September 2014

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

« Mario Rizzo: Born-Again Macroeconomist | Main | Ad Against Deficit Spending »


TrackBack URL for this entry:

Listed below are links to weblogs that reference Why Say's Law Does Not Preclude Gluts and Shortages:


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

I agreed until Steve got to "sticky prices." The way prices work is by being "sticky." Alchian and Leijonhufvud were excellent on the issue.

I don't know what the maximum flexibility of prices means. Define, please. Prices CANNOT be infitinely flexible, and it would be undesirable if they were.

The problem with Say's Law have exactly nothing to do with price flexibility. Nothing.

I agree with the monetary equilibrium point. But, like Jerry, I have problems with the price stickiness argument.

Sounds right, like a coordination issue. Bad economic symptoms would appear as a noticeable build up of inventory.

But, I don't buy the idea of systematic banking error such that the economy would be fooled into mal-production for any length of time. The effect of out of balance production is very visible to us, in my theory, more visible than price or interest rates

The systematic error of central banking is the constant playing of catch up with the economic average, the lag effects of having to estimate an aggregate. But they never systematically fall farther and farther behind, except to the extent global trade is advancing, the bigger aggregate leading to longer central bank lags.

This recession is the real economy, and there is a correlated shock, but it isn't banking, in my theory.

jerry says: "The problem with Say's Law have exactly nothing to do with price flexibility. Nothing."

If prices weren't sticky, real balances could adjust immediately to any level of aggregate supply, that is, aggregate supply equals aggregate demand always and by necessity. That would be Say's identity.

... and there would be no scope for monetary policy (like stabilizing Mv).

I agree with Jerry and Mario, and after just going through this material for teaching my graduate class I have to say that I think the way you and George have put this doesn't clarify the issues. Yes money matters, I am with you. But take a look at Blaug's chapters (5th Ed) on Say, classical monetary theory and neoclassical monetary theory and the original text and the context of the debates. And don't forget to read Say and Malthus. Your meta interpretation is right I'd say, but I disagree with your specific presentation points.

As far as I remember Blaug says that classical economists had Say's Equility in mind (an equilibrium condition) and not Say's Identity (Lange's interpretation). All who argue that classicals opposed the argument that stagnating aggregate demand will cause long-run limits to growth are right. Today, no serious economist believes this. Thus, long-run versions of Say's Law are uncontroversial today. Short-run versions of Say's law need the qualifications given by George and Steven:

#case 1: flexible prices - Say's Identity in long run and short run. No need for monetary policy since money always neutral.

#case 2: sticky prices - Say's Equality holds, that is, AD=AS in the long-run but not in the short run. system converges by deltaP. Since goods are traded for money and not for goods (Clower constraint), deltaP may cause a lot of adjustment costs + may induce noise in coordination process guided by relative prices. Thus, better performance by targeting Mv to keep AD constant. If deltaY > or < 0, targeting AD-level allows deltaP to signal changes in supply-side conditions.

This is how I understand Steven and George. To what extent does Blaug differ?


A quick response:

I think Arash is largely right. If you read Say, he comes close to assuming perfectly flexible prices, as that's the only way one can make sense of his claim that a "shortage of money" can never be the cause of "dull sales." If I wrote that chapter today, I probably wouldn't use the term "sticky prices" but rather just point out that prices don't adjust perfectly and instantly to monetary disequilibria.

Put differently, I don't think one can hold the following two ideas consistently:

1. Say's Law does not preclude general gluts when you consider the possibility of monetary disequilibrium.

2. Number 1 does not require imperfectly flexible prices.

If prices were perfectly flexible, why would an excess demand for money lead to a general glut? Wouldn't prices "just fall" (Rothbardian style) to clear out the potential glut before it materialized? How can you explain disequilibria in all of the individual product markets in the face of falling demand due to excess demands for money without arguing that prices are imperfectly flexible because sellers have go through a discovery process to figure out that prices need to be lower?

What Steve needs to mean is the tautology that prices must have that degree of flexibility such that excess supplies (demands) will be quickly eliminated. If prices were "infinitely" flexible (whatever that means) no one could gain arbitrage profit from adjusting them. So they wouldn't move or be infinitely flexible.

Somehow I feel that Say's Law isn't the area to develop this particular discussion.

But I basically agree with Steve.

As far as I understand it all economists before 1900 accepted that prices were not perfectly flexible. So, to say "sticky prices" just says the same thing earlier economists meant by "prices". Since flexible prices are impossible it would be nice to do something about that naming, and return to the old way of looking at things. But I think it's important to always keep in mind the problem, over on the LvMI site there are a few folks who argue like New Classicals when it suits them.

In his explanation Steve *begins* with a change in the relative demand for money and for all other goods. I think this is important. The problem is that there are two schools of thought about how money impacts Say's law. One is that the beginning of any recession must involve a relative supply problem amongst goods. The other is that the beginning must involve a relative supply problem amongst everything traded, including money. Think about Higgs' ideas for example. If you believe Higgs (and I certainly do) then the sounds that come out of Ben Bernanke's mouth can cause relative shortages and surpluses immediately. And, that includes the possibility that the demand for money will increase and the demand for everything else will fall.

This is important because many 100% reserve supporters think that if 100% reserves are introduced then no recessions could begin. They believe that no large relative shortages and surpluses could occur endogenously. That may be true in a long term libertarian society. But, in our existing world shocks come from external crises and external and internal politics.

Arash, I don't think your case 2 really follows directly from the rest. Lot's of other theories are needed to get there.


then you also may also agree with George: since prices cannot be infintely flexible, general gluts must occur in light of shocks to money demand, if monetary policy remains passive in respect to effective money supply. Since such short-run general gluts are meaningful, so must be the concepts of "aggregate demand" and "aggregate supply" (general glut = AS>AD in real terms).


I think it's a bit more complicated than that. Agreeing with Say's law in this form brings into existence Mises "money relation" : "the demand for money compared to the stock of it". But, if we want to talk about "aggregate demand" in the normal sense then we need a theory of how trade of GDP constituents relate to trade for money in general.

Suppose an economy with constant nominal income. In this case, the distribution of nominal spending on alternative goods and services is a zero-sum game. Further suppose there are two classes of goods and services: P and C. (Hereafter, 'spending' and 'income' are nominal).

Suppose that spending on P increases; it follows that spending on C decreases. Since spending is equal to income, the initial consequence will be a rise in income to producers of P and a decline of income to producers of C. This should prompt relative price changes, and a redistribution of resources toward producing more P and less C. Let changing price signals occur with falling wages in industries that produce C and rising wages in industries that produce P.

But suppose that prices are more "sticky" (or stubborn, insensitive, inflexible, etc.) downwards than upwards. Why are prices more sticky downwards? Unions, governments, contracts, or just the money illusion. I do not think the particulars matter so much in this context.

As wages in industries that produce P begin rising, they do not decline by a corresponding amount in industries that produce C. The change in relative prices that is supposed to reallocate resources is frustrated, because the economies' price changes are lop-sided. Eventually this disequilibrium will be smoothed out as the force on downwardly sticky prices becomes too great. (Such as businesses with strong unions becoming bankrupt).

This is one of the reasons I am sympathetic to something like a 3% NGDP level target as a least bad option for monetary policy. Gradually rising nominal spending can allow real wage declines in C to occur despite nominal rigidities, and perhaps create less disturbance to the price system than a NGDP stabilisation target. But I digress.

In the example above, nominal income remained constant, so any fall in income in one part of the economy is offset by a rise in another. It seems quite trivial to extend this analysis to a case where nominal income declines for the economy as a whole. The same nominal rigidities, or 'sticky prices', would contribute to a general disequilibrium or 'general glut'.

This is what I understand "sticky prices" to be. Clearly, it is not simple to identify when price stickiness is frustrating economic coordination. Prices may appear sticky for all kinds of reasons, but actually be better coordinating the allocation of resources.


I think Mario is right (and Current as well) you (and George) are painting w too broad of brush strokes, and also I would add not giving appropriate weight to the intellectual context within which positions were laid out. Again I think my Say's Law and Quantity theory frame provides a better interpretive frame from which to make your point.

I hate to get "narrow" in my focus. However, the issue, for me, is whether price (in)flexibility has much to do with *Say's Law*. Frank Taussig's pre-Keynesian understanding of Say's Law is that if we see lots of markets out of equilibrium it is because there has been a breakdown in "the machinery of exchange" and not a "permanent or deep-seated difficulties of finding an extensible or profitable market."

And why are we focusing on gluts when discussing "price inflexibility"? It is true that under certain conditions we can get a scenario -- intereacting with the demand for money -- that will produce widespread gluts. But if we are talking about price inflexibility per se then excess demands will result too.

My general point is: Let's not try to discuss everything at once.

The same old Keynesian and Banking School theory about the "excessive demand for money", "shortage of money", "monetary disequilibria" caused by "hoarding" etc, refuted time and again but, like any other theoretical zombie, refuses to die (I will not even comment on the 'sticky prices' "argument").

Here is maybe the best and most concise refutation of this "hoarding-shortage of money" fallacy:

"There is no fraction of time in between in which the money is not a part of an individual’s or a firm’s cash holding, but just in “circulation. It is unsound to distinguish between circulating and idle money. It is no less faulty to distinguish between circulating money and hoarded money. What is called hoarding is a height of cash holding which — according to the personal opinion of an observer—exceeds what is deemed normal and adequate. However, hoarding is cash holding. Hoarded money is still money and it serves in the hoards the same purposes which it serves in cash holdings called normal. He who hoards money believes that some special conditions make it expedient to accumulate a cash holding which exceeds the amount he himself would keep under different conditions, or other people keep, or an economist censuring his action considers appropriate." (Ludwig von Mises, Human Action. pp. 402-403)

You guys, as "experts", obviously know better than the very people in question who keep the cash balances how large these balances should be, i.e. how large a portion of his money income every particular individual should divert to the protection against uncertainty, as opposed to spending and saving. And consequently, you feel quite comfortable in the role of the economists "censuring" the other people's actions and determining what in their actions is "appropriate" and what is not.

It is then really no surprise that Selgin thinks that a Keynesian basically does not have many reasons to object to his monetary theory.

