called upon to "defend" against the type of "attack" leveled against the theory (like the one in John Hicks' well-known essay on "The Hayek Story").

That is, that the rate at which any additional ("new") money will circulate through the economy after the initial "injection" would be too rapid to likely allow the type of sectoral shifts in capital and labor misallocations that result in "real" distortions in the structure of production that is at the center of the argument in "Prices and Production." Thus, while there might be some general "distributional" effects, the wider impact on prices in general would occur more quickly than the slow lags that Hicks argued would be needed to have all have the "real" effects in Hayek's "story."

That is, the time lag between the inject of money and its relatively full effects on prices in general is too short for there to be the capital and labor re-directions of resources that is the heart of Hayek's presentation.

Now, Hayek tried to defend his theory in his 1969 article on "Three Elucidations of the Ricardo Effect," in the pages of the JPE.

Roger Garrison's response is to use the production possibility frontier to argue that the artificially lowered rate of interest, due to the monetary expansion, both (more or less simultaneously) increases the quantity demanded for loanable funds by borrowers and decreases the quantity supply of savings by income earners. Thus, there results a more or less simultaneous "push" on the demand for investment goods and consumer goods.

But, given scarcity, how does the economy accommodate an attempt to be at two conflicting points along the given production possibility frontier at the same time? Well, Roger's answer is a variation of the standard macro distinction between the short-run and the long-run "aggregate supply curves."

That is, the economy can temporarily go beyond its long-run production capability (that is, operating "outside" the production possibility frontier), but there will be forces bringing the economy back to it's long-run production potential.

However, if one reads Hayek's "Monetary Theory of the Trade Cycle" and the chapter in Mises' "Human Action" on 'Interest, Credit Expansion, and the Trade Cycle,' there is another explanation of the the potential for both "malinvestment" and "over-consumption."

Hayek was concerned with explaining how an increase in the quantity of money and credit to maintain a "stable" price level in a growing economy could be "destabilizing" because it requires lowering the rate of interest (under the assumption that the new money is injected via the banking system) below that rate that is consistent with a balance between real savings and investment.

Now the key phrase here is a "growing economy," that is, the production possibility frontier is moving "out." It is not a "given" production possibility frontier. Now the point along a higher, to the "right," production possibility frontier to which the system would be moving, given real savings, investment, and the resulting lower general level of prices (due to the output and productivity improvements), is a different path and point than the one resulting from the monetary induced lowering of the market rate of interest.

Thus, the economy is growing, with the potential for both more consumer goods and investment goods, but the interest rate manipulation gives the combination between these two a "wrong twist," in terms of a point along that hypothetical higher production possibility frontier than the one towards which the economy would be moving, if not for the monetary expansion.

Given the "real" gross savings, the economy over the next "period" would be able to, say, produce 100 units of more "stuff." And if not for monetary manipulated, the appropriate "equilibrium" amounts of this additional "stuff" at the end of the "period" would, say, 60 more units of consumer goods and 40 more units of capital goods.

But the interest rate distortion results in producers in the consumer goods and investment goods sectors attempting to produce, respectively, 70 more units of consumer goods and 50 more units of capital goods.

It is the attempt to be moving toward two incompatible points on that "higher" production possibility frontier that results in, as the old song says, "Something's gotta give."

And both consumption and investment spending "fall back" after the upper turning point of the cycle, because of the discovering of the "wrong twist" that has resulted in this growing economy.

This also explains that while there may be a fall in GDP during the downturn, it rarely declines to the level before this process began, because the misdirection of resources has been overlaid on a really growing economy, and with some real, sustainable increases to GDP not "wiped" by the collapse of the "boom."

If you read Mises' account in a part of that chapter in "Human Action," that assumption of a growing economy, and how that results in the potential for both "over-consumption" and "malinvestment" during the boom phase of the cycle is fairly clear.

Thus, Roger Garrison's account is a logically consistent version (given that assumption about short-run and long-run aggregate supply relationships), but it is NOT the "story" that Hayek is giving in "Monetary Theory and the Trade Cycle" and Mises in a part of "Human Action."

Richard Ebeling

Arash, you make a good point. However, as the Hayek quote shows the ABCT anticipates increased investment and consumption at first at the start of the boom. This leads to higher prices and forced savings, but only in the latter stages and just before the bust. Higher prices lead to greater profits in the consumer goods industries and cause the Ricardo Effect to kick in. Investment switches from capital goods to consumer goods and that causes the bust in capital goods. So if investment is aggregated as just “investment” you won’t notice the switch from capital to consumer goods investment.

Cowen does make a good point that at a hypothetical equilibrium it would be hard to have both increased investment and consumption at the same time. Equilibrium assumes full employment and full resource usage. Credit expansion at that stage would be quickly reversed. But Cowen should show us in his data that he thinks is embarrassing to the ABCT at what point equilibrium occurred. Of course, he can’t. Or does he assume that the economy is in equilibrium at every data point? The obvious explanation is that the economy reaches full employment and resource usage only late in the expansion, just before the bust.

Daniel, actually I'm basing my claims on what I know from PR and marketing research in a former life.

"Equilibrium assumes full employment and full resource usage."

Not all equilibrium theories assume this...

I agree with Jonathan earlier - this comovement is not as embarassing as Cowen thinks it is. But for it not to be embarassing, Austrians oughta think more seriously about underemployment equilibria (and some do, to be sure).

The answer is more empirical papers. Where are they?

"What is the best responses to Krugman, Cowen, etc."

Just note. The risk element was always there. So Cowen's "revision" was really much of a "revision".

Mario writes:

"One reason I like the Cowen "revision" of the ABCT is that it brings risk into the picture more explicitly than the traditional theory."

Cowen has misled folks claiming that risk wasn't explicitly there in Hayek, et al. Cowen can explain why he did that.

Make that:

"So Cowen's "revision" was really not much of a "revision"."

In Time and Money, at least, it clearly shows consumption going up during the boom due to the artificially low interest rates. What happens during the boom is too much consumption and too much spending on the most roundabout methods of production and too little spending on everything in between. During a recovery, you need to have less of the first two and more of the latter. I don't really understand why the comovement is such a cutting attach against ABCT. It shows more than anything that Krugman and Cowen are unable to think in terms of anything but aggregates like "investment" and "consumption."

Prof. Boettke- I also don't completely understand the call for more empirical papers. By "empirical," do you mean what the mainstream would call "history"? Or do you mean econometrics? For the former, there are already published papers applying ABCT to basically every crisis in American history. These types of papers are exactly NOT what has convinced the mainstream about the validity of the theory and I don't see why applying the theory to each and every crisis across the world will change minds. And if you mean econometrics, what data should we use? The type of "empirical" paper that would really get a New Keynesian to question their worldview is to throw a bunch of variables (some "Austrian" and some "Keynesian") into a regression and seeing which are significant. "Austrian" data has never been collected, and due to subjectivism quite possibly can never be collected, in such a way that the theory can be tested in a neoclassical sense.

Ryan Murphy,

As an econometrician who has an interest in Austrian economics, there is a degree to which we can approach testing the hypotheses laid out by ABCT. Also, as an econometrician, I know econometrics can never say "your hypothesis is correct." The best we can do with econometrics is to say "we fail to reject your hypothesis."

