Lewin on Prychitko, Minsky, and the Financial System

Steven Horwitz

While in Dallas this weekend, I had the chance to spend sometime with my good friend Peter Lewin and his lovely wife.  Peter recently posted a nice entry at his blog reflecting on Dave Prychitko's recent RAE paper on the Minsky Hypothesis.  Worth a read.

Comments

[Greg Ransom]

Note well, Hayek argues against what Lewin identifies as the "Austrian" position, in Hayek's book _Monetary Theory and the Trade Cycle_, i.e. Hayek explains how -- even without government mismanagement -- cycles are inevitable. Hayek provides something not too different from a "Minsky Cycle" in Book IV of his _Monetary Theory and the Trade Cycle_, without the detail or helpful distinctions, but with very helpful additions in the theory of production through time, linked to credit and interest rate and leverage issues.

William White (former chief economists of the BIS) famously combined the work of Hayek, Selgin & Minsky to identify the central malinvestment, credit, and monetary pathologies of the 00s -- and directly warned Greenspan and the central banking community of what was ahead at the Jackson Hole conference in 2003 -- and again and again year after year in the detailed reports of the BIS.

[Steve Horwitz]

I don't have my MTTC handy Greg, but my recollection is that Hayek says that real factors can cause the natural rate to deviate from the market rate and trigger the cycle. Is that what you're referring to?

If so, then I think the argument is more subtle than that. Such a cycle would not result from "government mismanagement" but from a sin of omission: a properly functioning banking system should adjust the market rate to match the changing natural rate.

Can you give us a little chapter and verse in MTTC so we understand what we all need to "note well?"

[Greg Ransom]

Links of William White and his Hayek/Selgin/Minky based empirical work on the malinvestment crisis of the 00s here:

http://hayekcenter.org/?p=444

and here:

http://hayekcenter.org/?p=1688

and here:

http://hayekcenter.org/?p=3354

and here:

http://hayekcenter.org/?p=931

[Greg Ransom]

Hayek says that the existence of the banking and finance system, the "elasticity" of reserves, and cascading interrelations of deposits and loans among the banks and financial institutions allows the market rate to deviate from the natural rate, creating a natural instability in the system.

The discussion is Book IV of _Monetary Theory and the Trade Cycle_, particularly section 7 through 11.

Here's the punch line:

"The fact, simple and indisputable as it is, that the "elasticity" of the supply of currency media, resulting from the existing monetary organization, offers a sufficient reason for the genesis and recurrence of fluctuations in the whole economy is of the utmost importance — for it implies that no measure that can be conceived in practice would be able entirely to suppress these fluctuations."

[Greg Ransom]

What Hayek says is that real factors _can_ be the spur to an "optimism" cycle, or a banking distortion money cycle, but it doesn't have to be the spur.

Here's

"It must be emphasized first and foremost that there is no necessary reason why the initiating change, the original disturbance eliciting a cyclical fluctuation in a stationary economy, should be of monetary origin. Nor, in practice, is this even generally the case. The initial change need have no specific character at all; it may be any one among a thousand different factors that may at any time increase the profitability of any group of enterprises."

[Greg Ransom]

Hayek is saying that this is impossible -- there there is no perfect automatic adjustment mechanism as in the pure barter model lacking money and credit, so there will always be business cycles.

Steve writes,

"a properly functioning banking system should adjust the market rate to match the changing natural rate."

[Greg Ransom]

Hayek identifies non-perfectly equalibrating changes in leverage, risk position, liquidity, and so on as inherent parts of a system with money, banking, and financial institutions. Knowledge problems make the signaling system through the financial system and the profit and loss system inherently imperfect, esp. because things can build across time (due the non-optional time dimension of variable production processes) before signals of a lack of coordination assert themselves in non-avoidable ways (e.g. scarcity of inputs to production, and the collapse of profits, etc.)

In other words, Hayek identifies the open barn door were "Minsky" type mechanisms function -- as William White discusses.

Minsky doesn't have to have the mechanisms exactly right for anyone to acknowledge the existence of such things as changing leverage, bandwagons of optimism and pessimism, mis-read profit signals, and large groups changes in leverage and liquidity and risk and interest rate stances forcing changes on individual banks and firms.