Lee Kelly,

I don't really think that a NGDP+X% target can really help with what you describe. The problem is that the situation you describe is essentially long term. We suppose that there is some regime of monetary policy in place and then the change in relative prices occurs. Here we should remember Mises' discussions about Keynes' view on Union and government induced wage inflexibility. If the response to it is to move into a monetary policy regime where consumer price inflation is generally around X% then unions will move their goalposts accordingly. They will consider anything less than an X% wage rise to be a wage cut, and they'll be right. During the period of change between a lower level of price inflation and a higher one the unemployment rate may go down and output may go up. In that case unionised employees gain at the expense of everyone else. But, that situation won't be persistent afterwards. Another rise in the generally expected inflation rate would be needed in the future to create the same effect. This is grabbing the Tiger by the tail, it's essentially the Fisher effect applied to other sorts of contract.

It is quite clear that Nikolaj has no idea what is meant by the expression "an excess demand for money." That last comment sounds like he's been studying under the brilliance of one DG Lesvic.

@Jerry and Mario: were I to write that today, I probably wouldn't use the language of "sticky prices." I'd probably just make the point that at as long as prices don't adjust perfectly, money matters.

@Pete: I have/did read Say when writing that chapter, so I'm not concerned that I've done violence to his big picture argument.

However, I do think Say implicitly assumed something close to perfect price flexibility when he argues that a shortage of money can't be the cause of "dull sales." The only way that can be true is if one assumes that prices can adjust instantly and proportionately in the face of excess demands for money.

In other words, I don't think one can hold the following two positions consistently:

1. Say's Law does not preclude general gluts if one accounts for the possibility of monetary disequilibrium.

2. The degree of price flexibility doesn't matter for Say's Law.

How else can you explain how the falling demands for goods and services across the economy due to excess demands for money can lead to excess supplies of those goods and services unless you view sellers as engaged in a time-consuming process of discovery to find more profitable prices for their goods?

@Mario: fair enough on the "everything at once" point, but if we ARE trying to explain how the existence of general gluts is still consistent with the truth of Say's Law, I don't know how we can avoid the excess demand for money scenario, which in turn requires that we make note of the fact that price adjustments take time.

Steve is right that Say comes close to claiming that money demand doesn't matter in "Treatise on Political Economy". He wrote:

"Thus, to say that sales are dull, owing to the scarcity of money, is to mistake the means for the cause; an error that proceeds from the circumstance, that almost all produce is in the first instance exchanged for money, before it is ultimately converted into other produce: and the commodity, which recurs so repeatedly in use, appears to vulgar apprehensions the most important of commodities, and the end and object of all transactions, whereas it is only the medium. Sales cannot be said to be dull because money is scarce, but because other products are so. There is always money enough to conduct the circulation and mutual interchange of other values, when those values really exist."

This is essentially Rothbard's line: any amount of money satisfies demand in the long *and* short terms. But, then he goes on to discuss how endogenous money production could avert and problem:

"Should the increase of traffic require more money to facilitate it, the want is easily supplied, and is a strong indication of prosperity—a proof that a great abundance of values has been created, which it is wished to exchange for other values. In such cases, merchants know well enough how to find substitutes for the product serving as the medium of exchange or money:*35 and money itself soon pours in, for this reason, that all produce naturally gravitates to that place where it is most in demand. It is a good sign when the business is too great for the money; just in the same way as it is a good sign when the goods are too plentiful for the warehouses."
Note 35 says:
"By bills at sight, or after date, bank-notes, running-credits, write-offs, &c. as at London and Amsterdam. "

However, someone told me that elsewhere Say is more careful.

I agree with Mario that too many issues are on the table at once. But "disproportionality" has been forgotten, which was the heart of Say's Law (see Mario's original post).

The problem with Say's Law is that money is an independent source of demand. There can an excess demand for some goods without an excess supply of others. Hayek states this most clearly in Monetary Theory and the Trade Cycle.

Prices adjust to the excess demand in a normal market fashion. But price signals now no longer reflect real factors only. "Distorted" not "sticky" prices are the problem. Relative prices are inconsistent with real factors. The price level is not necessrily wrong.

So with elastic credit, you keep the disproportionality part of Say's Law. There is not a general excess demand for "goods," but a change in the composition of demand. An excess supply of money working through the banking system interferes with equilibration. Steve Horwitz has a great article exactly on that issue.

Issues like central banking vs. free banking are clearly relevant, but, I think, separate issues. Sticky prices are also a separate and related issue. Without a common text, i.e., Alchian/Leijonhufvud, however, the discussion will be at cross purposes.

I will only add that nothing I just wrote is peculiarly Austrian. In recent years,Leijonhufvud has been telling a similar story using Wicksell. And, of course, he told it using his interpretation of Keynes. Not to mention classical economists when they discussed monetary disequilibrium. (Mises always insisted his theory was classical.)

Say's Law with modern money is a coordination story with disproportionalities. It still by itself doesn't get you general gluts.

The article by Steve that I had in mind is his 2003 paper in RAE on "The Costs of Inflation Revisited." He makes the case that, in modern banking systems, it is inevitable that monetary shocks distort relative prices and resource allocation.


I think that your and Selgin's points about money are not so original or subtle to create a serious confusion on anyone's part: supply and demand must be "matched", money must be "right", and in order to be "right" the supply of credit must be "elastic", if money is too "rigid", then there is a fatal "mismatch" where demand "exceeds" supply and so on and so forth. The same old, Banking School, mercantilist, Keynesian inflationism, coupled with the Fisher-Friedman crude, 'macroeconomic' quantity theory of money (MV etc) Mises and Hayek exploded time and again. Not much misunderstanding here.

I'm confused- is "always in equilibrium" economics being invoked to make these points? What I mean to ask is that the degree of entrepreneurial alertness is just as important as the degree of perversion which results from the veil of money.

Does even the none-caricatured Say's Law assume the ERE?

Interesting, so an oversupply of money can cause, not only inflation, but shortages of goods and services.

That is reminiscent of the late 1970's.

Current writes: "Should the increase of traffic require more money to facilitate it, the want is easily supplied, and is a strong indication of prosperity—a proof that a great abundance of values has been created, which it is wished to exchange for other values. In such cases, merchants know well enough how to find substitutes for the product serving as the medium of exchange or money:"

I wonder how well that fits into the history of money, as when in the 19th century, the money supply was controlled by gold strikes and new sources of bouillon.


When you interpret "excess demand for money" to mean that the observing economist is saying that people are holding "too much" money (i.e. "hoarding"), you demonstrate utter ignorance of what that phrase means.

The notion of an excess demand for money is totally subjectivist. All it means is that individuals' desired money balances *as judged by them* (that's subjectivism right there) are greater than their actual money balances. And George and I don't claim any great originality. What's causing your confusion is your ignorance of monetary theory, including Mises's.

After all, it was Mises who gave us the cash balance approach to the demand for money and the notion of an excess demand for money is completely consistent with that view. Nothing in your earlier quote from Mises about hoarding is germane to the excess demand for money concept. Nothing.

Mises uses examples of people who wish to increase their cash holdings. Those people have an excess demand for money - their desired cash holdings at the current price level are greater than their actual holdings. THAT is what is meant by "an excess demand for money."

Seriously, if you want to sit at the table with the grown ups, you really should know what we're talking about. Your attempt to change the subject in your most recent comment doesn't help persuade me you know what you're talking about.


how can someone ever be in favor of the banking school AND be a proponent of Fisher-Friedman type of QT? What in Steven's or George's comments makes you believe that they favor the banking school? Why do you invoke excess aggregate demand in a discussion on general gluts and the role of price rigidity?


The relevant explanation for sticky prices in Alchian (1969) is there are three ways to adjust to unanticipated demand fluctuations:
• output adjustments;
• price adjustments; and
• inventories and queues (including reservations).

There is no reason for price changes to be used regardless of the relative cost of these other options:
• The cost of output adjustment stems from the fact that marginal costs rise with output.
• The cost of price adjustment arises because uncertain prices and wages induce costly search by buyers and sellers seeking the best offer.
• The third method of adjustment has holding and queuing costs.

There is a tendency for unpredicted price and wage changes to induce costly additional search. Long-term contracts including implicit contracts arise to share risks and curb opportunism over sunken investments in relationship-specific capital. These factors lead to queues, unemployment, spare capacity, layoffs, shortages, inventories and non-price rationing in conjunction with wage stability.

Walter Oi has also written on slack capacity as being productive and he included references back to W.H. Hutt.

Oi’s work on retailing and supermarkets spends a lot of time explaining how an empty store is efficient because the owners are waiting for a mass of customers to arrive at unpredictable time. Oi redeveloped the term the economies of massed reserves to describe this. He through this was a better term than Hutt’s pseudo-idleness.

Oi argued that all resource idleness could, in principle, be eliminated, but, to accomplish this, the synchronization of the arrival rates of customers, clerks, and just-in-time inventories would be prohibitively expensive.

Robert Shimar and James Hamilton both wrote on rest unemployment where laid-off workers with industry and occupation specific human capital investing in waiting for conditions to pick up in a depressed sector or industry because moving to another sector required extensive retraining and scraping of human capital. Rest unemployment is a cousin of recall unemployment.

Benjamin Klein also wrote a nice short paper in the AER in the early 1980s using hold-up of specialised assets to explain rigid wages. He drew a parallel with an exclusive dealing contact in franchising.


they accept the elastic money supply doctrine, which is part and parcel of the Banking School's theory, the idea that the credit supply should fluctuate in order to match the "needs of trade" (although they usually don't use that term). As for the QT, they accept that stabilizing MV during the recession is an appropriate monetary policy. Asked them - they repeated that 200 times already. This is a typical stabilizationist doctrine, actually monetarism. Steve and George Selgin agree with Friedman that one of the main reasons for the Great Depression was a failure of Fed to sufficiently inflate the money supply in order to stabilize MV and avoid the "dangerous deflation". Steve thinks seriously that Forgotten Depression of 1921 would be much milder if Fed inflated more to prevent deflation.


get serious, please. It is plain as a day that you and George Selgin advocate stabilization of MV, in other words, advocate a doctrine that the solution for the "monetary disequilibrium" is not to allow the increase of the purchasing power of money, as Mises thought, but to increase the credit supply to accommodate the increased demand for money.

You consider hoarding a problem which needs to be addressed by the bank credit expansion, while Mises thinks that this "solution" would only worsen the problem. Your solution is based on a simple misconception, so typical for all the inflationist schools, that "hoarding" is unproductive use of money and a reason to increase inflation. And the quotation I provided directly and clearly exposes this basic error of yours.

In order to be able to characterize the Mises's approach as "consistent" with yours you would have to find where Mises said that in the recession, or in deflation, increasing the credit supply could be deemed desirable. To find i a word something consistent with your slogan "I will let people use their money as they see fit, provided the MV is stable". I challenge you to do that.