The correct approach, and what I think Prof. Boettke is getting at, for an econometric investigation of ABCT is to go through ABCT in great detail, including all the variations on the theme, and pull out every possible empirical claim one can. Furthermore, one must identify which of the claims are inconsistent with other theories of the business cycle. You also must be careful to figure out the multiplicity of feedback and confounding effects (this will be the hardest part) that can cause the results of the model to be wrong because of incorrect specification. Then you can develop a battery of testable hypotheses and run the appropriate econometrics on them and see what does and doesn't stick.

The most difficult part aside for empirical identification is dealing with comparative institutions over time and across sections (i.e. countries). This is where econometrics will fail.

I think it's important to remember, too, that Mises, Menger, and others were creating this theory based on their observations. So while they didn't do empirics (obviously) I believe they were inspired by their observations.

Nevertheless, there is a Hamiltonian (i.e. time series) approach one can take here. There are some working papers right now at Mises.org testing ABCT with a variety of countries. The trick is to get these papers into journals other than RAE or QJAE.

The elements are: 1) money growth creates a boom, 2) but the boom is doomed to bust, 3) the bust endures for a bit because there must be "recalculation," and 4) we'd have been better off without the artificial boom because it distorted resource allocation.

The recalculation problem gets you an unemployment rate that doesn't just disappear right away. To get the the need for recalculation you need misallocated resources, which means some price was wrong. It could be the exchange rate as in Cantillon, but the usual suspect in the modern world is "the" interest rate.

For wrong interest rates to create a recalculation problem, you need at least a little bit of basic capital theory in the model. That's a weak point in the theory IMHO, because capital theory is hard and complicated. We should cut the Gordian knot with some notion of the "duration" of an industry or sector, rather than impenetrable concepts such as "stage of production" and "average period of production." The minimal element of capital theory in the model also lets you say that the whole cycle was wasteful and should be avoided in the future. Without misallocation, unemployment would be the only problem, which we could fix with socialized unemployment insurance.

I don't see why we can't repackage those elements in a language that makes sense to mainstream macroeconomists. Am I missing something?

It seems to me that mainstream economists don't want to understand the ABCT. Take Friedman, for example. He called Pure Theory incomprehensible. I don't get it. I'm certain Friedman could have understood Hayek if he wanted to.

J. Oxman, I agree. I would love to see Austrians do more econometrics. I have promoted Structural Equation Modeling before. It's great for testing competing theories. But Austrians should also adopt neural networks, trees, support vector machines and other non-statistical methods that are far more powerful than traditional statistics and aren't encumbered by the long list of assumptions required for statistical analysis.

@Other Roger
Hayek's The Pure Theory of Capital *is* incomprehensible, as Hayek pretty much affirmed in Hayek on Hayek. We must cut the Gordian knot! We must abandon the usual versions of Austrian capital theory when doing macro. Give it up; game's over. Next topic. How 'bout we consider "duration" as a tool for an empirical research program in Austrian economics?

J Oxman-

Please provide links to the working papers. I'm under the impression people since Hicks in the 1930s have been trying to come up with a formal model to no avail. I don't want to name-drop, but I also heard pretty prominent people laugh the idea of econometrically testing the theory because all available data is designed with Neo-Keynesian theory in mind, so it's a square peg / round hole situation.

Koppl, I don't agree. It took me a couple of reads to understand it, and there are a few things beyond me. But it is far from incomprehensible, especially if you read PII first, which I think is Hayek's best explanation of his cycle theory.

As for research, I think Hayek could be simplified by just dividing capital into raw materials, capital goods, transportation and consumer goods production.

I teach at a small private college and the text chosen by the school starts off with PPF. I teach the standard stuff which includes a discussion of a PPF of investment vs consumption. I use the opportunity to slip in Garrison's take on business cycles.

@ Roger McKinney

Sure, there's a huge battery of statistical techniques one can use. I'm just a specialist in one of them, so that's what I talk about.

@ Ryan Murphy

Here is the link to all their working papers: http://mises.org/periodical.aspx?Id=7

I just recently read the Bjerkenes et al. paper and thought they had a reasonable approach.

I don't agree that all available data is designed with Neo-Keynesian theory in mind. To me that reflects a lack of understanding of just how much data is available.

One must note that testing the ABCT should not, under any circumstances I can see, follow a DSGE approach of high-level aggregation. Austrian economics is about subjective decision making - let's look at what households and businesses actually do, why not?


"Catlans, I don't get it. Consumer credit? Yes, but what about scarcity? If you can produce more of everything, then you are saying there is no resource constraint."

There is scarcity. The degree of scarcity represents the degree of malinvestment (see that figure I provided in my first comment). The more consumption, the more malinvestment you'll have with any given degree of new capital intensiveness (whether it's a widening structure of production or a lengthening structure of production).

There's no reason anyone here should take me seriously, because I'm not much in the way of an "accomplished economist."

However, I want to put something out there for consideration...

I think a lion's share of the confusion comes from the fact that ABCT is **NOT** a comprehensive description of all macroeconomics. There is differs from other schools of economics in a vital way.

A Keynesian view of business cycles involves synthesizing a description of business cycles in a way that jives with the rest of Keynesian theory. Obviously, the Robert Lucas stuff was a big part of that, too, and it lead to a different view of macroeconomics.

But ABCT can be taken "as-is," without implying something about Y=C+I+G+NX. That's important. ABCT is simply a description of what happens during business cycles. There is no comprehensive "Austrian macroeconomics." The Austrian School takes each question on an issue-by-issue basis. Descriptions can be improved by questioning the logical premises and conclusions, and perhaps numbers play a role there (why does X increase when theory suggests it would decrease?)

But one of the major benefits of ABCT is that the theory may speak about e.g. interest rates, but not every business cycle movement is going to have a big implication about interest rates the way it does with Keynesian theory.

Am I making sense?

I think an effort to change the minds of people like Krugman is a fool’s errand, but changing the dominant paradigm in economics is doable. That doesn’t mean changing mainstream econ. It means having more students in Austrian programs than in mainstream ones with businesses preferring Austrian grads.

To achieve the above, one would need to hire a good PR firm to survey potential econ majors to find out what appeals to them about econ. The PR firm would identify hot button issues in which Austrian econ has an advantage and which Austrian colleges would push in their promotions to potential econ majors.

In PR and marketing there are essentially two strategies for dealing with a dominant competitor when you’re the little guy: 1) show how you imitate the dominant competitor but add value, such as being cheaper or 2) differentiate yourself from the competitor. Austrian econ would definitely want to follow #2.

However, Austrian econ lacks credibility for a successful launch of #2 due to its small following. Credibility comes from status. Mainstream econ has the status from 1) the size of its following and 2) existing at the most respected schools, such as Harvard. In order to appeal to prospective econ majors, Austrian econ needs to borrow credibility from influential people. Ron Paul has loaned the school a great deal of credibility and he is a chief reason that many people are learning Austrian econ today. Krugman has finally begun to address Austrian econ only because Ron Paul has persuaded some Republican pols to embrace it. Glenn Beck has helped, too. Beck has high credibility with conservatives.

A PR firm would do it better, but in my opinion using Ron Paul, Glenn Beck, anyone in a respected financial firm or the Fed who follows Austrian econ, actors, journalists and anyone with high credibility to recruit students to Austrian schools will increase the sheer numbers of people studying Austrian econ. Credibility will increase again with numbers. But we’ll always have to do without a voice at the top private schools because they have staked out the territory of providing economists to tell the emperor how beautiful his new clothes are and politicians will always demand such.