Hayek specifically focuses in on how changes in the stances of other firms force individual firms to change their own stance.

All of this is made possible by the "loose joint" of money and credit, the time dimension of production and consumption, and the knowledge problem.

[Peter Lewin]

Thanks to Greg for this information. I am sure that many if not all "Austrians" would not deny my point about the existence of fluctuations even in the absence of money mismanagement. And the quotations provided by Greg convince me that certainly Hayek does not. (I expect similar quotations could be found in Mises.) Non-Austrians, especially critics, (conveniently) misunderstand the "Austrian position" on this; and shorthand characterizations like, "a properly functioning banking system should adjust the market rate to match the changing natural rate" encourage this kind of interpretation.

[Barkley Rosser]

Minsky and Kindleberger both looked to Mill as a source on "initial displacement," especially for bubbles, although this may not hold for strictly pure financial market evolution. In any case, that argument says that a bubble gets going because of some real side initial displacement that is probably temporary that pushes a price upwards. Then it keeps going as expectations change and one moves into speculative financing.

[Jerry O'Driscoll]

Very good post by Peter Lewin. On "people sould know better," I would add several points to Peter's excellent discussion.

First, there is the "greater fool theory." Anyone who has participated in a speculative bubble will tell that it was driven by that. They knew prices were unrealistic, but thought they could get out in time. Recall the now infamous comments of Citi's CEO, Chuck Prince, on needing to keep dancing so long as the music plays.

Second, Keynes addressed the issue in his beauty contest story. Everyone is guessing about the "average opinion," not real factors.

Third, as Lachmann was fond of remarking, we know people learn from experience but what do they learn? After the dotcom bubble, many learned that stock markets fall. So they invested in real estate for the next cycle. Now what? Bonds, gold and other commodities. Plus a dollop of Asian real estate. (RE isn't bad, just American RE.)

[Jerry O'Driscoll]

Hayek made the point about the inevitability of cycles in a number of places in MT&TC. He identified the gold standard plus an elastic currency. He even thought clearing houses could add to the problem.

In Prices and Production, written later, he introduced the idea of neutral money as a normative concept. I don't think it was successful, but it was an attempt to avoid cycles through policy.

[Ryan Murphy]

To add to Greg Ransom's point, Caldwell in a footnote on pg 15-16 of Contra Keynes and Cambridge says

"Mises and Hayek provided different accounts of the origins of the cycle. Mises blamed the cycle on the actions of bankers, especially central bankers eager to finance government activities by printing money. For Hayek, the cycle could be due to spontaneous action by bankers. But it also could be induced by an increase in the demand for loanable funds by firms (that is, an increase in the velocity of circulation) caused by increased business optimism, or the discovery of some new product or process, or a change in the structure of production, or a change in the form of credit: any of these could render money 'non-neutral'. See Hayek, Monetary Theory and the Trade Cycle, op. cit. chapter 4; Prices and Production op. cit., Lecture 4.)"

Although I guess the modern viewpoint is that free banking would offer incentives to counteract the forces Caldwell/Hayek are talking about?

Is Caldwell wrong that Mises did not understand the possibility of "endogenous" Austrian cycles?

[Lee Kelly]

Reading Lewin's account of Prychitko's take on Minsky's theory of the business cycle (whew!), I can't help but be reminded of the following cartoon:

http://www.epicidiot.com/evo_cre/images/science_miracle.jpg

[Richard Ebeling]

In response to Ryan Murphy's question about Mises and "endogenously"-caused cycles.

In the first (German-language) edition of Ludwig von Mises' "The Theory of Money and Credit," he did believe that there was a potential and a frequent basis for this sort of "endogenous" expansion of credit within the banking system due to the ability of the banks to extend "fiduciary media" (Mises' term for money-substitutes in the form of bank notes or demand deposits not 100 percent covered by "real savings" -- deferred consumption in the form of savings deposits out of income transferred to lenders through the financial intermediaries) in the face of "optimism" about current and future market conditions on the part of potential borrowers desiring to undertake investment and other business projects.

