September 2022

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

« If You Can't Get Enough of Debates over Fractional Reserve Banking | Main | On Keynesian Economics and the Economics of Keynes »


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

I was curious if he would put together a response. Thanks for the link.

I've read Selgin's draft after reading Bagus and Howden.

Even if I think Bagus and Howden may not provide too much new critics to FRB in their paper, I found Selgin's piece very worth reading.

I getting getting errors with that link for Selgin's response, even using a proxy. "Bad Request (Invalid Hostname)"

SSRN went down for maintenance shortly after my paper was posted there--drat! So keep checking; with luck it will be available again soon.

Same here. Now I am left alone with an intuition of the "pissed off brilliance" of Professor Horwitz.

SSRN is up again.

I remain mystified as to how complex theories involving the demand for money can be sustained without recognizing that the demand for money cannot logically be denominated in dollars alone, but must also include a consideration of the time of possession of those dollars. (dollar-days per month)

For the supply of money to mean anything, it must be true that money that I hold cannot be simultaneously held by any one else. My demand for money must limit everyone else's.

If $1000 out of my imaginary paycheck goes to paying my rent, the effect on my demand for money depends critically on exactly when the rent is paid, whether one day or thirty days after the receipt of my paycheck.

For this and other reasons, most individuals have the ability to adjust their demand to hold money
over a wide range at little more than whim.

Purchasing power resides not in money alone, but also in any asset that can be relatively easily converted into money by exchange.

Any policy that has the demand for money as an input variable is highly unlikely to be a stable and successful policy.

Regards, Don

Don Lloyd, what is the point of your observations? What are you talking about? What "complex theory" is it to which they are directed? What "policy"? Everything you say in your comment is a commonplace of the received their of money demand, so I wonder who it is that you imagine thinks otherwise, and what your reasons are for thinking so.


I didn't get very far in the original paper, and only read your response this morning.

I think it is pretty clear that the development of an interbank clearing market is in the interest of banks and that cooperative efforts to improve that market are possible, and that the result of this would be a reduction in the demand for bank reserves and so a higher price level. This could easily be episodic, and so the occassional shifts in the price level disruptive.

Yes, from reading your response it seems that the original authors were focused in the implausible scenario of a coordinated expansion of credit. (I will read it too.)

It seems to me reasonable obvious that they are assuming that an increase in the demand for money results in an immediate decrease in the price level (at least sometimes) and so an increase in the real quantity of money. And then, from this new equilibrium price level, the credit creation causes an increase in the price level, returning the price level to its initial level.

I agree entirely with your discussion of Wicksell. If the real quantity of money equal the demand to hold it, the market rate equals the natural rate. If prices are perfectly flexible, the market rate can't be below the market rate regardless of what is happening to the nominal quantity of money. But certainly, this is controversial. That somehow, injection effects distort relative prices even when prices are prefectly flexible. Some argue that, right?

In his response to B & H, did Selgin meant to write: "Consequently, critics of 100-percent reserve banking must make up their minds. They cannot have their cake and eat it, too." Or did he mean "...critics of free banking.."?

"Any economist with a heartbeat..." Ouch!

George Selgin,

I am encouraged to find that everything in my comment is a fully accepted part of the theory of money demand, but the lack of ever seeing any indication that the units of money demand are not money, but rather a rate of the time of possession of money gives me pause.

The 'theory' and the 'policy' to which I allude is the belief that the demand for money must be responded to by a change in the supply of money by some variety of an institutional agent or array of agents.

Even if the demand for money were reliably reported on daily stone tablets delivered by Moses, it still seems likely that individuals are in a much better position to adjust their own demands for money within the limits of their abilities and assets.

But the demand for money is likely even more of a problematical quantity than the supply of money.

If you would fight global warning by having a bureaucrat control a space-based variable density solar filter in response to daily newspaper snowfall reports, maybe it's all OK.

Just to mention something further that is obvious : when credit is expanded, the new loans that will be extended will almost always be more risky and imprudent loans than the ones that were funded by the original level of credit.

Thanks, Don

I read the paper.

I think there is a tradition from Mises to treat this proportional increase in bank liabilities so that the absolute level of reserves are not depleted. Thos of us with an orthodox background often start with thinking reserve demand as being in proportion to the issue of bank liabilities--you know, a reserve ration. If the demand for reserves is 10%, 5%, or 1% of liabilities, then, of course, an increase in issue raises reserve demand.

