Bryan Caplan  

Arnold, Money, and Nominal GDP

Modern Finance vs. Portfolio B... The Productiveness of Conversa...
Out of all of Arnold's macroeconomic views, only one strikes me as truly absurd: His skepticism about the ability of central banks to affect nominal GDP and other nominal variables.  The latest example:
[A]n increase in the supply of money will have neither the consequences predicted by Keynesians or by monetarists. In fact, the central bank does not even control the rate of inflation. Markets on their own can increase or decrease liquidity through innovation and adaptation. In terms of the equation MV = PY, whatever M the bank chooses to control, markets will take steps to move V in the opposite direction.
Frankly, I don't see why anyone would even start to hold Arnold's view.  The quantity theory of money is extremely intuitively plausible, as Hume's famous thought experiment shows.  And all of the clear-cut historical examples of large increases or decreases in the money supply support the view that large changes in the money supply change nominal GDP roughly proportionally.  I agree that matters are messier when you look at small changes in the money supply and short-term fluctations.  But if a theory's intuitive and passes the clean tests, why wouldn't you just embrace it?

Comments and Sharing

COMMENTS (10 to date)
Jon writes:

The basic question is what is 'M'. If you take the broad Austrian view, 'M' is any money substitute. So if the scope of money substitutes has come to be large because for instance treasury bills and notes and other AAA securities are regarded as good as money (they are certainly treated as such for collateral purposes). Or, if you regard secured bank deposits as being as good as money, i.e., CDs, acceptances, etc, then the Fed controls only a tiny portion of M.

Ordinarily, a healthy base-multiplier obviates this problem, but when banks are discouraged from lending by regulators, the minority position of base money with 'M' comes to the fore.

8 writes:

In the Roving Cavaliers of Credit, Steve Keen aruges that the central bank follows the banks in creating money supply.

Thus rather than credit money being created with a lag after government money, the data shows that credit money is created first, up to a year before there are changes in base money.

Mike Sproul writes:


Correction: The Quantity Theory is extremely IMPLAUSIBLE. When any bank, central or otherwise, issues a dollar, it gets a dollar's worth of assets in return. Bank assets rise in step with bank liabilities, so there is no change in the value of money. Every economist agrees that when a firm issues new shares of stock, the firm gets equal-valued assets in exchange for those shares. Firm assets rise in step with the firm's liabilities, so there is no change in the share price. That's true of all financial securities, and it's true of money.

The quantity theory implies that the money-issuer's assets don't matter. If that's the case, then somewhere in the historical record we should be able to find some example of a government that issued money (of positive value) without holding assets against that money. I don't know of any such case, and I don't expect to ever find one.

Joe in Morgantown writes:

These are interesting rebuttals to Bryan's mainstream view. Let me summarize the rebuttal to see if I get it.

What is money? Money is a set of financial instruments including cash, checking accounts, AAA bonds... etc.

When the Fed "creates" money it typically exchanges a bank balance for AAA securities. This creates nothing; the Fed and its counter-party have done a swap, but the volume of "money" in circulation is unchanged. Granted, some of it has just morphed from a money market/repo account into a bank account, but so what?

On the other hand, when I go to lunch and pay with a credit card money /is/ created. $10 that didn't exist before shows up in the restaurant's checking account.

I like it. Well, off to lunch. (Should I pay cash?)

Mike Sproul writes:


The creation of a credit card dollar is no different from the creation of paper or checking account dollars. You buy lunch with a credit card. The credit card company issues one credit card dollar to the restaurant, while the credit card company gets your IOU as an asset.

In 1710, the popular view was that paper money was not 'true' money, since a paper bill was ultimately paid in coin. In 1840, the popular view was that checking account dollars were not money, since every checking account dollar was ultimately paid with a note or coin. Since 1950, the popular view has been that credit cards aren't money, since every credit card dollar is ultimately paid with a check, bill, or coin. In every case, people forgot that there was a permanent float of the new kind of money that was never really paid off.

I predict that in 100 years, economists will count credit card dollars as money, but there will be some new kind of money that will be dismissed as an "economizing expedient" or "money substitute".

Joe in Morgantown writes:


When the Fed "prints" money, it matters what it gets in return, right? It could create money by giving away newly printed cash (or debit cards or bailouts). But if it buys US Bonds--- which were already being used as money in repos--- then no money is created on net.

This "the Fed makes no cents" thesis seems very plausible to me; in fact it seems to follow naturally from the credit-is-money observation.

If it is so, all the central banks' extensive efforts at "sterilization" are utterly without effect. That seems like it should be testable.

Mike Sproul writes:


The question isn't whether the fed creates money. What matters is whether the fed's assets move in step with the money it creates. It doesn't matter if the fed has assets worth 100 oz. of silver backing 100 paper dollars, or assets worth 200 oz backing $200. $1=1 oz. either way.

Thomas Cunningham tested that idea in "Some real Evidence on the Real Bills Doctrine...". Similar studies have been done by Sargent, Siklos, Calomiris, B. Smith, and Bomberger & Makinen, to name a few.

rjw writes:

why would one reject the quantity theory ?

three reasons :

- Money is not exogenous. The creation of money normally arises as a product of credit growth, accommodated by the central bank. So the causality assumption embodied in usual presentations of the quantity theory is highly suspect.

- The velocity isn't stable. Money demand equations broke down long ago. (And which definition of 'money' do we mean anyway? The QT gives no guidance)

- Output is not necessarily fixed, so a rise in MV (demand) would not in any case necessarily translate into a rise in prices, but could also affect output.

I expect I could think of others, but those ones seem fairly damning. One might also wonder no central bank rigorously and strictly targets money, if the QT is SUCH a no-brainer. I suspect there is a good reason.

rjw writes:

As a quick addendum to my post just above ... of course, one CAN point to certain historical episodes ... the effects of the flow of gold into spain in the early modern period is one ... where something like the quantity theory might be deemed a decent explanation. But that case is rather different from the case of a current monetary economy.

Also, under fixed exchange rates, one could make a rather stronger case for exogeneity. Although, as I understand it, the history of the classical gold standard in the late 19th century have shown that a variety of adjustments mechanisms were in fact in play, and the price-specie flow adjustment process rarely operated as the usual parable suggests.

Eye of Siva writes:

Take a look at II.V.27 from your own link to Hume discussing how the innovation of standing credit lines changed the velocity. I quote: "By this means, a stock of five thousand pounds was able to perform the same operations as if it were six or seven."

The private sector will innovate the use of money stock and I can see why Arnold is right.

I am from India and in some parts of the country it is difficult to come across clean notes. Torn notes are stapled together (sometimes wrapped in plastic sheets). Often the stapled parts do not come from the same note or even of the same denomination. In such cases, the practice is always to use the HIGHER denomination piece as the value of the note. What is the quantity of money in this case?

Comments for this entry have been closed
Return to top