I think the major problem is that there is no widely accepted theory of the pricing process. The Austrians have one, and I believe the most accurate and comprehensive one, but it's not widely shared by the profession. Without an explanation of a working pricing process, how can anybody accurately comment on the shape of an economy ('shape' meant both literally and figuratively)?

Capital theory is a specific part of this pricing process and coordination theory. You have to understand the basics before you understand the more complex concepts.


I'm going to disagree with you that there isn't an Austrian macro. It's just that Austrian macro looks way different than the other schools of thought's macro.

ABCT is a very visible part of the Austrian school, but I think it fits into the Austrian as an element of the study of interventionism. Maybe it's the most interesting, or most controversial, I don't know. But, to my view, it's just another facet of how an exogenous force (e.g. the gub'ment) interferes with the catallaxy.

Daniel: "But for it not to be embarassing, Austrians oughta think more seriously about underemployment equilibria..."

Using mainstream terms, assume that a depression is an underemployment equilibrum with massive underused resources. The expansion is a movement toward full employment equilibrium with scarce resources. The journey from one equilibrium to the other may take years, for most of which there will be unemployment and underused resources. Investment and consumption can increase at the same time without rising prices in consumer goods as long as unemployment and underused resources exist.

The Ricardo Effect requires that consumer goods prices rise relative to capital goods. But that won't happen until underused resources become scarce. Productivity increases also mask rising prices. So the Ricardo Effect has a trigger that doesn't pull until the expansion gets close to the latter equilibrium.

That's what Cowen is seeing in his data series.

RPLong, I think that makes sense. I characterize the ABCT as a medium term model. Mainstream models, like Solow's growth model, are accurate for the very long run, but can't explain medium term fluctuations with the long run.

Roger, how about deregulation of the labour market as a second line of attack on the unemployment problem?

On the seachange in development economics, how much more data was required to get the message that Peter Bauer was sending in the 1940s? How much data did Paul Samuelson need to understand the soviet economy when he was writing about it in the 1950s?

It would help to have a seachange in the use of data, from the "confirmation" approach to the critical approach. Hence the need to revolutionise the teaching of philosophy, the "long march" through the philosophy schools.

J Oxman,

What I mean is that, for example in a Keynesian system, your AD is lagging, so you increase G so that Y will also increase; but of course there are big implications as to how that is going to impact your IS-LM, etc. etc. etc...

Austrian macroeconomics doesn't work this way. I don't have the quote handy, but there's a great paragraph in Human Action after Mises describes the crack-up boom, where he takes some time out to say, basically, "Just because credit expansion works this way doesn't mean you're going to see inflation every time the central bankers expand credit -- there may be something else going on that stops inflation from happening, but it's a separate issue."

The point I'm trying to make is that just because business cycles can be described using ABCT doesn't mean that every increase in the money supply is going to create a massive market bubble. Rather, credit expansions *in general* tend to encourage malinvestment *in general*. There may be something else at play, but that "something else" will have to be analyzed separately, outside of ABCT.

This is so different from the Keynesian theory where everything is connected by 4 or 5 equations that I feel it's necessary to keep in mind when discussing macroeconomics with Keynesians or Monetarists. They're going to suck an Austrian into a Keynesian rathole, because they can't conceive of market forces in any way except the Keynesian way. It's a rhetorical limitation.


I get your point now. Yes, I agree with you, now that I understand you. It is a difficult gulf to bridge, no doubt about it.

One explanation for the co-movement of consumption and investment (over consumption coupled with malinvestment) is that when market interest rates are below the natural rate, investors take on projects that they would not if market rates were higher (which are later revealed as malinvestments, then written off or restructured), while consumers save less and fund more consumption by taking consumer loans or using their home equity as ATMs.

Mises and Hayek developed their theories at a time when the consumer loan market was much less developed than it is now, and when no one borrowed from his home equity to fund a party. The interest rate hinge is critical in this; I see little discussion or even acknowledgment of it in the comments.
Have we arrived at a kind of hydraulic ABCT?

Machlup considered the capital theory stuff Hayek's best work, and cited it as the central reason for giving Hayek the Nobel Prize, in Machlup's brief for Hayek requested by the Nobel committee.

Machlup was a President of the AEA, and as well respected as economists come.

Roger writes:

"Hayek's The Pure Theory of Capital *is* incomprehensible"

It's not incomprehensible. It requires effort to understand. Any complex theory should require effort to understand. Perhaps somebody can re-write it in more understandable English, but unfortunately I don't think that "somebody" has come along yet (all modern reinterpretations I've read have been subpar).

Greg: Yeah, Machlup used to say that in class too. I hate to disagree with him, but I just don't see it.

A lower interest rate increases the demands for all sorts of goods.

The demand for consumer goods rises and the demand for capital goods that help produce consumer goods in the distant future rise. But the demand for capital goods that produce consumer goods in the nearer future also rise. This is both due to a projection of the increase in current consumption into the future as well the lower interest rate discounting the future revenues from those consumer goods.

It is not at all obvious that the production of capital goods that help produce consumer goods in the near future will fall in the face of rising demand.

By lower interest rate, I was speaking of one due to an excess supply of money. If there is an increase in the supply of saving or a decrease in the demand for investment, the lower interest rate will be combined with a decrease in demand for either consumer or capital goods that are only dampened by the lower interest rates.

Why is there are presumption that a lower market interest rate causes this to be confused with increase in saving supply? Why isn't the confusion with a decrease in investment demand?


Demand for capital goods doesn't just rise because of a "projection of a future rise in consumption" and "lower interest rates discounting the future revenues", but because the pricing process adjusts along with changes in expenditure patterns. There is a fiscal incentive, provided by rising prices in any given stage of production, which causes non-specific goods to flow to the preceding stages of production. Hayek's Prices & Production deals explicitly with these changes in relative prices.

One of the major reasons I was let down by Garrison's exposition in Time and Money is precisely because he fails to elucidate (or even really mention) this pricing process. Instead, he alludes to the Hayekian concept of 'spontaneous order', which is completely misleading (almost as mystical of a term as the 'invisible hand'). Changing prices is how the structure of production changes and takes its shape.

Might the comovement occur because most of the consumption is in fact to create capital goods, which are in turn not being consumed? Mark Skousen argues here


that most consumer spending is of this sort.

Also, the artificially low interest rates that make money cheap is also discouraging saving. Some of that money will go into investments, but some will go into spending, since it's not worth saving the money. Might this also not contribute? Also, is the data being disaggregated?

"Might the comovement occur because most of the consumption is in fact to create capital goods, which are in turn not being consumed?"

By consumption we mean unproductive consumption; i.e. consumption without further production.

I've being busy recently, I only just got around to reading these the three top threads.

I think that to some degree people are making things too difficult for themselves. Let's say we have an expected rate of price inflation. Then, let's say that we have an unexpected monetary injection that will later lead to a rise in prices higher than the expected rate. Surely in the interim until that extra rise is discovered agents will treat the new money as they treat all money. How can this not cause plan discoordination?

I agree entirely that capital triangles, forced saving, and discussions about the split between I and C and their composition are useful. But, I don't see how anyone can argue that it can't possibly have an effect.