Indeed, this actually led Mises to also formulate a form of a "cost-push" theory of inflation that is briefly discussed in a subsection of the 1924 second edition of "The Theory of Money and Credit" (see, Part 2, Chapter 8, subsection 14 of the English-language version of Mises' book.)

Mises also presented this theory of a cost-push inflation in his inaugural lecture as a "privatdozent" (unsalaried lecturer) at the University of Vienna in February 1913, entitled "Rising Prices and Purchasing Power Policies," and in a lengthy article that he published later in 1913 (in German) entitled, "The General Rise in Prices in the Light of Economic Theory."

(Both of these articles will appear in English in Volume 1 of "Selected Writings of Ludwig von Mises," of which I'm the editor, from Liberty Fund in 2011. The volume also includes most of Mises' writings on Austro-Hungarian monetary and fiscal policy from the period before 1914, as well as some of his writings on these and related themes during and immediately after the First World War.)

However, by the early 1920s, his experience with and interpretation about government monetary policy (especially from his studies of Austrian monetary policy before the First World War, and the monetary events during and after the First World War in Europe in general) led him to believe that there was a stronger "activist" and often ideological factor behind the initiation and maintaining of below-market rates of interest and expansions of money and credit than "just" passive accommodation by the banking system to changing demands for investment borrowing due to the elasticity of fiduciary media.

But he never completely disavowed his belief that even in an environment free from political and ideological control and influence over the monetary system that economy-wide fluctuations were possible under certain particular circumstances.

But due to what he considered the "actively manipulative" direction of monetary policy before and after the Second World War, and therefore his tendency to critically challenge this development in monetary theory and practice, he rarely referred to economy-wide fluctuation potentials in a more fully free market monetary order in later decades since it seemed of little relevance in a policy context in which government activism was dominating the economic policy setting.

Richard Ebeling

[Current]

> Indeed, this actually led Mises to also formulate a
> form of a "cost-push" theory of inflation that is
> briefly discussed in a subsection of the 1924 second
> edition of "The Theory of Money and Credit" (see, Part
> 2, Chapter 8, subsection 14 of the English-language
> version of Mises' book.)

It's in Part 2, Chapter 2, subsection 14.

[Peter Lewin]

@Jerry: The additions to "people should know better" are very nice complements to what I said. Of particular note is the connection between the dotcom bust and the move to the housing market (spurred no doubt by the plethora of specific regulations encouraging home ownership and speculation in houses). This is not scientific evidence, but I remember clearly the many assertions that the housing market is not like the stock market - "we are in a different world now."

@Richard: I am thinking that over the course of the 20th century the monetary mismanagement component of cycles became more significant, hence Mises shifted his emphasis.

Nevertheless, I would still claim a large role for technology (aspects of a real business cycle?) that sets the stage for inappropriate central bank responses and the credit-induced cycle on top of "normal" fluctuations. The latter looms so large relative to the former that it dominates our analysis - rightly so.

[Richard Ebeling]

Peter Lewin:

The influence of technological and other innovative demands for credit is, of course, at the heart of Joseph Schumpeter's theory of "cyclical" change, as he developed it in his "The Theory of Economic Development" (1911).

Schumpeter not only considered it likely, but essential for the banking system to show such credit-creating accommodation if economic change was to be financed.

In my recent book, "Political Economy, Public Policy, and Monetary Economics: Ludwig von Mises and the Austrian Tradition" (Routledge, 2010), I have a chapter in which I compare Mises and Schumpeter on the causes and sequence of the cycle, Chapter 8: 'Two Variations on the Austrian Monetary Theme: Ludwig von Mises and Joseph A. Schumpeter on the Business Cycle,' pp. 273-301.

Richard Ebeling

[Bill Woolsey]

While I can't speak for Steve, I saw nothing in the comments that was inconsistent with his intitial statement that these cycles (the booms) involve an increase in the natural interest rate and the banking system keeps the market interest rate from rising.