"I will let people use their money as they see fit, provided the MV is stable".

This is the sort of gross mischaracterization of my views that I've come to expect from you.

In the meantime, consider the following as evidence for the consistency of Mises's views with mine (and George's):

Salerno challenged me to find Mises making the following argument: "2. in the absence of the creation of fiduciary media, an increase in the demand for “inside” (bank notes and deposits) money will lead to a rise in the loan rate of interest above the natural rate of interest thus causing a depression (or conversely, a fall in the demand for money unmatched by a contraction of fiduciary media will depress the loan rate below the natural rate and precipitate a business cycle);"

Horwitz: In Human Action (p. 547), Mises writes:

“Cash-induced changes in the money relation can under certain circumstances affect the loan market before they affect the prices of commodities and labor. The increase or decrease in the supply of money (in the broader sense [which includes fiduciary media – SH]) can increase or decrease the supply of money offered on the loan market and thereby lower or raise the gross market rate of interest although no change in the rate of originary interest has taken place. If this happens, the market rate deviates from the height which the state of originary interest and the supply of capital goods available for production would require. Then the market rate of interest fails to fulfill the function it plays in guiding entrepreneurial decisions.”

Although my langauge is different, it might be worth comparing that passage with something from my 2000 book in which I’m discussing the way in which inflation and deflation cause the market rate to deviate from the natural rate in the way Joe’s point 2 describes:

“[M]onetary equilibrium is that situation where relative price signals, particularly intertemporal ones [i.e. interest rates – SH], are accurate enough to allow entrepreneurs to create a potentially sustainable capital structure. … One of the primary effects of both inflation and deflation is to distort the price signals that lead to the integration of the intertemporal structure of production… A systematic unsustainability in the capital structure suggests that the ruling market rate of interest is not equal to the natural rate.” (Horwitz 2000: 82)

I submit that my argument is identical in its essentials to Mises’s.

Furthermore on pages 548-50, Mises has a discussion of the effects of inflation upon originary interest. He concludes by saying “The same is valid, with the necessary changes, with regard to the analogous consequences and effects of a deflationist or restrictionist movement.” That certainly suggests strongly that Mises saw deflationary process as reversing all the relationships of the inflationary one, including the relationship between the market and natural rate.

And on p. 566-7 (my emphasis) of Human Action, Mises discusses deflation explicitly:

“In all of these cases [the ways in which deflation might happen] a temporary tendency toward a rise in the gross market rate of interest ensues. Projects which would have appeared profitable before appear so no longer. A tendency develops toward a fall in the prices of factors of production and later toward a fall in the prices of consumers’ goods also. Business becomes slack. The deadlock ceases only when prices and wage rates are by and large adjusted to the new money relation. Then the loan market too adapts itself to the new state of affairs, and the gross market rate of interest is no longer disarranged by a shortage of money offered for advances. Thus a cash-induced rise in the gross market rate of interest produces a temporary stagnation of business. Deflation and credit contraction no less than inflation and credit expansion are elements disarranging the smooth course of economic activities.”

Mises then goes on to explain why deflation and inflation are not “simply counterparts,” a discussion that I largely agree with as he points out that deflation is generally not as destructive as inflation.

It’s also worth noting the italicized sentence, which suggests that Mises saw that deflation led to a downward “tendency” on prices and wages but that there was first a “deadlock” that only ceased after some period of adjustment. This is more or less exactly what monetary equilibrium folks have been arguing about the imperfect flexibility of prices during a monetary deflation.

The bottom line is that Mises did recognize that a monetary deflation (understood as a supply of money less than the demand to hold it, as he defined it in TTMOAC) could drive the market rate above the natural rate and lead to a recession. I think that addresses Joe's proposition.

Whole discussion here:


I think you nailed that one. Good job.

JIm Rose,

Thanks for jumping in with a discussion of Alchian. Just to reiterate and elaborate on what you wrote.

Except in auction markets, prices do not and ought not adjust constantly to ever changing and shifting demand. That would be very costly to all concerned. Posted prices provide valuable information to buyers, and that information is degraded by frequent changes.

Moreover, what tells a seller to change his price? Inventory (quantity) changes. Sellers don't know the correct price and are constantly searching for information about it. Inventory changes are one source of information. Cost changes are another.

Markets handle quantity adjustments every day. Why normal quantity adjustments morph into coordination failures requires something more than less-than-infinitely flexible prices.

Leijonhufvud used Alchian to make sense of Keynes. If sticky prices were Keynes' contribution to economics, then Keynes made no contribution. Pre-Keynesian economists knew prices did not move instantaneously with every shift in demand. To make sticky prices (or wages) "Keynesian" is to trivialize Keynes' work.

> I'm confused- is "always in equilibrium" economics
> being invoked to make these points? What I mean to
> ask is that the degree of entrepreneurial alertness
> is just as important as the degree of perversion
> which results from the veil of money.
> Does even the none-caricatured Say's Law assume the
> ERE?

This is what Jerry O'Driscoll was getting at earlier. Say's law doesn't really require that entrepreneurs are superhuman. It just requires that they can recognise what's in their interest reasonably quickly. As I see it, the point is: from the local perspective of each entrepreneur the adjustment he or she has to make is like the normal sort of adjustment made when a relative price that is relevant to their business changes. The difference for macroeconomics is that there are many entrepreneurs making these changes at once.

This is shows the problem with thinking of Say's law as being like Walras' law. That immediately brings up Walras' stylised picture of the economy.

In the debate with Nikolaj there is a typical confusion that comes from the 100% reserve side. There are two distinct aspects to holding money. There is the question on the one hand if it is good for the person holding it. We can't doubt that it is. Since it provides the person holding it with a service then, in a free-market economy, it thereby provides society with a service. But, by saying this we haven't proven that a changing demand for money has no negative external effects and cannot cause lost productivity. In fact, it's not the actual holding of money that's the issue, it's the excess demand as Steve said that becomes an issue if there is no market to match it with a rise or fall in supply. That's because that demand brings about the need for a great many price changes.

This is very different from the Keynesian view. To Keynesians money doesn't supply a psychic yield, or at least, holding it provides nothing to wider society. They don't understand that by providing ourselves with this psychic benefit we assist ourselves and consequently everyone else.

Keynesians only see the productivity issue caused by changes in demand, Rothbardians only see the psychic yield issue, we recognise both.

Rothbardians are fond of saying that any amount of money can provide the service of money. Certainly this is true in the long run, but it doesn't mean that the economy can quickly adjust to any particular amount of money. Just like, it doesn't matter in the long run what my phone number or my friends phone numbers are. But, if the phone company changed them all every day then that certainly would start to matter.


you've got nothing still. Where is the claim in Human Action that the appropriate "cure" for deflation is to increase the credit supply? Deflation could be temporarily "destructive" in a sense that it could lead to even more liquidation than the "equilibrium" would require. Mises agrees with that (obvious) point. But, guess what, he says that is "unavoidable".

You are trying to obfuscate the obvious point: Mises thought that the very notion of "right quantity of money" as Selgin charmingly says was a pure nonsense:

"As the operation of the market tends to determine the final state of money’s purchasing power at a height at which the supply of and the demand for money coincide, there can never be an excess or a deficiency of money. Each individual and all individuals together always enjoy fully the advantages which they can derive from indirect exchange and the use of money, no matter whether the total quantity of money is great or small. changes in money’s purchasing power generate changes in the disposition of wealth among the various members of society. From the point of view of people eager to be enriched by such changes, the supply of money may be called insufficient or excessive, and the appetite for such gains may result in policies designed to bring about cash-induced alterations in purchasing power. However, the services which money renders can be neither improved nor repaired by changing the supply of money. There may appear an excess or a deficiency of money in an individual’s cash holding. But such a condition can be remedied by increasing or decreasing consumption or investment. (Of course, one must not fall prey to the popular confusion between the demand for money for cash holding and the appetite for more wealth.) The quantity of money available in the whole economy is always sufficient to secure for everybody all that money does and can do." (Human Action, p. 421)

So let me repeat: "The quantity of money available in the whole economy is always sufficient to secure for everybody all that money does and can do", which is to say, the idea of increasing money supply to make it "match" demand does not make any sense (for Mises).



In response to your earlier comment, I partially agree. However, to the extent that downward ridgities are caused by the money illusion, what you describe is not so. Your argument assumes that workers have overcome the money illusion to understand how real wages may fall while nominal wages remain constant. I do not think monetary policy should attempt to compensate in such a case. In other words, my recommendation cannot neuter the misallocative impact of unions -- at best, it can reduce the distortion.

But perhaps this discussion is better left for another time and place, since these comments have already digressed enough!


Reread the beginning of what you just quoted:

"As the operation of the market tends to determine the final state of money’s purchasing power at a height at which the supply of and the demand for money coincide, there can never be an excess or a deficiency of money ..."

To me, this can be paraphrased: "When there is monetary equilibrium, there can never be too much or too little money."

In other words, everything Mises wrote that you quoted is said under the assumption of monetary equilibrium. There is nothing for Horwitz, Selgin, or any other free banker to disagree with in this instance.

If Mises meant something else by that first sentence, then please explain. At the moment, I am struggling to think of any other sensible interpretation.

Sorry for the triple post, but I just want to clarify my previous comment.

"As the operation of the market tends to determine the final state of money’s purchasing power at a height at which the supply of and the demand for money coincide" = "the market tends toward monetary equilibrium"

"at [the] height at which the supply of and the demand for money coincide, there can never be an excess or a deficiency of money." = "when there is monetary equilibrium, any quantity of money will do."

If you own a business, and you see the prices of the goods you produce increasing, you are going to interpret that as increased demand. Now, you may not be seeing your products flying off the shelves, but you will also logically conclude that others making the same product must be selling lots of their product, meaning they may face a shortage soon. Good news for you, if true. That means that you will be in a position to provide the goods in question when your competitors run out. In fact, it might make sense to ramp up production in anticipation of demand.

With natural inflation, this makes sense. With monetary inflation, everyone is making the same interpretation of the data -- none of their supplies may be going down, but one does have to anticipate future demand, so one has to make more. This is the boom.

It can also happen with artificially low interest rates -- where money becomes cheap, and people are willing to make riskier gambles, since the money is so cheap.

When prices drop over a long period of time, people wait for the deals to get better. Lower demand causes prices to drop further.

When the first producer realizes he has an oversupply of his product, he offers it for sale. The drop in price causes people to demand his goods. This causes others, and yet others, to drop their prices -- we now have system-wide deflation.

Boom and bust -- caused by monetary policy.