To be clear, I think that more empirical research would be very useful to find out the magnitude of the problem. Though I very much doubt it, it could be buried in the noise. But, I don't think that Bill Woolsey and Arash Vassei can deny that *theoretically* there could be a problem.

Many posts above Daniel Kuehn wrote:

> There is absolutely nothing about "full employment by
> accident" and a liquidity preference theory of interest
> rates that eliminates the Austrian point about the
> relationship between the interest rate and the
> roundaboutness of production.

Yes there is. You have to choose one or the other.

If interest rates are determined by the Keynesian liquidity preference mechanism then they don't reflect a rational allocation of resources across time. The roundaboutness of the production process is accidental, it has nothing to do with consumer preferences.

As I understand it in Keynesian theory progress occurs through technological progress which allows greater output to be produced for the same inputs at any interest rate or degree of roundaboutness. With the liquidity preference theory of the interest rate nothing else makes sense.

Jonathan Catalan has written a good explanation of his argument against Bill on his blog. I agree with him.


As I understand it, the ABCT is based on austrian microfoundations. Therefore, economists that don't know or agree with austrian micro, will not understand/agree with the ABCT.

The ABCT explains the business cycle as a process where a non-anticipated monetary expansion will trigger increased consumption and production plans. As the individuals subjective perception of his feasible consumption set (present and future) increase with added money stock (and prices didn't increase since the expansion wasn't anticipated). When these individuals proceed to execute their consumption/investment plans, the supply of money increases and prices increase, reducing their budget sets. As result, expected utility falls and action plans cannot be completed. Therefore, we have a crisis.

This theory is based on the assumption that the expectations of individuals regarding the data of the market can stay disconnected from the underlying data for some time. But eventually, they discover the real data. In standard macro we have the new classicals that assume that people's expectations are always glued in the underlying data and the keynesians, that don't think people can be rational.

The most important part of the ABCT is not the structure of production/intertemporal production plans/capital theory part (Hayek even explained economic fluctuations in the 70's without reference to capital structure). The capital theory part explains why it takes a long time for the market to return to the state before the monetary injection and why monetary expansion can produce so much trouble, as explained by Roger Garrison.


Your argument fails because of the green and blue money thought experiment.

Suppose all of the newly created money is blue and the existing money is green.

You are assuming that if we look at the blue money as it passes through the economy, we are seeing distortions.

But this is obviously false. It is easy to conceive of situations where there blue money follows the pattern that money (the green money) would have followed without any money expansion.

New money issued by banks. The banks are conservative and only lend to business. They only lend to high yield projects. Every penny of the new, blue money funds projects that would have been funded if their hand been no new money created.

If there had been no new money created, the high yield projects would have been funded by bonds. The firms that are borrowing blue money from banks don't sell any bonds, and so bond prices rise and their yields fall.

Firms that with higher risk, more speculative projects (or perhaps more roundabout projects that are only profitably at a lower yield) sell bonds instead. Those who would have bought the bonds of the firms with the higher yield projects, to some degree, buy these newly issued bonds with their green money. They also purchase consumer goods with their green money.

The change in the flow of expenditure all green money. The old money is shifted so that there are more consumer goods purchased and risky (or longer maturity) projects. The new money is solely used to fund projects that would have been undertaken if there had been no increase in the quantity of money.

In most likely scenario is that a mixture of old and new money will be spent. But it is an error to focus at all on the particular money issued by the banks.

The best way to see this is that the excess supply of money is matched by an increase in the supply of credit, and a lower market interest rate. This should raise the demand for just about everything, but demands for goods with more interest elastic demand rise more than the demands for goods with less interest elastic demand.

Now, if you want to go with scenarios were the money is not lent into existence, the blue/green money still shows that tracing the new money is pretty useless, but only after the first step. If the government creates new money and buys tanks, it does stimulate the tank market.

If the banks create new money and lend it, it does impact the loan market and interest rates, but there is no reason to think that change in the pattern of demand has much to do with what kind of loans the banks happen to make.

And, in the credit market, interest rate scenario, the flows of demand are based on interest elasticities. That __is_ the effect on flows of expenditures. Imagining that there is the blue money flowing out there causing problems is an error.


Consumption is spending on consumer goods. Most consumption is in fact to product capital goods that are not consumed? I think you should try to explain this in another way.

Catalan: By consumption we mean unproductive consumption, consumption without further production.

Who is "we?" And why would you use consumption to mean "unproductive consumption." And why is consumption without any further production unproductive?

Easy -- if a miracle occurs.

Otherwise, impossible to imagine.

We call science that depends on magic _nonscience_.

Bill wrote,

"It is easy to conceive of situations where there blue money follows the pattern that money (the green money) would have followed without any money expansion."

This is a version of Kaldor's argument against Hayek on forced savings and the boom / bust cycle.

Kaldor also depended on the "science" of .. "and then a miracle occurs".

Bill writes,

"It is easy to conceive of situations where there blue money follows the pattern that money (the green money) would have followed without any money expansion."


I don't see the point of your "blue and green money experiment", sorry. Yes, if newly created fiduciary media doesn't enter circulation then there won't be distortion of prices caused by this excess fiduciary media. Austrian cycle theory presupposes that this fiduciary media will enter circulation by the issuance of bank loans to investors.

Yes, there are many things that this new money will be spent on. Yes, not all of it will be malinvestment (I'm not sure who ever claimed it all was, anyways). Yes, some of it will be spent on projects which may have been committed to anyways.

What Mises and Hayek predict, though, is intertemporal discoordination, where the degree of capital intensiveness doesn't match the volume of savings. You can invest in what seem very low-risk investments during a boom period and still see your investment fail when prices readjust (housing, for all intents and purposes, seemed "low risk" as long as you didn't realize it was a bubble and as long as you figured that if it was a bubble it would pop much later than it actually did).

I agree with Greg, your "thought experiment" doesn't correspond with reality.


Put on your thinking cap.

Do you really think that my argument depends on a miracle?

The "most likely scenario is that a mixture of old and new money will be spent. But it is an error to focus at all on the particular money issued by the banks."

The most likely scenario is that there will be an increase in the flow of expenditure on all interest sensitive product regardless of whether newly issued money rather than old money is spent on them. That doesn't mean that this pattern of expenditure is sustainable or desirable.

Suppose all credit is bonds. Then bank loans appear for the first time, funding loans with banknotes. Every single banknote represents an excess supply of money. The old money is all gold coins.

The particular places the banks make the loans are not important. Some of those projects would have been funded by bonds. Those who would have purchased the bonds instead purchase other bonds issued by firms that would not have sold bonds before or else purchase consumer goods. They are all using "old money," the gold coins. While some of the projects funded with the banknotes would not have been made otherwise, some of the projects funded with banknotes would have been made anyway. It is only those that would not have been made that represent a distortion of the pattern of expenditure.

But, the added spending on consumer goods (using gold coins) would also be a distortion. Because those buying the consumer goods would have purchased bonds funding some of those projects that where instead funded by bank loans. That is added spending due to the excess supply of money and the credit creation. But only "old money" is being spent.

Simiarly, the projects funded by firms that sell bonds (for gold coins) to people that would have purchased bonds from firms that instead borrowed from the banks, are also a distortionary increase in demand, even though it is gold coins that are used to buy those bonds and then used by the firms to purchase capital goods.