The process doesn't require an increase in the quantity of base money and instead could occur by greater incentives to economize on reserve holdings by banks in response to higher market rates. It is even possible that it could be associated with no increase in the quantity of money issued by banks (I worry about some of the discussion above, particularly Caldwell on Hayek that looked a bit like money and credit are confused,) but that would require a lower demand to hold money by firms and households, perhaps motivated by higher expected profits.

And suppose money expenditures on current output is kept on a stable growth path? Can one of these endogenous cycles start? And how much do they matter?

[Dave Prychitko]

Peter:

I say in my paper: "But if the hypothesis is supposed to explain business cycles, can Wray and other followers of Minsky find examples of macroeconomic booms and busts that were not caused by monetary disequilibrium?" What I had in mind was under the current monetary institutions (as in Mises above, discussed by Richard).

Are you suggesting that they (and Austrians, too)can find empirical examples? If so, my position in the concluding section of my paper is false. Moreover, it does make our theory of the trade cycle much more Post Keynesian than even I let on.

[Pietro M.]

Hayek's argument in MTTC is that in case of a positive productivity shock (an increase in the natural rate of interest) which causes an increased demand for savings, banks may be unwilling to raise the loan interest rate in order to increase savings or starve the now submarginal investment plans of funds.

If they don't raise the interest rate, they don't signal the higher demand for funds to the rest of the economic system, causing a coordination problem.

A perfect banking system should set the price of capital to a level that takes into account the larger demand for savings due to the new investment opportunities, and eventual variations of the total supply of credit due to changes in the income structure caused by the innovation itself (will the new income be saved or consumed? This effect is likely to be negligible at least for small innovations).

Because there is no market mechanism assuring this equilibrium, Hayek claims that crises are unavoidable.

In Prices and Production he is explicit in the fact that not all increases in the supply of money are dangerous: he claims that it is impossible to distinguish, however, an increase in the demand for money stimulated by malinvestment and a genuine increase in the demand for money.

He also adds, however, that the persistence and extent of these cycles may not be very large on the unhampered market, if I remember well, so that the Misesian "exogenous" explanation of cycles as the result of an inflationist ideology may be a more relevant approximation, which is likely to hold almost always, and to better explain the persistence and severity of real world business cycles.

Both Mises (in "Causes of the economic crisis") and Hayek (in "Monetary nationalism...") explicitly said that moral hazard due to suspensions of convertibility and liquidity creation in times of crisis by the central bank were an important element in causing business cycles.

Although I believe it is true that there is no direct immediate market feedback regarding the sustainability of investment plans - because what should be estimated by market agents is the future availability of savings relative to the demand for complementary capital goods - it is likely that, absent policy moral hazard, banks would learn how to cope with the problem in an "ecologically rational" way.

However, occasional deviations are unavoidable because of the maturity mismatch inherent in banking, where investments are decided before saving decisions are made. Although it is likely to be a small problem in the free market, but moral hazard is probably the engine of monetary non-neutrality in the present world.

PS Congratulations to Prof Lewin! It's a very interesting post and touches very important points.

[Bill Woolsey]

By the way, I define the natural interest rate as the one that coordinates saving and investment at full employment of resources. The natural interest rate that would exist if everyone had perfect forsight of real investment returns (and maybe lots of other things too) really plays no role in my thinking.

Anyway, if there really are better returns to investment, and this is recognized to some degree, I would think that market response "should" be a an increased demand for investment--capital goods, a higher natural interest rate, a larger quantity of labor demanded and labor input today, a lower quantity of consumption demanded, and more resources available for the production of capital goods. More labor means the production of capital goods can expand more than the contraction in the production of consumer goods. Real income and real output are higher. And, in the end, the payoff of these greater returns--even more output and income allowing for more consumption.

If, on the other hand, the returns were an illusion, then extra output, that final boost to output, income, and consumption never materialize. The leisure and other consumer goods that were sacrificed come out of the pockets of those who pursued these imaginary returns. Their net worth falls. If there was not enough net worth, then bankruptcies, and lenders share the losses.

Presumably, the lower investment demand results in a lower natural interest rate and a higher quantity demanded of consumer goods. Real wages fall and labor input decreases and leisure returns to normal. etc.

