Why Deflation Is Not Inevitable (Sadly) Part 3: Why Currency Withdrawals Do Not Matter

Recently by Gary North: Keynes and Bernanke on Bubbles and Manias: Blame the Free Market

In Part 1, I replied to the issues raised by John Exter a generation ago:

1. The coming rush for liquidity2. Gold as the most liquid asset3. Currency and T-bills will do almost as well as gold4. Gold as money, despite a commission5. Asset prices influence consumer prices

In Part 2, I went on to deal with the myth of price deflation in Japan. First, there has been no monetary deflation. On the contrary, there has been mild monetary inflation (M2). Second, there has been price stability in Japan, despite massive declines in the real estate market and the stock market after 1989.

In response, I received a letter from Mike Rogers, a contributor to Lew Rockwell’s site. He has lived in Japan for decades. He is the author of a book, Schizophrenic in Japan. Here is what he wrote.

I read your article. You are absolutely correct. There is no deflation in Japan. The average Joe-Blow on the street uses what we in Japan call the “Bento-index.” Bento is those boxed lunches that Japanese people pack and eat. The price of a good cheap bento in 1996: about Y500—Y700 yen (maybe you can find a Y450 place — I knew of one. It was good too!) The price of a good cheap bento in 2010: about Y500—Y700 yen (maybe you can find a Y450 place — I know of one. I ate there last Friday. It was good too!)

He added this piece of evidence that I have seen nowhere else. There has been a change of consumer buying habits in Japan. Japanese consumers have become price-sensitive.

One thing you might want to know: It seems, as opposed to many years ago, only cheap things sell well in Japan anymore. So maybe people’s perceptions have changed. A good example: There used to be no such thing as Dollar Shops or discount clothing (on the level of a K-Mart or something like that) in the late 1980’s. Now, they are everywhere and they are booming. Because these places, and places that sell discounted goods are popping up all over the place, maybe this fuels the perception of deflation.

This is a major change. For decades, the Japanese government restricted entry of American firms, such as Wal-Mart, which are highly price-competitive. Price competition was regarded as a threat to traditional Japanese business standards. This attitude made possible political and social controls favoring wage stability, promotion by seniority, and the practice of lifetime employment. In short, it made possible government laws that in turn made Japanese corporations more like government bureaucracies.

When price competition becomes prominent, consumer authority has more clout in the retail markets. This has forced changes in Japanese corporate policies. This pressure will continue to work its way up the corporate chain of command.

This brings me to Part 3: currency’s role in the inflation vs. deflation debate.


A piece of paper money that is not a legal warehouse receipt for a specific quantity and quality of gold or silver is fiat money. All paper money systems today are fiat money systems.

The question is: How is a paper dollar issued? It has these words on it: “Federal Reserve Note.” This tells us where the money came from.

Paper dollars are printed by the Treasury Department’s Bureau of Engraving and Printing. The Treasury runs the BEP. You can see this on the BEP’s site.

What is not generally known is that the Treasury does not issue currency directly, on its own authority. It acts as an agent of the Federal Reserve System. We read in the Federal Reserve’s FAQ list on currency:

Does the Federal Reserve produce bank notes and coins?

No, the Federal Reserve does not produce bank notes or coins. The Bureau of Engraving and Printing (BEP) produces currency and stamps, and the U.S. Mint produces our nation’s coins. The Federal Reserve issues Federal Reserve notes and places them in circulation.

How are coins and currency put into circulation?

Coins and currency are placed into circulation through depository institutions, which obtain coins and currency from their Reserve Banks. The Federal Reserve, the Bureau of Engraving and Printing (BEP), and the U.S. Mint do not provide coins and currency directly to the public for circulation.

This is crucially important information for understanding how currency affects the money supply.

There is relatively little paper money circulating inside the United States. It is well known that paper money is sent back home by immigrants. This money is withdrawn from circulation here. Estimates on the percentage of the total U.S. currency supply held outside the United States range from 50% to 70%. The Federal Reserve and the Treasury estimated that in 2005, this was about 60%. (

With this in mind, consider the words of a booklet produced in the 1970s by the Federal Reserve Bank of Chicago, Modern Money Mechanics. It is no longer in print, but it is on the Web. The booklet begins its discussion of currency with a widely shared fallacy.

Money has been defined as the sum of transaction accounts in depository institutions, and currency and travelers checks in the hands of the public. Currency is something almost everyone uses every day. Therefore, when most people think of money, they think of currency.

Yet relatively few transactions take place through the use of currency. That was true 30 years ago. It is far more true today, because of the use of credit cards. When the booklet was written, customers used checks.

Contrary to this popular impression, however, transaction deposits are the most significant part of the money stock. People keep enough currency on hand to effect small face-to-face transactions, but they write checks to cover most large expenditures. Most businesses probably hold even smaller amounts of currency in relation to their total transactions than do individuals (p. 14).

The booklet then describes what happens to the total money supply when currency is withdrawn from a bank by a customer.