Thanks Jerry,

The relevant quote from Alchian and Woodward’s 1987 'Reflections on a theory of the firm' is:

“… the notion of a quickly equilibrating market price is baffling save in a very few markets. Imagine an employer and an employee. Will they renegotiate price every hour, or with every perceived change in circumstances? If the employee is a waiter in a restaurant, would the waiter’s wage be renegotiated with every new customer? Would it be renegotiated to zero when no customers are present, and then back to a high level that would extract the entire customer value when a queue appears? … But what is the right interval for renegotiation or change in price? The usual answer ‘as soon as demand or supply changes’ is uninformative.”

Alchian and Woodward then go on to a long discussion of the role of protecting composite quasi-rents from dependent resources as the decider of the timing of wage and price revisions.

Alchian and Woodward explain unemployment as a side effect of the purpose of price rigidity, which is the prevention of hold-ups. Alchian and Woodward note that unemployment cannot be understood until an adequate theory of the firm can explain the type of contracts that the members of firm contract with one another.

Seriously? There was a nice discussion going on here among top scholars, why do you guys always have to ruin everything?

I think it is demonstrable that money supply has a great effect upon aggregate economic movements, as described by Troy Camplin.

This is a major flaw in Mises' theories. While at the same time I think the Monetarists overstate the action of money supply.

Take our recent economic downturn. While I might take issue with Tarp and some of the fiscal steps taken to deal with it. I will say that the action of the FED to pump money and liquidity into the system was exactly the right thing to do at the right time.

You cannot prove a negative and so we cannot know it another policy, or no policy would have worked as well, but I certainly do not think so.

Pete: "I think the way you and George have put this doesn't clarify the issues."

My post was pretty brief, but the gist of it was the following statement: "...I emphatically believe general gluts _are_ possible, Say's Law notwithstanding, so long as by 'general' one means 'pertaining to goods in general' and not 'pertaining to both goods and money.'" Could you tell me, Pete, precisely what's cloudy about this statement, which I had considered unexceptionable, or how I might have rewritten it had I taken the time to re-read Blaug, Say, and Malthus?

Or was it something else I said that was confusing?

Lee Kely,

Mises does not say "in monetary equilibrium" as you assume, but "ALWAYS". And he says: "there can never be an excess or a deficiency of money" That's a key. He meant what he said: there CAN NEVER be an excess or a deficiency of money. Any supply of money is always optimal. Period.

Re-read the quotation on hoarding I provided earlier. It confirms clearly that he thinks that even during the "unanticipated increase in demand for money" the dogma of any supply of money as "optimal" still holds. The only way Mises had ever imagined the supply and demand for money can "equilibrate" was via the increase in the purchasing power of money, i.e. liquidation and recession, never through the new credit creation. If you claim that he ever considered the new credit creation as a viable alternative to the increased purchasing power of money and painful liquidation, please provide a quotation, I would me more than happy to learn and revise my understanding accordingly.

However, until you do that I cannot help but to repeat that, in his framework, this notion of an "equilibrium" as a consequence of the elastic money supply adjusting to the increased demand does not make any sense. He repeated many times in HA that sometimes sharp contractions of credit were "unavoidable" and that it would be very harmful and dangerous to try to prevent them by the new credit inflation. I am unable to see how the following two approaches could be ever reconciled:

"During the deflationary recession, an appropriate monetary policy (in CB or FB) would be to stabilize MV" (Selgin, Horwitz and co.)

"As the operation of the market tends to determine the final state of money’s purchasing power at a height at which the supply of and the demand for money coincide, there can never be an excess or a deficiency of money." (Mises)


do you know the difference between (comparative) statics and dynamics? Do you know that only infinitely flexible prices transforms the latter kind of analysis into the former? Do you know that only in this case it is permissible to argue that real balances are 'just right' at each and every moment? And even if Mises had argued that way (Steven showed that he didn't), why don't you make up your own mind? Think! Don't be a believer!

Since there is nothing like instantaneous price adjustment for *any* good, why don't the free-market ME theorists fret about changes in demand for non-monetary goods? What's the case for markets here?

All over the place in "Human Action" Mises talks about the "money relation". And all over the place in "The Theory of Money and Credit" he talks about the "demand for money compared to the stock of it". If Mises really thought that the supply and demand for money immediately equilibriated as Nikolaj claims then where would be the sense in all these remarks?

I think Lee interprets Mises correctly, Mises begins by assuming that the supply and demand for money is matched by market processes, so the short-run problem isn't present. He then goes on to explain how if supply and demand are matched then it makes no sense to talk about a shortage. This was important for debates at the time because some Keynesians believed that creating more money beyond the demand for it could raise output and employment permanently.

Also, Nikolaj writes:
"He repeated many times in HA that sometimes sharp contractions of credit were "unavoidable" and that it would be very harmful and dangerous to try to prevent them by the new credit inflation"

Once malinvestments are discovered they cannot support credit. As such the credit that they did support cannot continue to exist. So there must be a "credit crunch" of some sort. But, this doesn't necessarily mean that there must be a corresponding "monetary crunch". But, certainly, if banks don't have enough assets on their balance sheets to meet the demand for money then the only way the market can proceed is through price deflation.

A long time ago Steve Horwitz taught me long ago that believing that the market for money is perfectly flexible means denying that ABCT can occur. The question is whether prices are transmitted quickly across the economy. If they are then not only would there be no secondary recessions there would be no ABCT booms and busts. Because, as new money is created that money would bid up prices in asset markets, the higher-order goods, which would spread quickly through the economy raising the price of consumer goods, the low order goods.

It is implausible to suggest that prices adjust very slowly when they are being bid up, but that they adjust extremely quickly when they are being bid down. It's inconsistent to be a "New Classical" for the downward direction but not the upward direction as Rothbardians attempt to do.


Perhaps you are right. I have read quotes that suggest Mises did see credit expansion as equilibrating when there is an excess demand for money (in the broader sense), but I cannot remember where those quotes were from precisely (though I believe some came from TToMC). Steve probably knows the relevent citations.

But if you are right, then what does Mises mean by:

As the operation of the market tends to determine the final state of money’s purchasing power at a height at which the supply of and the demand for money coincide

What does it mean for the 'supply of and the demand for money' to 'coincide' if not for the supply of and demand for money to by in equilibrium? What does Mises mean by 'coincide' and why would this statement be there if 'any suppy of money is optimal', because what happens when the 'supply of and demand for money' do not coincide? Presumably you would answer that prices rise or fall. But isn't the ABCT premised on price inflation driven by *too much money* (relative to the existing structure and level of prices)? Wouldn't the distortion of monetary expansion not be considerably lessened if it coincided (to use Mises's term) with a general increase in demand for money? Better yet, suppose the increase in demand for money is not general but very specific, so that people who receive the new money just reduce their spending in proportion. In this case, the monetary expansion has no consequences for the existing structure and level of prices, and no Austrian business cycle is instigated despite the increasing money supply.

If your interpretation of Mises is correct, then I believe Mises was a bad economist. He misunderstood the basic principles of monetary economics without which his theory of business cycles is nonsense. Fortunately, I do not believe your interpretation of Mises is correct.


The claim that prices are imperfectly flexible is a positive claim not a normative one. I don't "worry" about it. It's simply a fact of life in real world economies. That fact has implications for how monetary disequilibria play themselves out. The normative claim is about whether allowing those imperfect price adjustments to bear the burden of removing monetary disequilibria is more of a problem than allowing a free market in money to prevent the costs of price adjustments from happening in the first place by preventing said monetary disequilibria.

Let me throw the question back at you:

If you think downward price adjustments in the face of an excess demand for money are not problematic (because people always hold the "right" amount of money or for whatever other reason), why do you think UPWARD price adjustments in the face of an excess supply are so awful? Folks like you and N worry so much (and rightly so) about the ways prices do not adjust perfectly and smoothly to credit expansions, but if you think the supply of money is always right and worrying about imperfectly flexible prices is silly, why are you so worried about inflation?

After all, as Current says, there wouldn't be a cycle if prices adjusted upward the way you all think they work going downward.

What gives? If nothing else, the MET folks at least treat price adjustment processes symmetrically and understand that both directions are discovery processes involving bumbling, stumbling, Mengerian man.

Here's a thought experiment.

Suppose there was a counterfeiter who hoarded every dollar he printed. His home was jammed full of hundred dollar bills, but he never spent any of them. Unbeknowst to the outside world, the money supply was significantly greater than any of the official estimates. Yet the relative structure and level of prices was entirely unchanged, because the counterfieter's hoarding prevented any of the currency from 'circulating'.

The new money was 'sterilised', because the counterfeiter's demand for *real* money balances exactly offset the increasing supply of money. In other words, his increased money demand was a reverse 'injection' effect.

Now suppose the counterfeiter's house burned down with him in it; he died and all his counterfeit money burned with him. The significant decline in the money supply wasn't noticed by the outside world, because the relative structure and level of prices remained unchanged. When the counterfeiter died, the falling money supply was matched exactly by a fall in the demand for *real* money balances.

I think this thought experiment is instructive for many purposes.

The key to understanding Nikolaj's arguments is the question, "what creates the most effective propaganda for the gold standard?" In organizing a revolution that will, among other things, institute and maintain a gold standard, what is the best thing to say to the more intelligent of the prospective cannon fodder? While "the truth" as best we understand it is one possible answer, Noble lies (or half truths) motivating them to hit the barricades is another.

The existing recession is all the Fed's fault, and after we abolish the Fed, the market will avoid recessions, and so more or less full employment and prosperity will be our glorious future. Just create a 100% reserve gold standard...

Surely, this is the most motivating message. Why confuse the cannon fodder with the truth?

Excess demand for money? You must mean that ordinary people should not be free to choose how much money they will hold.

And, of course, what might happen if there were a shortage of money and how the market process would work out at a fixed quantity of money always morphs into... but an increase in bank issued money is always bad.

Well, if we imagine that it might help, and that in some scenarios, that would be inconsistent with maintaining gold redeemability (much less 100% gold reserves) then that is giving up a bit of ground to the enemies of the gold standard! When ordered to charge the barricades, perhaps some of the cannon fodder will have doubts.

Well, maybe an increase in the quantity of bank issued money always has even worse consequences than price and wage deflation. But that doesn't mean that there is no such thing as an excess demand for money and that the market process that corrects it at a given quantity of money isn't painful. I am certain that a gold standard has some disadvantages and I grant it has some advantages.

As far as I know, no monetary equilibrium/disequilibrium theorist argues that an excess demand for money creates a permanent unemployment equilibrium. Everyone understands that there is a market process that allows for a readjustment even if the quantity of money is fixed.