Some of the projects funded with banknotes are likely to only have been funded because of the credit creation, but some of them would have happened anyway. And some of the projects funded with bonds sold for gold coins would only happen because of the credit creation, though others would have happened anyway. And some of the consumption funded with gold coins would have only occured because of the credit creation, even though it involves the expenditure of gold coins. And, of course, much of that consumption would have happened anyway.

That the banks lend the money in certain places doesn't matter. They could stick to business loans, short term business loans, consumer loans, it doesn't matter (or not much.) The major effect is lower interest rates, which impact the demand for all sorts of consumer and capital goods.

But it is the interest elasticity of demands that determine the disturbance to the pattern of demand, not the pathway taken by the new money through the economy.

To me, the greatest element of complication is that expected money expenditure on consumer goods at various future dates plays a more important role that the interest rates in determining the demand for various capital goods.

And, of course, impact on the pattern of production depends on the price elasticities of supply and the price elasticities of demand, as well as the changes in demand.

For example, if all the supplies were perfectly elastic, the pattern of production would stay the exact same. There may be a change in the pattern of prices, but there would be no change in the pattern of production or employment. (Not a likely scenario, except in the very short run and long run.)

I am confident that Hayek takes all of this into account in some kind of complicated manner. I am not sure that you do.


The important part of the thought experiment is that the credit creation impacts the pattern of expenditure of the green money. That is, the already existing money.

But these changes are in response to changes in prices--interest rates. Remember, I was responding to your reference to the "fiscal" impact of flows of expenditures. (I think fiscal is an awful term to use in this context.) I was responding to your claim that I was mistaken to identify the impact on the pattern of expenditure with interest elasticities of demand and further, expectations of future consumption demand.

I thought you were focusing too much on where the new money goes. Well, any excess supply of money and matching increase in the supply of credit, and reduction in the market interest rate will impact the flow of "old money" too.

This is old hat. Long ago, advocates of the Austrian business cycle theory have said that since banks now make consumer loans, perhaps credit expansion increases the demand for consumer goods. Whereas, when Mises and Hayek wrote, banks made business loans, and so, the credit creation increased the demand for capital goods.

Well, that was wrong headed. The credit creation should impact the demand for consumer goods even if banks make no consumer loans. Consumption is impacted because households would have purchased bonds to fund some of the projects that are instead funded by bank loans. The households who don't purchase those bonds will either purchase other bonds (or other financial assets) or expand consumer expenditures. Other firms will sell new bonds (to those who don't buy from firms borrowing from banks) which will fund investment projects for those firms. These are distortions created by the excess supply of money, increase in the supply of credit and decrease in the market interest rate. These distortions should not be identified with the flow of new money.

A lower market interest rate results in a decrease in the quantity of saving supplied and an increase in the quantity of investment demanded. With perfect credit markets, where the banks choose to lend will have no effect. In reality, there will be some effect, but there will be shifting so that you should just look at this as a decrease in the interest rate. But, also, what will be the effect on consumption expenditures at various future dates.

As I explained earlier, you are making a serious mistake to assume that I don't understand the Austrian business Cycle Theory. I understand quite a bit.

Richard Ebeling's comments make a lot of sense to me.

Jonathan and Bill,

I wasn't sure if both kinds of consumption or just final consumption was being considered "consumption" here. That's why I asked. Thanks for clearing it up for me.


I don't know how that "not" got in there. What I had meant is that consumption of goods to create capital goods might be considered consumption, and would explain the comovement even if final goods weren't being consumed at a higher rate. But if we are calling consumption the consumption of only final goods, then my point is moot.

I would suppose that lower interest rates make it easier for people to borrow money to spend, which might explain the comovement. Also, I still wonder what happens when the data is disaggregated.


The problem I see is that you are failing to take into consideration interest rates. Low interest rates make money cheaper, so people are more likely to take risks. The more risks people take, the more risks don't pan out (even if the percentage is higher, which is probably not the case). High interest rates make money more expensive, so people are going to be more conservative with their borrowing and investments. The injection of more money into the economy has the effect of lowering interest rates. If prices are going up at a faster rate than the interest rate, it makes a lot of sense to borrow to buy now and sell later -- whether that be stocks, to buy capital goods to produce a product, etc. After all, if prices are going up due to monetary inflation, no matter what happens, you can always sell off the capital goods you purchased at a higher price than you bought them at and pay back your loans and still come out. This creates an effective negative interest rate, with which one gets inflation. So the creation of money isn't non-neutral, as you seem to be implying. If monetary growth slows or the interest rates go up, you get your recession triggers, as either of these will burst the bubble.

Mises and Hayek wrote between 100 years and 70 years ago. The economy is an ever changing dynamic moving entrepreneurial discovery process. This means that historic thought processes may well need to be continually updated.

What was observed and described then by these Great Teachers may well have moved on to the unique position we have today. This should be expected. Therefore , it is for the teachers of today and their pupils to carry the research program forward and expect new theories and new empircial data to emerge and embrace it.

I as an entrepreneur have always read the Mises - Hayek ATBC to mean government inspired artificial inducement of the lengthening of the structure of production causing the production sector to bid away goods from the consumer sector. When the production sector factors of production get paid for the goods they produce, then they spend on consumption on the whole. This bids up prices in the consumption sector. Like vultures feeding on a dead body, both the production sector and the consumption sector start to compete with the same finite resources. So expect co movement of both the production and consumption sectors, with a lot of it happeing at the same time! Economists, please remember, production does not happen, then time lag, then consumption of goods. This dynamic process is for ever evolving . In the current case of disruption, we have had nearly 2 decades of elongation of the structure of production and nearly two decades of inflated consumption, both competing V each other. All simultaneous and it would be impossible to seperate out the two.

HOWEVER, if you observe the production sector of the economy, you can see that the osilations in production are far greater (as predicted by classic ATBC) than the movements in the consumption sector. Mark Shousen's work is very good and demonstrating this. So is Sean Corrigan's that is used in his commercial research. THIS I submit is the salient point ATBC guys should be focusing on.

The crack up boom does then crack as it is revealed, the Emperor is indeed naked and there is not enough savings to provide the money to buy the goods and services thus produced. This is the bust.

It is all very well economists arguing empirical points here and empirical points there, but keep your fire power focused on what is indeed practical and useful. The ATBC is very much so a practical tool and it needs to be brought up to date for each specific change in the economy as each has its own special sets of circumstances.

Pete , well done for posting this, but make sure you keep the research real and not navel gazing about some tiny detail of theory and empirics.

To the degree we take Garrison's static analysis literally, it is absurdly wrong. But who would do that? When I do similar analysis, I surely realize that it has to be superimposed over a growing economy.

Ebeling's point that Hayek and Mises assumed a growing economy is correct.

If, however, this has anything to do with Cowen's critique... you must be kidding.

As I wrote above, _of course_ Cowen, like everyone else, understands that the production of consumer and capital goods is increasing all the time.

_Of course_ any misalloscation story is going to be translated into spending on capital goods growing more quickly that it should (in some sense,) and production of capital goods growing more quickly too. The question is whether some kind of tight scarcity constraint means that the production of consumer goods must be growing more slowly than they should in the short run, or else, somehow, the constraint is lose in the short run, so that when spending on consumer goods grow too quickly, the production of consumer goods grows more quickly too.