When deposits, which are fractional reserve money, are exchanged for currency, which is 100 percent reserve money, the banking system experiences a net reserve drain. Under the assumed 10 percent reserve requirement, a given amount of bank reserves can support deposits ten times as great, but when drawn upon to meet currency demand, the exchange is one to one. A $1 increase in currency uses up $1 of reserves.

Suppose a bank customer cashed a $100 check to obtain currency needed for a weekend holiday. Bank deposits decline $100 because the customer pays for the currency with a check on his or her transaction deposit; and the bank’s currency (vault cash reserves) is also reduced $100.

Now the bank has less currency. It may replenish its vault cash by ordering currency from its Federal Reserve Bank — making payment by authorizing a charge to its reserve account. On the Reserve Bank’s books, the charge against the bank’s reserve account is offset by an increase in the liability item “Federal Reserve notes.”

The national money supply does not change immediately. “The public now has the same volume of money as before, except that more is in the form of currency and less is in the form of transaction deposits.” Things do not remain constant, however. The withdrawal reverses the fractional reserve process.

Under a 10 percent reserve requirement, the amount of reserves required against the $100 of deposits was only $10, while a full $100 of reserves have been drained away by the disbursement of $100 in currency. Thus, if the bank had no excess reserves, the $100 withdrawal in currency causes a reserve deficiency of $90. Unless new reserves are provided from some other source, bank assets and deposits will have to be reduced (according to the contraction process described on pages 12 and 13) by an additional $900. At that point, the reserve deficiency caused by the cash withdrawal would be eliminated.

This means that the bank must call in loans. It has to get its new, lower reserves balanced by a reduction in its assets: loans on its books. Or maybe it refuses to make new loans as old loans are paid off. It may have to borrow money in the federal funds market until it can reduce its loans.

This means that the withdrawal of currency is deflationary. This usually does not matter, because the person with the currency spends it at a business establishment. The business then deposits the currency at the end of the day. As the booklet says, “When Currency Returns to Banks, Reserves Rise.”

After holiday periods, currency returns to the banks. The customer who cashed a check to cover anticipated cash expenditures may later redeposit any currency still held that’s beyond normal pocket money needs. Most of it probably will have changed hands, and it will be deposited by operators of motels, gasoline stations, restaurants, and retail stores. This process is exactly the reverse of the currency drain, except that the banks to which currency is returned may not be the same banks that paid it out. But in the aggregate, the banks gain reserves as 100 percent reserve money is converted back into fractional reserve money.

The money supply then increases: “Since only $10 must be held against the new $100 in deposits, $90 is excess reserves and can give rise to $900 of additional deposits” (p. 17).

So, if there were a run on the banks for currency, this would be deflationary. This was why the U.S. government created the Federal Deposit Insurance Corporation (FDIC) in 1934. It created a similar agency for savings & loan associations. Ever since 1934, there has been no threat to the total money supply from currency withdrawals. The Federal Reserve can use open market operations to offset any decline in the money supply. It can buy assets. As Modern Money Mechanics says,

To avoid multiple contraction or expansion of deposit money merely because the public wishes to change the composition of its money holdings, the effects of changes in the public’s currency holdings on bank reserves normally are offset by System open market operations.


Robert Prechter is the most widely known promoter of the Elliott Wave theory. He is a deflationist. He has been consistent, unlike most deflationists, who promote gold ownership. In 2004, he said that gold at $450 would be the top.

I turned bearish early but I also said the upper limit was the low US$400 [per ounce] and that is where it stopped and so far I am very comfortable with everything we have been saying.

He also said this:

Gold is in a bear market so it’s not a good thing to own, but I think everyone should have some. In a bear market governments get desperate and it’s happened many times in history that they will seize gold, they will make it illegal so people can’t buy it, so if you already have some you are in much better shape.

Why owning gold would be better than owning paper money, which the U.S. government has never outlawed, he did not say. Gold falls in price in price deflation. Paper currency appreciates, income tax-free: more real wealth, no taxes.

On December 18, 2009, he wrote the following article: “The Fed’s Presumed Inflation Since 2008 Is Mostly a Mirage.” In this article, he did not mention excess reserves deposited at the FED’s regional banks by commercial banks. These reserves came close to offsetting the increase in the FED’s holdings of assets in its monetary base, which was the result of the bailouts in the fall of 2008. Instead, he said that the FED has offset its own policy. He did not say how this was possible.

When the Fed buys a Treasury bond, net inflation occurs, because it simply monetizes the government’s brand-new IOU. But in 2008, in order for the Fed to add $1.4 trillion new dollars to the monetary system, it removed exactly the same value of IOU-dollars from the market.

This is not true. The increase in its assets still is listed in its balance sheet.

Prechter continued:

It has since retired some of this money, leaving a net of about $1.3 trillion.

This is also not true. The FED does not “retire money,” as if money were some aging race horse put out to pasture. It sells assets, which destroys money in a fractional reserve banking system. It has not done this, as its balance sheet indicates.