Of course, I don't believe that there is any problem created by an expansion in the quantity of money that matches an increase in the demand to hold money. And, a fractional reserve gold standard allows an adjustment of the quantity of bank issued money to match the demand to hold that money in many circumstances.

I am very worried about the circumstances where a gold standard would intefere with that process. And so, I don't favor a return to the gold standard. On the other hand, I admit there is no perfect alternative.

One final note. My view is that much of the simple analysis of the impacts of "imbalances" in the economy assume monetary equilbrium. We import more, so we export more or build more capital goods to reflect the net capital inflow. Employment shrinks in some places, it grows in other places. If these changes were due to the end of some past perverse policy--we were building lots of tanks, and now government spending falls and taxes fall and we go out to eat more often, that doesn't mean that the inevitable disruption is much different than if new especially tasty restaurant meals motivated people to switch from cars to eating out. Ending a government policies that encouraged building too many new homes and now producing other more valuable things is disruptive, but no more than if people just decided they didn't want so many new homes.

If monetary disequilibrium develops due to these changes, the inevitable disruption will be compounded by a need to adjust the purchasing power of money so that the real quantity of money matches the demand to hold money.

The problem is with the monetary institutions that fail to maintain monetary equilibrium, not with some change in the allocation of resources.


This was important for debates at the time because some Keynesians believed that creating more money beyond the demand for it could raise output and employment permanently.

Exactly. Mises was not responding to the modern controversy about monetary equilibrium theory; he was responding to his contemporaries who argued for monetary expansion even when the 'supply of and demand for money coincid[ed]'.

Ironically, Nikolaj also quoted this statement by Mises:

(Of course, one must not fall prey to the popular confusion between and the appetite for more wealth.)
the demand for money for cash holding----quote----

Isn't this precisely what Nikolaj has argued in the past here when claiming that money demand is infinite because people always desire more money (assuming constant purchasing power)? And if Mises believes we shouldn't 'fall prey to the popular confusion', then what does Mises understand the proper understanding of the 'demand for money for cash holding' to be? Contrary to Nikolaj, hardly anything Mises wrote in the quote he provided is comprehensible except in light of MET.

Oops! That second quote is supposed to read:

(Of course, one must not fall prey to the popular confusion between the demand for money for cash holding and the appetite for more wealth.)


in your experiment there is an agent producing zero-velocity money (which you assume is a perfect substitute for 'real' money). His increasing money supply never becomes effective, that is, it never becomes part of the exchange nexus. Thus, burning down his stock doesn't affect any other market participant (obviously, the agent's consumption and asset purchases - that is, his economic interaction - are paid out of income + 'real' ownership). This case is trivial!

The discussion here focuses on the case, where CIRCULATING money supply drops due to higher money demand (v drops while M is constant => Mv declines). In this case the price level (and relative prices) must fall to establish equilibrium (otherwise convergence is accomplished by falling income = temporary crisis!).

The debate is focused on whether P drops fast enough to ensure that output losses are moderate. In this case, monetary policy can remain passive. Or, and this is Steven's point, does it take a while for P to drop, for whatever reason, such that output losses are comparatively high. In this case, it is possible for monetary policy to increase M (such that +deltaM=-deltav => PY constant).

Your experiment cannot resolve this dispute.

read: ... price level and relative prices must change ...

I don't really share Bill Woolsey's cynical view of the 100% reserve supporters or their intellectual leaders. I've never seen anything to suggest that they don't really believe what they're saying. Though I think many of them fear that a free market wouldn't stamp out fractional reserves.


I intended my thought experiment to be a jumping off point to understand how monetary expansion does not necessarily produce malinvestment. Austrians often claim that credit expansion, even when enabled by greater money demand, distorts relative prices. Thus, even when there is an excess demand for money, expansionary monetary policy is likely to produce malinvestment. I believe this view is mistaken: it merely seems plausible given a misapplication of 'injection' effects. My little thought experiment seems like a good jumping off point to explain why this is so.

This debate, unfortunately, has come to include many topics (I have not helped on this matter!) Perhaps I should never have posted my little thought experiment, so that we could focus on the topic at hand. On that matter, I agree with everything you and Steve have written so far.


What you're talking about is velocity, which I don't think anyone would deny is an important element. Even thought he Fed, for example, injected over a trillion dollars into the economy, they did so in such a way that they are hoping to get it all back before the banks begin to lend it out. If Bernanke can do that, no harm, no foul. But if he can't . . .

Steve Horwitz has elsewhere pointed out that there is a stealth inflation going on -- with slightly smaller packaging and/or less content at the same price. Recently I saw a brand of juice I was buying at Walmart go from $1.00 to $1.33. Might be higher fruit costs; might be velocity (meaning Bernanke's too late). If we're starting to see higher prices everywhere, without increases in wages coming first, or higher employment, it seems likely that inflation's here.


Although I was thinking about velocity, that was not all I meant to talk about. I suppose I was trying to shift the analysis away from money and toward spending. A lot if made of where money 'enters' the economy when banks expand the supply of credit, and the distortionary consequences that supposedly follow. But looking at the situation like that is misleading, I think, because what really matters is relative shifts in spending.

When it comes to the kinds of distortions we have been talking about, one has to talk about money. As has been said here, money consists of half of every trade. Money -- and its relative quantities -- matters.


I tried, elsewhere, to elicit an answer to a rather simple question:

Assuming a fixed stock of base money, as apparently free banking does, in what way a hypothetical free banking system is any different from what we have today, during a period when the Feds do not conduct any open market operations ?

My understanding is that the fixed stock of base money is what is supposed to prevent uncontrollable growth of credit under free banking.

vjk: "Assuming a fixed stock of base money, as apparently free banking does, in what way a hypothetical free banking system is any different from what we have today, during a period when the Feds do not conduct any open market operations ?"

Let's be clear: "free banking" doesn't itself assume or imply any particular base regime. It refers to a banking regime in which banks are free to issue notes, to branch, and to choose their assets, including their cash reserves, free from special government regulation. Some students of free banking, myself included, have suggested that, were a free banking system combined with a frozen stock of fiat money, the combination might be conducive to monetary stability. (Milton Friedman also thought so, FWIW.)

It should be obvious from this summary that to imagine that free banking "means" a frozen base, so that any arrangement in which the base isn't changing must perform in the same manner as a free-banking arrangement with a frozen base, is to be guilty of a rather crude mistake.

It is also a mistake to assume that a lack of open-market operations means zero base change. (If recent experience doesn't disabuse one of _that_ belief, I daresay nothing can!). Finally, it is a mistake to suggest that there have been no (net) base growth in recent years. On the contrary: besides the gigantic spike of October 2008 there has been a pronounced upward trend since. Indeed, if anything the base looks much flatter for the years leading up to the Sept. '08 panic than it has since then.

It is true, on the other hand, that a fixed base will "prevent uncontrollable growth under free banking." But it will do so under almost any regime. On the other hand, any _positive_ growth of the base, so long as it isn't "uncontrolled" growth, will suffice to prevent "uncontrollable" growth of credit!


Some students of free banking, myself included, have suggested that, were a free banking system combined with a frozen stock of fiat money, the combination might be conducive to monetary stability

Is there a numerical model of some kind that would quantify credit expansion given a frozen stock of fiat money under free banking?

It is also a mistake to assume that a lack of open-market operations means zero base change.

Could you elaborate on that point? Without OMO, what would be the source of base money ?

It should be obvious from this summary that to imagine that free banking "means" a frozen base, so that any arrangement in which the base isn't changing must perform in the same manner as a free-banking arrangement with a frozen base, is to be guilty of a rather crude mistake.

I did not mean to say that the fixed stock of base money was the *only* feature of free banking. Sorry, if it came through this way. I am interested in seeing how well such fixed stock could control credit expansion.

Thank you.

vjk: "Is there a numerical model of some kind that would quantify credit expansion given a frozen stock of fiat money under free banking?" Yes. Call it the reserve ratio model. M=(1/r)B were B is the quantity of reserves. So long as r is finite, so is M. (For credit just consider that M is backed by assets consisting of reserves plus credit (loans and securities). As for the precise determinants of r, you'll find them elaborate in standard models of reserve demand, going back to Edgeworth's (1888) "mathematical Theory of Banking."

In fact no economist, so far as I know, has ever doubted that a fixed B would suffice to limit M! Indeed, most would assume that such an arrangement could only lead to deflation.

"Without OMO, what would be the source of base money." Suppose a central bank does lots of discount-window lending to other banks. Then B goes up. You don't need open-market operations. In fact, the Fed didn't make much use of OMOs before the 1930s.

I suggest you have a look at some basic text on money and banking to get some further information on these topics.


I think it is confusing to talk about general gluts period. We can talk about coordination problems caused by disproportionality and we can talk about government induced disturbances. But I think it clouds rather than improves our understanding to talk about general gluts, under-consumption, or over-production, or even bubbles as such. If you assume people are delusional and that prices cannot do their job, then of course the self correcting forces of the market will be none existent. But just make a commitment to thin rationality, and to price adjustment processes and all this stuff about "gluts" becomes nonsense. Market forces work, and they work through price adjustments. Coordination problems confront us everyday and every moment of everyday, but economic actors guided by price and the discipline of profit and loss work to solve these problems. Unless they are prevented from doing so. IF that is the case, then the issue is one of policy induced problems NOT general gluts and whatnot. In the Chicago interviews, I have to say Cochrane was pretty good, and Casey Mulligan has actually made a lot of sense throughout this entire period.

Perhaps we just disagree --- wouldn't be surprised I actually don't believe there is such a field of study as macroeconomics. There is price theory, there is our understanding of particular markets, and that is it. Say trumps Malthus, Hayek trumps Keynes. Old fashioned I know, simple I know, but being old and simple is not necessarily wrong.

Pete, tell me if this helps clarify.

You're used to thinking in disproportionality terms because Austrians usually talk about inflation, during which we get the opposite of a "general glut". There's "too much" spending going on. The problem is that this can't really manifest as a "general shortage" because in the short run (think Roger's point beyond the PPF) we can make it look like we can produce more to meet the excess spending. But the interest rate effects also mean we get distortions in WHAT we produce alongside that temporary "too much." But the key is that real resource constraints make the disproportionality become the real problem.

When we talk about an excess demand for money, we don't face the same sort of constraints. It IS possible to have some amount of "everything" going unsold (looking like underconsumption). Yes, that "everything" will include some disproprotionality, but the most obvious aspect of the problem will be lots of goods going unsold, i.e., a general glut.