If you assume that the demand for reserves is a dollar amount, say $1,000,000, then an increase in lending by all banks in proportion could leave actual reserve holdings unchanged.

Of course, Selgin argues that neither is correct, but rather the demand for reserves is a function of the variance of payments, which depends on total expenditures.

Anyway, I am not sure why one would expect that the demand for reserves would be a specific dollar amount, and if not, then this entire "all banks increase in proportion" thought experiment is pointless.

Iwaaks: that's a typo, and thanks for alerting me to it.

Don, demand for any asset necessarily has a "time dimension" people demand assets until they don't demand them any longer! Money's not special in that sense. s for bureaucratic control of M, I am no fan of it, and neither are Bagus and Howden, so again, to whom are your remarks directed (give the topic of this posting?). I do believe it is sometimes desirable for M to grow; but I also argue that under free banking the growth happens more or less automatically. All that's necessary is that bankers maximize profit. If there's a part of the theory you don't like, perhaps you can refer specifically to it rather than interject seemingly unconnected obiter dicta.

Bill: I understand that banks will strive to economize on reserves; I assume as much in supposing that they set up clearinghouses and settle multilaterally after netting. But B&H aren't making the banal observation here. They offer three specific scenarios, and those are what I criticize. As for other possibilities--well, of course, there are others; and perhaps some "might" result in sudden and substantial reductions in base demand. But this is one of those observations that merits a "so what"? along the same lines as "a gold standard might allow for rapid inflation owing to new discoveries" and "the public might stage bank runs simply because they think runs are about to happen; and so on." The question is, what empirical grounds are there for fearing those possibilities? Can you think of an inflation caused by a private clearing innovation? I can't.

At very least, if we are to judge arrangements by what is "possible," let us agree that, since a discretionary fiat regime is one in which any M adjustment is "possible," free banking clobbers it on this score! (I know you agree, by the way.)

Finally, do re-read B&H. You take them much more seriously than they deserve. The article's arguments are not based on subtle distinctions: it is a botched attempt at a hatchet job.

From Bagus and Howden (2010):

"Selgin (1988, p. 55) invokes a monetary misperception argument, also used by real business cycle theorists."

My inclination is to find a charitable reading. But I am afraid it would be quite a stretch.


The time dimension that matters for money is the scarcity of exclusive ownership. The same would be true of other goods or assets, but only if they also are purely exchange valued, provide no benefit to society of a larger supply, and have existing stocks that are devalued by a larger supply. The above is true even if we set aside the question of the sign of the net possible benefits of a dynamic increase in supply.

I may well be wrong, but my perception is that all free bankers believe that there is merit in actively trying to adjust money supply to money demand. The demand to hold money is not a response to a single purpose, and it is not at all clear that a supply shortage vs demand will be reduced by an increase in supply. For example, if an increase in supply increases prices, my paycheck will run out earlier, and my demand for holding money will increase as a result for the part of money demand that is so affected.

Thanks, Don

Don, I'm sorry, but you are arguing about monetary economics without apparently having mastered the basic principles of the subject; and a blog isn't the place for learning those or for expecting someone to go over them. Suffice to say that when an increase in M occurs in response to an equivalently increased demand to hold money, the monetary growth doesn't raise prices.

Since Bagus and Howden commit the same crude error in their article, and I point the fact out there, it also appears that you haven't read my comment on the paper. Why not read it, learn what free bankers really are arguing, and then decide whether we are saying things that don't make sense?


Thank you. I have downloaded your comment, and am finding it of interest. A specific question, if you don't mind --

“Bank borrowers,” I observe there, “generally acquire money balances [“bank liabilities”] only to spend them immediately on goods and services. The demand for money, properly understood, refers to the desire to hold money as part of a financial portfolio. A bank borrower contributes no more to the demand for money than a ticket agent contributes to the demand for plays and concerts…” ..."

Your quote above sounds reasonable, but does it go far enough? The funds loaned and spent will presumably eventually end up in someone's money holding even if it has to first make multiple trips through banks.

Thanks, Don

The comments to this entry are closed.

Our Books