Prechter then brings up the issue of currency.

In currency-based monetary systems, the creation of new banknotes causes — indeed forces — inflation.

This is irrelevant. The United States operates in terms of debt money created by its purchases of debt. Very little currency circulates inside the United States. When this increases, the money supply falls, due to the reversal of the fractional reserve process.

Likewise, the monetization of new government debt creates permanent inflation practically speaking. (Theoretically, the government could retire its debt, but it never does.) But when the Fed simply swaps money for previously existing debt, there is no net change in the amount of dollar-based “purchasing power” on the planet.

What does he mean, “swaps money”? The FED swapped assets at face value: liquid Treasury assets for illiquid toxic assets. That did not change the money supply because money was not involved. When the FED swaps money for a previously issued debt, it must create the money used to swap. It does not have “money” in reserve.

By the way, the phrase “net change in the amount of dollar-based ‘purchasing power'” is meaningless. The issue is this: Does the monetary base increase when the FED “swaps money for previously existing debt”? The answer is yes. To deny this is to deny both money and banking practice and theory.

So investors, who previously held $1.3t. worth of IOUs for dollars, now hold $1.3t. worth of dollars. They are no longer debt investors but money holders. The net change in the money-plus-credit supply is zero.

This is incoherent. This is why you cannot follow it. It is not your fault.

Investors are not money holders unless they hold currency or a liquid account in a bank. The total of these is M1. There was no increase in bank reserves held by the public. That is because of excess reserves. So, this leaves currency. There was not $1.3 trillion in currency to hand over to them. Currency is held by the public or in bank vaults as reserves. Conclusion: there was no offsetting swap by the FED of money for previously existing debt.

Prechter went on:

The Fed simply retired (temporarily, it hopes) a certain amount of debt and replaced it with money. In fact, if the Fed is to be believed, it desperately wants to sell the rest of these (in)securities and retire the new money. I doubt it will happen, but it doesn’t much matter to inflation either way.

This is also incoherent. The FED cannot “retire debt” and thereby “create money.” The only way that it can create money directly is to buy debt. Indirectly, it can create money by lowering the legal reserve requirement or by imposing a fee on excess reserves. So, the only two ways for the FED to offset its own actions in 2008 are these: (1) increase legal reserve requirements; (2) sell all of these assets. It has done neither.

The economic effect of the commercial banks’ voluntary increase in excess reserves is the same as an increase in the legal reserves. But the banks determined this; the FED did not.

Prechter is clearly confused about how the FED works or what the FED has done. Yet he has been publishing ever since 1979.

He then compounds his error.

The theory among monetarists is that these new dollars are hot money that creditors can now re-lend. Thus, it will multiply throughout the banking system. At first it might seem that new money in banks’ hands should be more powerful for creating inflation than the previously-held FNM mortgages. But this is not the case, because the main thing for which the Fed wants banks to lend out the new dollars is new mortgages.

Monetarists and Austrians and all other schools of economic opinion say that banks can legally lend any reserves made available by central bank purchases of debt — new debt or previously existing debt. (The word “re-lend” has nothing to do with anything.)

The banks did receive the money spent by the FED when it purchased assets in 2008. Bankers have decided not to lend this money. This stopped the fractional reserve process, just as an increase in the reserve requirement would have stopped it.

The FED has no policy of banks’ buying mortgages. It buys Fannie Mae and Freddie Mac bonds for its own account, but it has remained silent on who banks should or should not lend to. In any case, banks can legally lend to any borrower they choose. Raising the issue of mortgages merely confuses the issue. What issue? The issue of excess reserves: the bankers’ fear of buying anything, including T-bills.

Today, bankers and other creditors are afraid of mortgages, and they don’t want new mortgages any more than they want the old ones. In the mortgage-intoxicated, pre-2008 world, there would have been little significant difference in the paper, because banks were creating new dollars any time they wanted by taking on new mortgages. In the mortgage-repelled, post-2008 world, guess what: there is still little significant difference in the paper, because virtually the only thing banks can use it for is to fund mortgages! The only other outlet for the Fed’s new money is to fund market speculation, which is one reason why the stock and commodity markets rose.

Got that? “. . . virtually the only thing banks can use it for is to fund mortgages!” This is not true. The exclamation point makes it even more untrue.

It is very important that people who talk about Federal Reserve policy understand how money and banking work, both in theory and practice. Prechter does not.


The forecasters who predict inevitable price deflation, yet do not predict inevitable monetary deflation, do not understand monetarism, or Austrian School economics, or supply-side economics. They are not economists. They are journalists.

There can be monetary deflation. That could result from deliberate FED policy: increased legal reserves or sales of assets. It could result from the collapse of banks, due to the unwillingness of the FED to intervene to provide Congress with the fiat money necessary to fund the FDIC. Neither policy is inevitable.

January 14, 2010

Gary North [send him mail] is the author of Mises on Money. Visit He is also the author of a free 20-volume series, An Economic Commentary on the Bible.

Copyright © 2010 Gary North