In other words, what's most visible in inflation is the disproportionality, which disguises what could be termed a "general shortage" as compared to the flow of expenditures, and what's most visible in deflation is the excess supplies of goods sitting around unsold, looking like a general glut.

I wouldn't deny that disproportionality is a problem, but in the case of EDM, we do, in fact, get what looks like "too much of everything." The problem, of course, is with the monetary system and left to its own devices, the price system will eventually clear up the excess supplies, but not without much pain.

As far as "macroeconomics" goes, all I'll say is "money is different." Once we start talking about disequilibria in money, the effects are systematic in a way that goes a step beyond *just* price theory. Obviously I agree that the problems of monetary disequilibria manifest themselves in the price system, but the insight that too little money leads to what looks like "too many" goods in general and that too much money leads to systematic disproportionalities is a distinctly "macroeconomic" one.

I'd frankly like to get rid of the word "macroeconomics" as, like so many others, it was coined by those who have abused it. Maybe "monetary theory" is enough, but even that doesn't get at the system-wide effects of monetary disequilibria.


I don't really see that much difference between the comments that Steve, George, Mario, Jerry and I have made. It's perhaps best to never exclude money from the items considered in the "general" part of "general glut", because if money is excluded then it's not really general. The "general" part should mean "all supply" (and not "all reproducibles"). But apart from that, it all seems reasonable to me.

Notice that as you've mentioned yourself a general glut is conceivable. Overall the argument for the Austrian viewpoint has to proceed by first explaining why Say's law is true, then explaining that that means that relative problems are the real problems and that leads to the Coordination problem.


" Call it the reserve ratio model. M=(1/r)B were B is the quantity of reserves

I do not think it's a satisfactory model since in many countries, e.g. Canada, r = 0 (zero).

In the US, the "r" role is greatly diminished as it only applies to transactional deposits and not to savings (unless factually used as a transactional account), or term deposits of any kind.

In fact no economist, so far as I know, has ever doubted that a fixed B would suffice to limit M
What exactly is such belief is based on ? Given fixed aggregate quantity of base money B and no imposed reserve requirement, how would the credit expansion be limited ?

I suggest you have a look at some basic text on money and banking to get some further information on these topics.

Although I am not an economist, but just a humble practitioner of the "art", I do know how base money creation works under the Fed regime. Besides, it is unlikely that the de haut en bas recommendation to go and read a book is really helpful to educate hoi polloi on benefits of free banking.

So, my original second question stands: what other sources of basic money creation, other than the Feds OMO, do you have in mind ?

Pete: Steve anticipates much of my response. Let me just add that I don't see how jettisoning the notion of "general gluts" clarifies: it is simply an old-fashioned name for an excess supply of goods-in-general which, far from being a meaningless concept, is perfectly meaningful, being properly understood (as so many of us have insisted) as the flip side of an excess demand for money balances. Now, it's true that here as in other instances of market disequilibrium nominal price adjustments will tend to eliminate the excesses of demand and supply in question. But the comprehensive adjustments needed in the case were it is money that's in short supply, combined with fixed contracts and other sources of nominal rigidity that make some prices particularly slow to adjust, makes it possible for "general gluts" to persist for what can be painfully long intervals.

I also find little merit in the attempt to distinguish between coordination problems and such "macro" obsessions as general gluts. A general glut just happens to be a very intriguing sort of coordination problem precisely because it doesn't lend itself so readily to the usual price-adjustment solution.

Now as you well know, this is just re-hashing in a nutshell what Yeager, Leijonhufvud, Clower, and untold other monetary economists have argued at great length; it is, indeed, a majority perspective, the minority consisting of a small number of die-hard Rothbardians and the more intransigent New Classical economists. Most significantly, it is an implicit part of the great Wicksellian framework, for were p-adjustments always sufficient to clear the market for money balances, that is, were such adjustments as straightforward and expedient a means for eliminating monetary shortages as they are for eliminating shortages in most other markets, there could be no divergence of actual from natural interest rates and (by implication) no Austrian cycle.

As for macroeconomics, the routine bashing of the field frankly bores me, implying as it does that macroeconomists are not at least endeavoring to use precisely the same basic economic logic as everyone else, but with the particular aim of understanding economy-wide developments such as inflation, business cycles, and mass unemployment. Fifty years ago, to have voiced opposition to "macroeconomics" might have meant nothing other than to insist on building upon solid microeconomic foundations. Today it can't mean that anymore, for the simple reason that every macroeconomist at least pays lip service to such foundations, and would blush to have specific features of his theory inconsistent with such pointed out. Instead, macro-bashing amounts to suggesting that the special problems macroeconomists concern themselves with aren't worthy of study.

Consider the following mental experiment: you kill all the macroeconomists. Now, will the self-styled "pure" micro-types--that is, those who are pleased to never have soiled themselves by conceding the possibility of general gluts and such--step up to the plate to try and shed light on the business cycle and such? I doubt that they would, and I think that the theory of demonstrated preference supports my conjecture.

Finally, I want to insist that to share the particular Austrian preoccupation with relative price distortions needn't compel one to reject out-of-hand the suggestion that prices (that is, goods prices) might generally be too high or too low. It is bad economics, and not simply bad macroeconomics, to jump from insisting that one not overlook the relative price effects of monetary disturbances to insisting that such effects are all that matters, that is, that prices must _on average_ be at their market clearing levels, as if every too-low price must have its too-high counterpart somewhere.

> So, my original second question stands: what
> other sources of basic money creation, other
> than the Feds OMO, do you have in mind ?

George has already answered that question - discount window lending.

I have a simple question --- can you have a general glut in a free market? Note, I am not asking if we can have a general glut in a barter economy (which we know is impossible), I am asking can we have a general glut in a free market economy.

If not, why? If so, why?

P.S.: On macroeconomics, I believe there is a pernicious influence that impacts even people who know better to start talking about systems of relationships unmoored by human choice, rather than chains of causation that can always be traced back to individual choice. I do believe money is one half of all exchanges, so if you screw with money you screw with all exchanges. Thus, don't screw with money. But note that the reason the manipulation of money and credit causes problems is through screwing up the exchanges in the economy.

P.S.S.: I don't buy the cumulative rot story unless there is someone disturbing the ability of market prices to adjust. This isn't an Austrian issue, it is a price theory issue. Lets make sure that we emphasize the other half of that UCLA equation -- namely Alchian, Demsetz and Hirschliefer, and not just Clower and Leijonhufvud. BTW, is the argument that I am making that much different from the argument made by Margett or by Hutt? If I appear to be, then either I am expressing myself wrong, or I don't really understand their position.


Sorry, somehow missed this:

Suppose a central bank does lots of discount-window lending to other banks. Then B goes up

The Feds lend through the discount window primarily overnight to sound banks. So, the base money is destroyed upon the overnight loan and repayment.

In some cases, the discount window loan maturity can be extended while remaining extremely short-term to facilitate the return of the otherwise solvent bank to healthy liquidity or to resolve the bank in an orderly way.

One has also to keep in mind that the discount window is literally the lender of the last resort due to the extra points (50 bps currently) the bank has to pay over borrowing in the interbank market, as well as the stigma associated with resorting to this sort of borrowing that may lead to real or perceived impression of the bank insolvency.

In any case, the discount window created base money is very short-lived, typically one day, and cannot possibly contribute to the base money sustained growth.

vjk: The lack of mandatory reserves doesn't generally mean r=0. Canada's zero (overnight) reserve arrangement is unusual. In essence, banks there rely on overnight overdrafts as a substitute for overnight reserve balances, where the overdraft rate is targeted by the BofC in much the same manner as the Fed targets the federal funds rate. To achieve its target it adjusts B. Consequently it, too, relies on the existence an implicit base money multiplier, albeit one in which aggregate overnight settlement reserves (and hence the settlement reserve ratio) is zero, making the currency ratio its main determinant.

When I referred to a simple multiplier relation M=(1/r)B I specifically had in mind a free-banking frozen base arrangements of the sort you originally inquired about, where there is no central bank capable of administering a system of overnight overdrafts. I did not mean to suggest that that's the appropriate way to conceive of the multiplier in every instance.

As for my referring you to a textbook, I did not do so with the intend to be condescending. Its just that one can find in such a book a far more detailed discussion of the various ways, apart from open-market purchases, by which central banks can increase the monetary base, as well as more detailed treatments of the base-money multiplier, than one can reasonably expect to be communicated on a blog.

There happens, nonetheless, to be a good and pertinent discussion of the multiplier, in Canada and elsewhere, on another blog--Nick Rowe's Worthwhile Canadian Initiative.

vjk: "In any case, the discount window created base money is very short-lived, typically one day, and cannot possibly contribute to the base money sustained growth."

Usually, yes. But not necessarily, and certainly not recently!


I actually think with respect to your thought experiment, that IF say someone like John Cochrane ( had been in charge in the Fall of 2008, a market correction would have been, well, a market correction, and not become an economy wide crisis. I actually think the Butos and Koppl model of how Keynesian Big Players can create a Keynesian world is a very important insight.

I don't blame DSGE models for our current mess, I blame G --- nothing else. G is at fault. And how G is at fault is by screwing up the price signals, including the manipulation of money and credit. So I am with you, government as monopoly supplier of the currency can do untold damage to the economy. But I really don't understand how money in a truly free market can cause problems.

I boil down the older debate about business cycles and general gluts to be about the self-correcting tendency of the FREE market economy. That is what Malthus was challenging and that is what Keynes resurrected.


I've just commented on the short-lived nature of discount window created base money.

Another important circumstance that one may be unfamiliar with is that the Feds, simultaneously, conduct neutralizing OMOs to counter-balance discount window borrowing by repo'ing T-bills and thus extracting base money from other depositary institutions.

In effect, the discount window lending means base money redistribution among depositary institutions, as described above, rather than injection of additional units of such money.

However, my main question is: how would a fixed stock of base money prevent credit bubbles from happening given that the reserve requirement is non-existent theoretically under free banking regime and practically in countries like Canada ?

A numerical model of credit expansion dynamics under free banking would be useful, but a verbal exposition would be interesting as well.


The lack of mandatory reserves doesn't generally mean r=0. Canada's zero (overnight) reserve arrangement is unusual

Not that unusual as the same zero(non-negative) reserve requirement exists in the UK, Australia and Sweden. Perhaps elsewhere too, I am not sure. And it is not an overnight requirement. In regulatory terms, a canadian bank is not limited by available cash on its reserve balance when granting a loan.