Then, there are these claims the the production of some capital goods grows too quickly, the production of consumer goods grows too quickly, and the production of other capital goods grow too slowly. That sound fine in theory, but when the two types of capital goods are added together, they shouldn't show up as growing too quickly and booms would look like consumer goods growing faster and capital goods growing like usual. While more durable capital goods vs. less durable ones could probably be measured and they should show the problem, other parts of the problem of higher order vs. lower order capital goods is only in terms of plans divided up among many entrepreneurs.

Anyway, my understanding is that everything is usually growing, and the production of consumer goods grows more quickly in booms and more slowly in mild recessions, but shrinks in the really bad ones. The production of capital goods grows more quickly in booms and faster than consumer goods. The production of capital goods shrinks in recessions.

As I see it, the traditional ABCT theory says that spending on consumer goods should accelerate at the peak. The firms producing them should face bottlenecks of key resources, (because the appropriate capital goods were not produced, the resources instead used to produce extra durable capital goods ready for the distant future or capital goods that cannot be used to produce consumer goods but only other capital goods that will be able to produce consumer goods in the future.

Anyway, the bottlenecks means the production of consumer goods can't expand. We should see inventories of consumer goods dropping. And the prices of consumer goods should rise.

This is usually missing in the real world. (I think.) Instead, spending on capital goods slows then falls, then spending on consumer goods begins growing more slowly. Bottlenecks in the production of consumer goods don't appear.

In the Garrison approach, spening on all goods rises too fast and production of all goods grows faster for a time, but as the price level grows faster, real expenditure and real production returns to trend. The excessive growth in spending and production will be general, but more for those goods that have more interest elastic demands. The problem with this approach is that it is hard to see there being any misallocation. Production and employment are too high in the boom (like the more convetional monetarist story,) and the excessive production and employment are concentrated in some sectors, but what needs to happen is that spending and output just grow more slowly to return a sustainable growth path.


I don't think the bust must come when some hard real constraint is met. It can come before that because of expectations of its imminent arrival.

That is, if the inflation rate rises and real returns on long term projects falls then people may well come to the conclusion that the real value of their assets has fallen.


I don't ignore interest rates.

I know the difference between real and nominal interest rates.

I only seem to be implying that money creation has neutral effects on the economy because you have pigeon-holed me and someone told you that monetarists (I guess) believe that money creation is neutral.

My actual view is that only excess supplies of money cause any problem. As long as money expenditures on output grow at a slow steady rate, there is no monetary disequilibrium, no deviation of the market interest rate from the natural interest rate, and no imbalance between saving and investment. I have no expectation that the quantity of money or the market interest rate will be especially stable in such a scenario.

Avoiding monetary disequilibrium and keeping money expenditure on target never requires any kind of bubble. Bubbles can exist even if there is no monetary disequilibrium and money expenditure continues to grow at a slow, steady rate. And bubbles can burst, without there being any monetary disequilibrium and money expenditures cam grow at a slow, steady rate.

When there is an excess supply of money, I expect the real market interest rate to fall below the natural interest rate. The demand for just about everything will increase, with the demands that are more interest elastic rising more than others. One reason for higher interest elasticity is when consumption is going to be generated by a good in the more distant future. Any change in the composition of output and employment, rather than just demand depends also on the price elasticities of supply and demand for various goods.

All of these effects only exist with the excess supply of money. Once the price level rises enough for the real supply of money to adjust to the demand to hold it, then the real supply of credit returns to where it was before, and the real market interest rate returns to the natural interest rate. The demands for various goods go back to where they were.

The supplies of goods tend to be inelastic in the short run. Further, how elastic they become over time is going to very much depend on how persistent the change in price is going go be. Firms won't change their productive capacity much in response to what they consider a temporary change in price.

Also, the real market interest rate right now isn't the sole interest rate relevant to the demands for various goods. How interest rates impact the demand for a good (it's interest elasticity of demand) depends on current and future interest rates.

So, assume that everyone is myopic and treats todays market interest rate as permanent. (Long rates change point by point with the Federal Funds rate.) Assume that the supplies of all goods are perfectly elastic in the short run. So, shifts in the demands for different goods shift the composition of output and employment.

And further, prices take a really long time to adjust, and so excess supplies of money persist for a really long time, and the market interest rate can be below the natural rate for a long time.

More importantly, if we turn this around, and assume that the central bank is targeting the real rate, or maybe the nominal rate where inflation is expected to be zero, then we can conceive of money expenditures growing, perhaps at an accelerating rate. These interest rates might persist for a long time and be projected into the future. And supplies could become quite price elastic as the pattern of demand persists.

However, the progressively more rapid growth in money expenditures is clearly inconsistent with slow, steady growth of money expenditurs. But it is also inconsistent with inflation targeting.

In the real world where a very short term nominal interest rate is targeting, and this target changes every month, and there is a target for the inflation rate, it is doubtful that the current value of the interest rate is taken as permanent or the pattern of demand that might be associated with any excess supply of money will be taken as permanent and so result in a low supply elasticity and so a prompt change in the composition of output.

I don't favor the status quo approach to monetary policy, but I doubt that it plays an important role in creating malinvestment. And so, I don't think liquidations of malinvestment caused by interest rates below the natural interest rate plays an important role in real world recessions.

It is exactly the position of Milton Friedman. It could be. But it probably isn't that important of an effect. It is more or less what Krugman says. It is more or less what Cowen says.

By the way, I think housing was overproduced for a variety of reasons, and that this is depressing the productive capacity of the econom and employment, and raising structural unemployment at this time. I just don't think that an excess supply of money was a significant cause of the problem.

I don't understand the objection. It's true that consumption and investment, in the short-run, simultaneously rise, but such investments (malinvestments) will not be completed on time, at all, or will be completed only at the expense of other, more warranted productions, precisely because scarcity exists. It's also why forced savings is required in order to complete the malinvestments and ease the liquidation and restructuring process (the bust).

In other words, the concept of malinvestment is meaningless in a world without scarcity because all such investments could be completed without a problem (even as society simultaneously elevates consumption). We must distinguish between completable investment and investment projects which begin (due to an expansion of money in the broader sense beyond the demand for cash holdings) but which cannot conceivably come to fruition in the "long run."

In fact, consumption and investment will simultaneously fall during the restructuring process in order to restore "balance" (move towards a sustainable PPF). So again, I fall to see how the Austrian theory of Cycles (the Mises-Hayek-Lachmann story) does not account for scarcity.

Thanks Bill. This helped in understanding your underlying point.


I made the statement not because anyone had told me anything about you or monetarism, but because what you were saying sounded like you believed in the neutrality of money. Nor could I tell that you had taken interest rates into consideration in what you had said to that point. That's why I stated my point as I did, that it seemed you held that position, rather than saying that you did hold that position.

Friedman argued that the money supply should grow at the same rate as the economy grew -- 3% for 3%, in the case of the U.S. That would prevent inflation, he thought, because that would mean that the same amount of money for the same supply of goods, on aggregate. The only problem with this is that the money does not enter the economy at the same rate in every sector. It is going to enter some sectors more than others. Continually dropping interest rates, as we saw, more or less, under Greenspan and Bernanke, encouraged riskier and riskier investments. U.S. housing policy directed the loans and, thus, the increase in money supply, into housing. Thus, the housing bubble.

@ Sheri:

Well put; I agree.


I've heard Friedman's claim that agents can "expectation" their way around the problem. I think that people overestimate how difficult this really is.