Obviously, there are complications, but something like the notion that there are really good returns only sometimes motivate the real business cycle theories, and the notion that the returns are an illusion is something like the Minsky theory.

Think about all of this in the context of stable growth of money expenditures. (Don't let the real or imaginary prospective returns create an excess supply of money and growing money expenditures on current output, or the reverse when there are no more great returns or else the past great returns turn out to be imaginary. No excess demand for money. Money expenditures remain on the same growth path.)

I think by far the best environment for this fluctuation to occur is in the context of stable growth of money expenditures. The illusionary one, is a mistake, of course. Better if people don't devote resources to capture returns that don't exist. However, I think it could use more analysis.

Do real wages in the boom solely rise through decreases in the prices of consumer goods? Do money wages rise more than usual, even given contant growth of nominal incomes and higher interest rates?

In the bust, could money wages actually need to fall? Even if real returns on capital are lower? Is it all a higher price level?

I really wish that more Austrian economists, particlarly those with a good grasp of monetary equilibrium, would think about these sorts of things more, and less about what happens when the quantity of money rises given the demand for money, the supply of saving, and the demand for investment.

P.S.

If you assume a central bank is maintaining a nominal interest rate target through all of some "real" boom-bust as above, the monetary disequlibrium and fluctuations in money expenditures, and the needed fluctuations in the price level and money wages would be extreme. And that is what is going on with Minsky/Keynesians.

[Bill Woolsey]

Obviously, careful study of Hayek's work from the thirties isn't my strong suit. However, I think the conventional language of an increase in investment demand, and a higher interest rate leading to an increase in the quantity of saving supplied (and reduction in the quantity of consumption demanded), is better than speaking of an increase in the demand for saving.

If you assume the supply of saving is perfectly inelastic with regards to the interest rate, then higher investment demand just raises the interest rate, at least, if there is no excess supply of money created by banks distorting the process.

If you assume that the supply of labor is perfectly inelastic with respect to the real wage, then higher expected investment returns won't result in more labor input.

Once you relax those assumptions, then expected returns, even illusionary ones, will result in higher employment and output and more capital accumulation. The illusionary ones will turn out bad for those pursuing them.

Is it the role of a monetary authority (or monetary institutions) to create an excess demand for money when wrong expections generate a boom? I don't think so.

And does maintaning money expenditures even when lots of business suffer losses create moral hazard? I don't think so. Letting those that made bad investments lose money in the context of growing money expenditures just about right, and that losses could be magnified for everyone because of a recession, while dampening investment demand, true, it would be an excessive dampening.

[Peter Lewin]

@Dave: Well this is an interesting question. I wish I were in a position to answer it authoritatively. At least in the 20th century I think it is probably very difficult to find cycle episodes that were not accompanied by monetary mismanagement. The problem of course is to know what would have happened absent the latter. I believe the technology surges of the dotcom period and of the 1920's would have produced a cycle - of much diminished amplitude had monetary policy been more "neutral" - more passive (or, preferably, if we had had a free banking system).

The (Post) Keynesians would probably respond to your question by shifting the burden of proof. "Can you prove that the cycle would not have occurred without the monetary policy that we observed, and that, indeed, might not have been worse?" The power and shelter of counterfactuals.

I do not think this diminishes the validity of the Austrian story at all, but it makes one aware of how carefully it has to be explained. It is so easy to construct an ABCT version as a straw man - one that denies the existence of any cycles without a mischievous central bank.

I am also not denying that central banks can, and probably have, precipitated cycles from nothing. Only that this is not necessary and probably not descriptive of many (most) real world episodes. Your excellent article clarifies much of this.

[Bogdan Enache]

The defense and restatement of the Austrian business cycle theory, in it's classical form, ultimately depends on a very difficult act of integrating real and monetary factors into a comprehensive analysis by way of capital theory. Absent this very difficult capital-based integrated analysis, a purely monetary or a purely real business cycle theory will always prove more instrumental and more persuasive in analysing this kind of phenomena whatever the context.

[Jerry O'Driscoll]

Bogdan,

My response is "coordination problem."