In Canada, the banks rely on netting interbank settlements and interbank borrowing primarily to accommodate interbank cash flows. There is as much stigma associated with, as well as punishing CB overnight borrowing rate, as in the US, conceptually, if not numerically. So, a persistently borrowing and repaying bank will be considered suspect, a defaulting on its overnight loan bank will be resolved by CDIC rather quickly.

Canada is a rather remarkable example of a well-functioning banking system not being constrained by the stock of base money. You are correct that the only tool at the BoC's hands to manipulate credit expansion is overnight *interbank* rate corridor influenced by OMOs.

The Feds primary tool is the same, i.e. OMOs influencing the interbank lending rate. Obviously, the tool did not work very well and the results of the Feds activity are here for everyone to marvel at.

When one looks at the free banking proposal, one cannot help but wonder: is the fixed stock of base money a reliable credit expansion governor ?

Imagine a hypothetical situation where two banks lend to the customer of the other bank $1mil or $1trillion. Further, assume that the banks use a DNS(netting settlement network) such as CHIPS. Then, the amount of base money is immaterial due to the netting effect. In effect, the settlement was achieved with credit money.

In a more realistic scenario, but with a more or less similar credit expansion ratio and more or less uniformly distributed interbank obligations, the stock of base money can, at best, smooth out fluctuations in mutual obligations, but in no substantial way does it appear to be capable of unlimited credit growth( Keynesian "lock-step" expansion) prevention.


I am surprised. The Fed actually operates by making very short term loans to security dealers secured by an unspecified selection of goverment bonds. Even though many of these loans are overnight, the total outstanding amount makes a substantial contribution of the monetary base. What counts is the total amount lent, not the term to maturity on any loan. Conceptually, on could imagine that the Fed just lends to first one borrower then another, but really, the same security dealers effectively borrow new money from the Fed to pay back the old money--the borrow for extended periods of time, obligated to pay back the Fed at any time. It is like any other repurchase agreement that is rolled over.


You know that monetary disequilibrium theorists are not claiming that there is an "unemployment equilibrium" that persists forever. That surpluses will eventually lead to lower prices which will cause the real quantity of money to rise to the demand for money, and simultaneously end any general glut of nonmonetary goods and services is part of the story.

But when we observe that spending on just about everything has fallen, and firms who sell those goods are cutting back on production and laying off people... well that is quite interesting and troubling. To me, it appears inconsistent with scarcity. Aren't all of these goods scarce? Why is just about everything being cut back?

Monetary disequilibrium theory says that the problem is an imbalance between the quantity of money and the demand to hold it.

You know that monetary disequilibrium theorists don't claim that there are never surpluses matched by shortages because the wrong goods were produced.

To me, anyway, matching shortages and surpluses is the way it "should" be. In a world of scarcity, we choose this rather than that. To produce what people value most, we produce more of this and less of that. Labor and other resources need to be devoted to producing more of this and less of that.

We don't want this, we don't want that, we don't want anything? Scarcity no longer exists? NO! The problem is money.

Monetary disequilibrium economists aren't claiming that there is no scarcity, or that growth in productivity gets ahead of the growth in income, or that saving isn't what allows for investment for the production of future consumer goods and services.

No, it is a disequilibrium situation that occurs with an excess demand for money. An excess demand for money can be caused by many things. It is not necessary that there be a sudden discovery that too much of some things were produced and too few of other things were produced.

Why in a world where the supplies and demands for most things change for all sorts of reasons should we expect that the supply and demand for money would somehow remain very stable or even constant?

When we think about prices being sticky, this isn't really a problem when there is a shift in demand between goods. We can imagine that the good with less demand has a lower price and the good with more demand a higher price--no shortage or surplus. Then, because of losses for the low demand good, and the profits for the high demand good, the demand for resources shift. And then, that change in demand for resources just lowers the wages and other prices for resources with lower imputed demands and higher wages and other prices for resources with higher imputed demands. And then, because of the differentials in wages, the labor and other resources shift. And then production shifts. Then prices change and profits and losses dissipate.

Now, if some of these prices are sticky, then we can have shortages and surpluses. We actually can have shortages of a good with an incease in demand. In many industries there is a backlog of orders that can develop. And certainly we can have surpluses. The firm with the backlog can expand production make new hires, and the like. And the industry with lower demand, cuts output, lays off workers. The resources shift.

In reality, prices and quantities shift in a variety of ways. But the resources move.

What is special about an excess demand for money and a matching "general glut" of goods, is that there is no need to shift resources. If the process that brings the market to equilibrium is a change in the price level so the real quantity of money adjust to the demand, then any decrease in quantities are just errors. It isn't that there is a need to produce less of everything. It isn't that resources are no longer scarce and we have too much of everything.

The market institutions that generally shift resources from less to more valued uses don't work right in this circumstance.

A decrease in the quantity of money or an increase in the demand for money doesn't mean that people want to work less and consume less now _and_ in the future. There is still scarcity.

So, perfectly flexible prices might cause all sorts of problems in many markets. And real world sticky prices work ok when the problem is shifting resources from less to more valuable uses. But only the perfectly flexible prices will work so that the real quantity of money will remain equal to demand to hold money without shifts in production from more valuable to less valuable things. Especially a shift from present _and_ future consumer goods to leisure.

I could just say "What Bill and George said" but I'll add one more thing in response to this:

"I have a simple question --- can you have a general glut in a free market? Note, I am not asking if we can have a general glut in a barter economy (which we know is impossible), I am asking can we have a general glut in a free market economy.

If not, why? If so, why?"

If by "free market" you include a genuinely free market in money, then the answer is no in my book. If you then want to go on to say that "see it's all G!" that's fine. But understanding WHAT "G" is causing the problem is what Bill, George, and I are talking about.

I would argue that people who thought general gluts were possible under laissez-faire weren't even considering the monetary question very carefully and thus concluded gluts were possible under "laissez-faire." The point of noting that EDM can cause a general glut is to then provoke the question of *why we have EDM in the first place!*

You want an answer to the "underconsumptionist" theory of the Great Depression? You need the analysis we've been offering. If you want to just keep repeating "it's all G" please go right ahead (you know I agree), but that doesn't help us understand the actual market processes by what *appear* to be free market general gluts actually are not.

And yes, prices as they really exist and adjust are "good enough" to ensure that markets work well *assuming the money supply is right.* But if we want to explain why getting money wrong leads to trouble, we have to talk about the ways in which prices do not respond instantly and perfectly to changes in demand. There's no normative judgment there, just a positive statement about what will happen with ESM or EDM. Without recognizing the *fact* the price changes will be less than instantaneous, due to the very institutional frictions that you so often remind us are the driving force of markets, there's no grounds for identifying welfare losses due to inflation or deflation.

The weird thing here is that if you want to argue that the fact that prices don't adjust smoothly and instantly is just part of how markets coordinate, then you are running a risk of sounding like Lucas. Consider the following:

We have a completely free market economy except for the monetary system. The central bank screws up and produces too little money. Demands begin to fall across the economy. Due to market institutions such as contracts, as well as the real time it takes for entrepreneurs to discover that the "objective" data have changed, prices and wages do not respond instantaneously to the falling demand and the result is unemployment. You seem to be saying that either there's some reason that unemployment won't occur (if so, tell me what it is), or that the unemployment that results involves no welfare loss in a comparative institutions sense because, hey, markets are doing as well as they can because Big G isn't interfering with prices and wages.

Let me throw a question back at you: do you think the Austrian cycle theory story can take place in a world where government has no role in interfering with prices and wages, yet there still is a central bank? If so, why? Isn't the ABCT story ALSO one in which prices don't adjust perfectly to the, in this case, excess supply of money? If inflation can cause relative price effects in the absence of government interference with prices and wages, why can't there be a cumulative rot during deflation in the absence of Big G?

The only way you can escape this dilemma is to either: a) admit the cases are parallel by agreeing that both involve welfare losses in comparative institutions context or b) admit the cases are parallel but DENY they involve welfare losses by just saying "well that's how markets adjust." I really don't think you want to take the second position, but I also think you nearly do based on your invoking Alchian et. al. above.

Yes markets adjust imperfectly, which is more than good enough if money's right. But if money isn't right, the fact of their imperfect adjustment is what leads to money causing mischief.

A lot of what has recently been said are fancier versions of my point above about how government-created inflation and artificially low interest rates cause distortions.

It seems to me that Pete is suggesting that without government involvement in the market -- if we did indeed have a free market -- then there would be no such thing (nor need to be any such thing) as macroeconomics. Perhaps. Or perhaps it would take on a completely different character. What would macroeconomics look like if we could have a truly free market?

I wish I could say to Pete, "Yes, we all agree that things only get (macroeconomically) screwed up because of G, so let's hug and forget we ever disagreed." Except I can't accept the G-theory as a starting point. It has to be something we arrive at by means of careful study of how a laissez-faire monetary system differs from one that government overseas. But having given some attention to this very question, my own opinion is that, although a LF monetary system may well be more conducive to avoiding macro disorders than a G-system, the difference is bound to be one of degree: there's no question of perfection under LF, so "macro" thinking doesn't become entirely irrelevant, although it might not be all that important. In that case, one could well kill all the macroeconomists without doing much damage. But we are a long way from LF and, as much harm as today's bad macroeconomists do, I firmly believe that having none at all would be worse, for your average pure-micro type would _not_ be likely to step into the breach to the rallying cry of "end the Fed!" Instead, the breach will quickly get filled by a new sort of half-baked macrotheorist whose beliefs will make the crappiest of today's professional Keynesians look pretty good. You can test this: just knock on the door of any run-of-the-mill "microeconomist" and ask about whether government should get out of the monetary system. (And I don't mean typical GMU microeconomist!)


The Fed actually operates by making very short term loans to security dealers secured by an unspecified selection of goverment bonds. Even though many of these loans are overnight, the total outstanding amount makes a substantial contribution of the monetary base.

According to the Feds' data, the repo contribution currently stands at zero (WREPO series) with no repo's during last week. Presumably, all repo loans have matured.

The discount window contribution stands at $0.032 bil as of the last week -- that's the "total amount lent". You can compute percentage wrt. the about $2trillion total base. It is not overwehelming.

that is rolled over
The Feds play with fixed maturity repos, most frequently 1 day, very rarely more than two weeks. There is no automatic roll-over.

The last batch was 5 day repos that have already expired (see "Temporary Open Market Operations" at the Feds site).