To begin with the prices which consumers are concerned about are the prices of consumer goods, not just consumer goods outputted in that period. The central banks only target the latter, and then quite loosely using price indexes that are somewhat dubious.

I'm not convinced that everyone has rational expectations about the interest rate, especially ordinary consumers. But, even if they do I'm not convinced that it fully solves the problem. It's not just a matter of expecting the interest rate to rise if it is particularly low, its a matter of disambiguating the real interest rate from the nominal interest rate. It's expectations about the real interest rate that are needed. Also, entrepreneurs must not be distracted by input prices that are cheaper than normal because those prices don't yet incorporate expected inflation.

Recently Frank Shostak wrote an article on the Cobden Centre site about whether expectations are "neutralising". Though I often disagree with Shostak I think it's very good:


I agree with Current. Most people go, "Wow! 2% interest rates! I'm going to buy me a house and flip it!" That is perfectly rational. Everyone acting rationally on low interest rates will create a bubble.

Purchasing a house using a variable rate mortgage planning to "flip it," is a speculative investment based upon expectations of future interest rates and home prices. Assuming that recently decreased interest rates will permanently stay low and that past trends in housing prices can be projected into the future may not be exactly irrational, but look to be poor entrepreneurship to me. And that it is that poor entrepreneurship that causes any malinvestment.

Of course, it isn't buying the house alone that leads to the malinvestment. It is rather that construction firms add to their capacity so that they can produce more houses because of the added demand. In other words, the elasticity of supply over time. Building sawmills because of a temporary increase in the demand for houses is an entrepreneurial error. And that is why there is malinvestment.

Poor entrepreneuriship, and now malinvestment.

What should happen is that short term interest rates fall, and long term interest rates fall to reflect solely the current long term interest rates. No one changes productive capacity except to reflect the temporary increase in demand. (supplies are perfectly inelastic.) Prices rise. The real supply of money and credit fall. Short term interest rates rise back to where they started.

Not likely, I know. But always keep in mind that this is what __should_ happen. And to some degree it is what will happen.

Suppose the natural interest rate falls for one year and then returns to its previous value. What happens? How does the market adjust. Suppose that it was something other than housing that caused the change, so that housing is just responding. What happens in the housing market?

I am more and more persuaded that interest rate targeting is especially bad, and interest rate smoothing just makes it worse.

One final consideration. Suppose entrepreneurs and the Fed both believe that the natural interest rate has fallen.

Compare that to a situation where the Fed thinks the natural rate has fallen, but entrepreneurs disagree.

Finally, what happens with a private monetary system when entrepreneurs wrongly believe that the natural interest rate has fallen? Or simply that it will be falling soon?

Troy, on Friday, I said:

"A lower interest rate increases the demands for all sorts of goods."

You claimed today that I don't take interest rates into account. Well, OK. I am sure that I have some posts where I didn't point that out. And I guess I shouldn't expect everyone to be familiar with my macroeconomic approach.

Your analysis of Friedman is wrong. It is true that a 3% money growth rate won't be inflationary, but not because the supplies of goods are rising 3%. It is that real income is rising 3% and Friedman believes that the demand to hold money will rise in proportion to real income--3%. Since the quantity of money grows with the demand to hold it, there is no monetary disequilibrium. The demands for various goods grow with income. The supplies of various goods grow with factor supplies and productivity. Supply and demand both increase at the same rate. The typical price stays the same. The price level remains stable. Of course, some prices rise and others fall.

With credit money, the increase in the demand to hold money is one element of saving. The supply of bank credit funded by money issue is one element of the total supply of credit. Generally, the growing income results in growing saving with added money balances just one part of it. Expectations of growing future income results in growing investment demand. There should be no particular trend in the natural intrerest rate. While firms often self finance, directly matching part of saving and investment, to some degree, the supply of credit and other sources of finance by households, including that provided by growing money demand, is matched by the growing demand for credit and other sources of investment by firms to fund investment. The market interest rate remains equal to the natural interest rate.

That it is credit money is important, but other than that, focusing on where the new money goes is a red herring.

Quasi-monetarists(like me,) would agree with the above, assuming the demand for money grows in strict proportion to real income. When it doesn't, we believe the quantity of money should grow or shrink the amount needed to keep the quantity of money equal to the demand to hold it.

While Friedman favored a 3 percent money growth rule, which implies 3 precent growth in money expenditures on output if money demand is strictly proportional to real income, the ideal growth rate for money expenditures is subject to dispute among quasi-monetarists. Some favor 3 percent growth, like Friedman, and others favor only 2% or 1% growth. And some favor the 5% growth we had during the Great Moderation.

I believe, at least, that the economy can adjust to any of these trends so that the real quantity of money grows with the real demand to hold money and the market interest rate remains equal to the natural interest rate. Real expenditures grows with productive capacity and saving and investment remain in balance. Your argument that a growing nominal quantity of money causes "inflation" in some sense, or that this causes malinvestment is wrong.

A shift to a new regime, or a monetary policy mistake, that results in an excess supply of money can cause malinvestment. But it is not a sure thing. That "new money" enters some sectors more than others is not a useful way to look at it. With credit money, it pushes down the market interest rate and raises the demand for just about everything to a degree that depends interest elasticity of demand. Projecting the low interest rates into the future and expecting the pattern of demand to persist are entrepreneurial errors that don't have to be made.

With a regime shift, this would be a necessary evil, and can partly be avoided by educating people about the new regime. Of course, given a monetary regime, excess supplies of money should be avoided.

The problem with many versions of Austrian Business Cycle Theory is that it focuses on the scenario of a constant demand for money, a constant supply of saving, and a constant demand for investment. Any increase in the quantity of money is treated as an excess supply of money. Any decrease in the market interest rate is treated as if it has fallen below the natural interest rate. Fortunately, most good Austrian economists know better.

Investment demand rises, supply of saving falls, the demand for money falls, and the quantity of money remains the same. What happens then?


I went back and re-read your comments on interest rates, and while you do bring up interest rates, you're not taking them into account in the same way I am suggesting you should. But of course I haven't been clear about that.

If interest rates are being determined by savings and loans, then of course the more savings you have, the lower the interest rates for loans. The higher the interest rates, the more people want to save and the less people want to borrow; the lower the interest rates, the less people want to save and the more people want to borrow. That is self-evident. The more money people save, the lower interest rates go, because the banks don't need as much money to lend out if there is already a high enough level of savings. This is also self-evident. But only if this is how banks get the money to make loans. If the Fed is creating money to lend out to banks for them to make loans, then of course the interest rate charged by the Fed is going to affect interest rates.

We saw, through the tenure of Greenspan and then Bernanke, a pretty steady drop in the interest rate the Fed lent money out at. Why offer high interest rates to savers to attract their money if the Fed is going to lend at such low interest rates? More, if you see the Fed fairly consistently dropping their lending rate over the length of time they did, it is rational to expect them to continue to do so. Of course, any inflation, regardless of source, above the interest rate makes that interest rate essentially negative.

