November 24, 2007
Inflation, we are told, comes from “printing money.” Well, maybe it does, and maybe it doesn’t. The value of a currency can be interpreted as the intersection of supply and demand for that currency. If currency is supplied in a fashion vastly in excess of demand, then the value of the currency will fall. So, you could say that “printing too much money causes inflation.” OK, how much is too much? For hundreds of years, people have hypothesized that some steady, stable rate of growth would be appropriate. Milton Friedman once said that 3% money supply growth per year should be made a Constitutional amendment — which merely demonstrates that Friedman, the so-called “monetarist,” didn’t understand much about money.
However, Friedman at least understood that a growing economy, and growing population, will tend to require more money to do its business, just as it requires more electricity, more pairs of shoes, more automobiles, and so forth. There is an even more retrograde line of thinking, particularly among “gold bug” type hard-money cranks, that claims that printing any money at all is inflationary. (A fair number of these types will also claim that any expansion of credit at all is also inflationary! Which explains why nobody listens to the “gold bugs.”) Well, someone must have printed it, or it wouldn’t exist today. Also, there are many long and enduring examples of currencies that remained linked to gold for many decades and even centuries.
Let’s take one example: the US dollar. It was repegged to gold in 1879, following a floating-currency period that began with the Civil War. In August 1880, with the dollar worth $20.67 per ounce of gold, the U.S. dollar monetary base was $897m. In August 1932, with the dollar worth $20.67 per ounce of gold, it was $7,850m. That’s a 775% increase in the monetary base, in response to increasing demand for money for use in transactions, resulting in a stable, gold-linked value of the currency. In 1775, the estimated total monetary base in the U.S. was $12m — with those $12m trading near $20/oz. of gold. So, from 1775 to 1932, the U.S. monetary base increased by about 65,300%, while remaining pegged to gold (at least on a point-to-point basis).
In 1934, the dollar was repegged at $35/oz. The monetary base was $9,491m. Here’s something to notice: between 1932 and 1934, the dollar’s value fell by 43%, or to put it another way, the “price of gold” rose by 69%. Did the base money supply rise by 69%? No, over two years it rose from $7,850m to $9,491m, which is an increase of only 20.9%. Although that is still a fairly high rate of growth on an annualized basis, the devaluation was not accomplished by wanton “money printing.”
In August 1970, with the dollar still worth close to $35/oz. of gold, the monetary base was $63,725m. That’s an increase of 571% in the monetary base, with no loss of currency value.
In February of 1980, the dollar reached its nadir of 1/850th of an ounce of gold. The monetary base was $132,785m, an increase of 108% from 1970. During the 1960s, the monetary base increased by 56%. The increase in the 1970s was certainly greater than the 1960s, but the decline in the dollar’s value by a factor of about 18 times (1/35th oz to about 1/650th oz average for 1980) is in no way proportional to the somewhat modest increase in “money printing” during that decade.
What happened in the 1970s? Why could the monetary base increase by 653x in the 1775-1932 period with no loss of value, but the modest 52% increase of base money during the 1970s (as compared to the 1960s) be tied to a currency collapse?
What about today? For October 2007, base money was up only about 2.0% from October 2006 — according to the Fed’s statistics. This is about the lowest rate of “money printing” in the last forty years. They could be lying, of course. But, the figures are plausible. So much for the “wild money printing” supposedly going on during this time.
Occasionally, governments do get into the practice of printing press finance, with predictably disastrous results. The results are so predictable and so disastrous that this almost never happens today.
What happens instead is a collapse in demand for a currency as a response to apparent currency mismanagement. A currency falls when supply is in excess of demand, which can happen even when the supply (monetary base) is actually declining! But this is no surprise, since the dollar is widely seen as a mismanged currency. The rather obvious decline in dollar value over the past several years has produced no indication of concern among the dollar’s managers. Why hold onto dollars? Thus the collapse in demand.
The rate of “money printing” (base money expansion) was a little high in the 1970s, but not by much, and basically in line with the disinflationary 1980s and 1990s.
The dollar is an international currency. Indeed, most dollars (banknotes) are actually held outside the United States. The Fed estimated, in 2000, that two-thirds of all US currency is held internationally. Thus, the “demand” is worldwide, making comparisons to domestic-only government statistics all the more meaningless. The dollar is now, apparently, in the process of becoming a much less international currency. People outside the US are not as interested in holding dollar bills. If they reduce their holdings by half, replacing them with euros for example, then the overall demand for dollars would fall by 1/3rd, and to maintain a stable currency value, the supply would also have to contract by 1/3rd. Thus, the real danger here is not “money printing” but the failure of the currency managers to respond to declining currency value and declining demand for currency. The demand falls even more when it is apparent that the currency managers are asleep at the wheel.
What about M3? It seems to be the indicator of choice these days for people who want to correlate “money printing” with inflation. Well, what is M3? It is the monetary base plus dollar bank accounts and CDs, money market accounts and repurchase agreements. In essence, it’s a measure of bank credit. Now, why should a money market account (an equity shareholding in an investment company that holds short-term debt) be counted as “money” but short-term debt itself is not counted as money? Why should a CD, which is essentially a bond issued by a bank, and which can be of long maturity, be counted as “money” but a bond issued by a non-bank corporation is not? Why is a 1-year CD “money” but a 1-year Treasury bill not “money,” unless it is owned by a money market fund, in which case it becomes “money”? Why is a repurchase agreement, which is the purchase of some fixed-income asset with an agreement by the seller to buy it back later, money, while the purchase of a fixed-income asset without an agreement by the seller to buy it back later, not money? What does rising M3 really mean, anyway — greater willingness to loan to banks rather than to make loans to non-bank entities? Greater willingness to hold one’s investments in the form of short-term debt rather than long-term debt or equities? Greater willingness by foreigners to loan money to foreign banks on a dollar-denominated basis rather than in some other currency? We begin to see the problems when dealing with measures of credit rather than money, known as “base money”.
There is an intriguing spike in M3 growth in the early 1970s, which might be interpreted as inflationary. Also, there is an intriguing spike in M3 in the 1998-2001 period, when the dollar was very deflationary. There was a huge dip in the early 1990s, when there was a recession and a high dollar value, and another smaller dip in 2004-2005, which was a recovery and a declining dollar value. Conclusion? Looks “indeterminate” to me.
So, I think we’ve determined by now that “money supply” figures are largely meaningless by themselves. The interesting thing is the value of the currency, which reflects the balance of supply and demand. The only kind of money which matters in this balance of supply and demand, which determines value, is base money, i.e., real money.