Steve, let me restate my question. Let's say there is a change in demand for some good, say people demand at prevailing prices more beer than ice cream. Factors of production must be shifted from ice cream production to beer production, yes? And this does not happen instantaneously, it takes time for any market
to clear. Same thing would happen if demand changes between present and future goods, for between money and non-monetary goods. If a case against 100% reserve banking is that price
stickiness precludes instananeous (or even just fast) adjustment, why doesn't that apply to any market-based solution for the production of any good?

Price stickiness isn't "the" case against 100% reserves. There's lots of other problems there that plenty of folks have documented.

The point here is that:

1) the price adjustments necessitated by EDM are economy-wide and lead to major social costs in the form of unemployment etc.. This is the Yeagerian point that for all the ways money is like other goods, it is also fundamentally different in really important ways.

2) As Yeager also argues, across-the-board price adjustments, such as those necessitated by an EDM, are much harder to get going than are adjustments caused by changes in relative demands or productivity. I make the latter point in my 1996 JHET piece. Productivity-driven price changes are sought after by entrepreneurs.

3) The costs of idled resources due to an EDM are avoidable in a way the costs of resource reallocation from one good to another aren't. And they are avoidable by a system that, its defenders would argue, does much else that is good and no other harm, not to mention the fact that it is more consistent with freedom of contract and related values.

I'll put the question to you: If you think the price adjustments necessitated by an EDM are no different than those that take place due to shifts in relative demand, why are the price changes that take place due to an excess SUPPLY of money any different than those that take place due to a shift in relative demand?

Thanks vjk.

So the Fed collected on all repurchase agreements in February of 2009. I should
watch these things more closely.

They still have a good bit of term auction credit.

I suppose we can call their purchases of mortgage backed securities "open market operations."

Let's see.

Are there IPads or potatoe or silos of corn going unsold?

Or perhaps houses and mortgage backed securities?

I'm with Pete.

I don't get this "general glut" talk.

And, if there is a "general glut", does this mean there is a "general glut" in every country, or every state or every city in the U.S.? Is there the same "general glut" in North Dakota and Texas as there is in California and New York and Nevada?


If the supply of gold were highly elastic, then we could imagine a decrease in the demand for ice cream and an increase in the demand for gold would require that resources be freed from icecream production and shifted to gold mining. Of course, with a highly elastic supply, there would be little need for a change in the purchasing power of gold.

But, in truth, the supply of gold is very inelastic. While some resources might well be devoted to additional gold mining and so need to be freed from the production of nonmonetary goods, this is a small effect.

Instead, what mostly has to happen is that the price level fall enough so that the real value of the stock of gold/money increases to meet the demand. The amount of resources devoted to the production of other goods don't need to change much.

If all the gold was already mined out, this would be the sole effect. With a fiat curreny with fixed quantity, that is the only effect.

There is no need to free up resources from the production of everything else to expand the production of money. The signal of reduced demand for goods and services should result with solely a decrease in prices. The quantity decrease, which is the right thing to happen when there is a shift in demand between goods, is entirely inappropriate. Only if prices are perfectly flexible is this inappropriate response avoided.

I can only repeat that the apparent signal of an excess demand for money is that scarcity no longer exists. Produce less of everything. We don't need to produce less of one scarce good so that we have the resources to produce some other more valuable and also scarce good. The signal of lower demand, which usually means the resources currently being are better utilized elsewhere doesn't mean that. It instead only means that all prices and wages need to drop so that real value of money rises to meet the demand.

Again, to the degree more resources are needed to mine gold, then this is not completelly true.

By the way, Buchanan's 100% reserve common brick standard might be instructive. And why use resources to make and then store up piles of bricks to back paper money?

The concept of a general glut is associated with theories which have little in common with the MET. Perhaps it is more accurate to say that an excess demand for money may create circumstances that appear like a general glut. Insofar as a general glut implies a general overproduction, or an inevitable lack of purchasing power, it has nothing to do with MET.

Other phrases like 'idle capacity' carry too many Keynesian connotations. Perhaps a new phrase or term should be coined for MET.

I think that the answer Steve gives to C of C's point becomes clearer when compared to a microeconomic situation.

Let's suppose that there is a a few suppliers of coal and there are, say 100, industrial buyers of coal. The buyers and the seller have agreed long term contracts with each other. These contracts stipulate terms and prices at which the buyers can buy coal from the sellers. If there is a rapid change in demand or supply of coal then the market could continue with this arrangement. Or, if changing the contracts and prices would be too costly or difficult there are other options. Exchanges on a spot-market could take place instead for example. Speculators could also play a role.

If the government were to ban some types of intermediating contracts or speculation then the price adjustments could still be made it would just be more difficult and possibly take longer.

The situation with money is similar. Like other financial assets and contracts there is always more than one way to skin a cat. If when a change in money demand or supply occurs then other entrepreneurs can adjust their prices quickly then that's fine. If the fractional reserve banks can respond more quickly to changes in money demand than other entrepreneurs altering prices can then they are providing a useful service.

One way or another, the money supply will increase to satisfy an excess demand for money. The issue is whether that increase will be nominal or real, and which is likely to be the most costly.

An increase in the nominal money supply will keep the price level constant, but many have fears about injection effects distorting the allocation of resources. Alternatively, an increase in the real money supply will force the price level downward, but others fear this will caused prolonged hardship, political instability, and government intervention.

To me, the preference is clear: an increase in the nominal money supply. I do not believe it necessarily distorts the allocation of resources, and I fear the consequences of an increase the real money supply much more.

Hayek says something like this directly against Keynes:

Things are CONSTANTLY being produced and consumed throughout the bust -- if there was a real "general glut" then the only assumption must be that every type of item in the economy can be reproduced essentially costlessly -- otherwise the supply on hand of many goods would quickly be zero (things even in the bust are constantly being produced and consumed).

Let's think about it.

Why would anyone trade any money at all for the input goods to that production process if there really was a "glut" of the promised output, and if they would prefer more money to any goods at all.

But people are trading money for inputs to production -- and goods are being produced and reproduced, and consumed via trades for money.

They could only prefer input goods to money because there is NOT a general glut.

Greg Ransom: "And, if there is a "general glut", does this mean there is a "general glut" in every country, or every state or every city in the U.S.? Is there the same "general glut" in North Dakota and Texas as there is in California and New York and Nevada?"

How tedious these reductious are. Really, Greg: "general glut" means that the excess demands for all goods sum to a negative number. That entails no uniformity of excess demands at all. Some could be positive. And none of us on this forum who have insisted on the usefulness of the general glut concept has ever suggested otherwise. Nor, for that matter, have we claimed (not all of us at any rate) that such a glut is necessarily what's ailing the U.S. economy today. That's a separate empirical matter.

Anyone can make any notion seem absurd (and thereby also make the notion's defenders seem so) by misrepresenting its meaning.

"If there was a real "general glut" then the only assumption must be that every type of item in the economy can be reproduced essentially costlessly." No, it means merely that production is artificially constrained by lack of means of payment. Relaxing the constraint by expanding the nominal money stock serves, not to make goods drop like manna from heavan, but merely to allow them to be purchased at prices that cover historic costs.

> Note, I am not asking if we can have a general glut
> in a barter economy (which we know is impossible)

I'm not sure this barter economy idea is really quite what we think it is. I know this thread contains quite a lot of drift, but this part is interesting.

If we consider a situation with a hypothetical barter economy and a hypothetical Walrasian auctioneer then Say's identity would be true at all times. But, this shows that the *barter fiction* fulfils this criteria.

Real barter economies aren't like that model. In real barter economies we have the problem that is called the "double coincidence of wants". It is more difficult to make a trade than in a money economy. An individual may proceed by indirect exchange but that is still difficult.

What we really mean by this is that relative surpluses and shortages are often not acted upon because of the knowledge problems involved. Any randomly chosen individual who has a surplus of shortage of some good may not know how to proceed through direct or indirect exchange in order to rectify the situation. Taking the indirect exchange case they may not know what to buy to use for indirect exchange or who to go to to sell their surplus. Even disregarding those immediate information problems there will be price stickiness. A man who is used to exchanging a cow for two pigs will not change his price until he knows that he must. He will on change his price after having being refused the normal price a number of times, and he may not relate that exchange ratio to the one between cows and chickens. As a result of all this compared to a market economy with the same capital a barter economy will underperform.

What we should recognize is that this isn't a separate issue to monetary disequilibrium. It isn't true that barter economies have this problem of "double coincidence" and that monetary economies have this unrelated problem of "fluctuating demand for money and sticky prices". In the barter case there are many instruments of indirect exchange being used in particular local areas which are, because of knowledge problems, trade at prices quite far from equilibrium. That causes a constant drag on the economy. This is essentially similar to the case of a money economy, except the problems are dispersed over time rather than concentrated. Each good used in a locality for direct or indirect exchange in a barter economy is subject to knowledge problems and the price stickiness of the goods it is traded for. When we talk about the double coincidence of wants problem we're talking about a very similar the demand for money but on a different scale.

(Note: here I'm considering the situation before issue of money substitutes by banks occurs, I'm not saying that the problems of a money economy can't be solved.)

That helps, George. My motivation is to understand this stuff/ I know I'm not all the way there yet, but getting closer.

A bit more drilling on two things you say might help.

I think I know the answer to this question, maybe. Economic thinking and the economic valuation is forward looking -- why does it make sense to take a backward looking "historic cost" perspective here? In other words, why do we want to cover past sunk costs? Why does economic coordination require us to look backward, rather than forward?

George writes:

"["general glut"] means merely that production is artificially constrained by lack of means of payment. Relaxing the constraint by expanding the nominal money stock serves, not to make goods drop like manna from heavan, but merely to allow them to be purchased at prices that cover historic costs."

Here's another I'd like to drill down on.

George writes:

"general glut" means that the excess demands for all goods sum to a negative number. That entails no uniformity of excess demands at all. Some could be positive."

This means that money reserves have increased relative to purchases and output streams -- and this is what people wanted, right?

Were playing off the insight that the total quantity of money doesn't matter (it could be anything) with the insight that the cash reserved, liquidity, debt position, risk position, and leverage demanded by different folks varies over time.

I think I've forgotten something basic right here -- or botched up everything -- but what is it, or how?

This seems wrong.

If the demand for money goes up in different ways for many and non-identical individuals, the the demand for different goods will change, and investment positions and taste for leverage and credit will change.

When the demand for money changes there is a need to shift resources.

Bill writes:

"What is special about an excess demand for money and a matching "general glut" of goods, is that there is no need to shift resources. If the process that brings the market to equilibrium is a change in the price level so the real quantity of money adjust to the demand, then any decrease in quantities are just errors."

The comments to this entry are closed.