What is the effect of this? Well, if low interest rates make people more willing to take risks, then you should expect more people to take more risks over time as the interest rates drop and drop and drop. If interest rates are kept artificially low -- whether through inflation or through various political pressures -- meaning they are not responding to the market, then there is a signal that more risks should be taken, even if people should be engaging in less risky behavior, as past risks haven't panned out. If interest rates were responding to the market, then as fewer loans could be paid back, interest rates would go up, encouraging more conservative investments. If downard pressure keeps interest rates low anyway, people spend more, whether it be on consumption or on investments or on capital goods. More risky behavior is going to result in more failures. It is also going to result in riskier types of investments, eventually leading to ponzi schemes and the like. All of this eventually leads to economic collapse down to the level where the economy should have been growing at (if we're lucky).

As for Friedman, I suppose after over 15 years I am bound to misremember his exact arguments. I'll have to put him higher on my reading list to refresh my memory. :-)


Draw a supply of saving curve with a posive slope and a demand for investment curve with a negative slope. Where they cross is the natural interest rate.

If there is no excess supply or demand for money, that is where the market interest rate will be. If there is an excess supply of money, then the market interst rate is pushed below the natural interest rate. Look at a lower interest rate. The quantity of saving supplied and the quantity of investment demanded rises. The amount invested (spending on capital goods) is greater than the amount saved (that part of income not spent on consumer goods. Income equals output, so it is that part of output that is not matched by spending on consumer goods.)

The Fed can push the market interest rate below the natural interest rate by creating an excess supply of money. If the Fed creates an excess supply of money, it pushes the market interest rate below the natural interest rate.

I find your "why pay higher interest rates to attract saving when the Fed will lend at lower interest rates" a bit confusing.

An increase in the supply of saving shifts the saving curve to the right. An decrease in the demand for investment shifts the investment demand curve to the left. Either or both together reduces the natural interest rate.

I wonder if Strigl's Capital and Production does not offer an important lesson in regard to the nature of capital and the structure of production -- one that seems, in my impression, to be missing from many discussions of malinvestment and the business cycle. This is the concept of the subsistence fund, the stock of consumer goods which makes it possible to divert production to higher order capital goods for the purpose of more roundabout production. If it will take Crusoe a week to build a net to catch more fish, he has to have a stored-up supply of food to sustain him while he diverts his efforts from the less efficient efforts at spearing fish that come within range. This stored food is properly viewed as capital -- i.e., as part of the structure of production.

In a modern economy, if an injection of fiat money pushes interest rates artificially low, it can of course make longer term, more roundabout capital projects appear more feasible than they actually are. If this sufficiently warps the structure of production, it leads at some point to a fall in output and living standards, due to the expense in labor and resources to re-build that structure. By analogy, if Crusoe cuts up his spears to fashion the framework for the net, but it turns out he was mistaken about the sufficiency of his subsistence fund and can't complete it before running out of fish, he can't just go back to using the spears. So he is forced back to a still less-effective structure of production -- catching fish by hand, or gnawing roots -- even just to get to the point where he can re-fashion spears and rise back to his earlier level of affluence.

In the modern economy, if the injection of money may also increase the sales of consumer goods, in effect it causes the subsistence fund to be depleted even as the new capital projects required a larger one. This is what makes it so counter-productive to try to stimulate the economy by stimulating demand.

Granted, the modern economy is much more complex that Crusoe's, but I think caution is needed to avoid losing sight of the essential underlying physical realities. Without descending into scientism, I think there is some similarity in the relationship of physics to engineering. I know damn well that you can't build a perpetual motion machine with a net output of work. You can go on all day with how the battery connects to the motor and the motor to the generator and the generator to the battery, with AC converters and transformers to raise and lower the voltage, etc, and it just doesn't matter because energy is conserved. Similarly, once the structure of production is distorted, through any combination of restructuring to unsustainable roundaboutness and/or depletion of the stock of consumer goods, a drop in living standards to below the baseline is going to result.


The broken Hayekian triangle represents the same sort of thing as the older subsistence fund idea. The subsistence fund in the older theories are those goods close in order to consumer goods but not consumer goods.

Toby Baxendale: "In the current case of disruption, we have had nearly 2 decades of elongation of the structure of production and nearly two decades of inflated consumption, both competing V each other."

You posted a good summary of the ABCT! Krugman assumes that historical data is the result of a controlled experiment comparable to the data from controlled experiments in physics. It's not. There are many many variables influencing the outcomes. There is a great deal of correlation and interaction going on.

On the other hand, the ABCT is a thought experiment holding certain variables constant. The ABCT follows the scientific method. Krugman doesn't because he fails to hold other factors constant.

Krugman and Cowen assume equilibrium for their critiques when the economy is never in equilibrium. Throughout most of the boom resources will not be scarce, so both investment and consumption can continue to grow. Only when full employment of resources is approaches is there any scarcity. Excess capacity is large until then. Then businesses consume capital in order to keep producing. But the bust comes shortly after idle resources have been used up.

I'm not sure that we can say that there has been 20 years of misallocation. I think that this gives expectations too small a role. I think for the past 20 years people have expected price inflation.

But, I don't think Bill's position is right either. I think he overestimates people's ability to think around the various problems of input prices and real interest rates I described above and act accordingly.

Expectations can get the economy some of the way around the problem of excess money creation, but I can't see how it can entirely remove it.

The market needs a new private gold standard bank.


Where do banks get the money to lend?

If you open a bank with, say $1 million of your own money, you can lend the money at interest to make money for the bank. Let us say, for simplicity sake, that you lend it all out at 10% interest, and that it all gets paid back at the end of the year. Let us say that expenses are $50,000. Once you have taken care of expenses, you now have $1.05 million to lend out. You do this over and over and over to make your money.

However, if you take in people's savings and pay them interest, you now have their money to lend out as well. So you have an incentive to take in savings and pay enough interest to encourage people to deposit their money. The interest rates you charge to lend are going to have to be more than what you pay out for savings if you are going to make money. Let's say, for argument sake, that you discover that the best rate to pay is 5% to get enough savings to lend out and make enough money to help cover expenses and add to the base money supply of the bank to lend out.

But suppose there were another bank that was willing to lend you money at 1%. It would make sense to borrow money from them and lend out at anywhere between 6% (to make the 5% in interest) and 10% (the maximum you had been lending at). On the one hand, the high rate will keep the same number of customers as you had; on the other hand, a lower interest rate, which you can now afford to offer, will attract new customers. The rational thing to do, then, is to borrow at 1% and lend out at lower than 10%, but higher than 6% to get the same percentage return or more. If you are borrowing from the other bank at such a low interest rate, you can afford to offer a lower interest rate to savers, because you don't need their money quite as much as you had before. You don't want to have much more in savings than you are lending out, because then you will lose money. So you have to balance savings taken in and loans made. If you can get the money at 1%, you have no incentive to take it in at a higher interest rate from other savers.

In other words, the Fed lending out money at the low interest rates they were lending to the banks encourages lower interest rates than would occur without their intervention. Thus, the banks won't pay higher interest rates to attract saving because the Fed will lend at lower interest rates. This is how you get interest rates below the natural interest rate that would occur in a market economy without a central bank artificially keeping interest rates down.


I understand the basics of central bank policy. I also understand financial intermediation.

Most credit is not provided by banks. The Fed, anyway, has usually made very few loans to banks.

An excess supply of money results in a lower market interest rate, and that results in a decrease in the quantity of credit supplied, which is the same thing as an increase in spending on consumer goods.

It is a mistake to focus too much on bank loans--either by banks to firms and households or by the central bank to banks.

The comments to this entry are closed.

Our Books