The Inflation Crisis, and How to Resolve It

13

What Determines the Value of Money?

The Velocity of Money

We cannot fully understand the present American and world inflations, and the consequences to which they are likely to lead, unless we fully understand the causes that determine the purchasing power of money. Perhaps the greatest obstacle to this understanding today is the continued prevalence of an old but false theory.

The strict quantity theory of money and its “equation of ex-change” have dominated and distorted the thinking of even some of the most respected monetary economists. A striking illustration is Bresciani-Turroni’s discussion in his otherwise admirable history and analysis of the German hyperinflation of 1919–23, The Economics of Inflation.

Bresciani-Turroni treats the equation of exchange as an inescapable axiom. In his version it is not MV = PT, but “MV = p1a + p2b + p3c ... where M is the quantity of money issued, V the velocity of circulation, a, b, c, ... the quantities of goods exchanged and p1, p2, p3 ... the respective prices.”

When he finds that in the late stages of the German inflation (and in the late stages of practically all other hyperinflations) prices of goods did not rise in proportion to the increase in the quantity of money but at a far faster rate, he assumes that this must have been—that it could only have been—because the velocity of circulation increased sufficiently to account for the discrepancy.

His method of arriving at the supposed increase in the velocity of circulation is as follows. He first assigns the presumed velocity of circulation of money in Germany in 1913 an arbitrary base rate of one. He then compares, for each year or month after the inflation began, the number of times the German money stock was increased with the number of times that wholesale or retail prices increased. He then divides the price increase by the money-stock increase, and assumes that the quotient must represent the increase in the velocity of circulation.

For example, at the end of 1922 the currency circulation of Germany was 213 times greater than in 1913. Wholesale prices were 1,475 times greater. The cost of living was 685 times greater. Therefore, he concludes, in 1922 the velocity of circulation in wholesale trade must have increased 6.92 times and in retail trade 3.21 times.

He applies the same formula to each year from 1914 through 1918, and then to every month of the five years from October 1918 to October 1923. His derived velocity rate begins to go up rapidly from August 1922. For the last month on his table, October 1923, he gives the velocity of money in retail trade as 10.43 times greater than in 1913 and in wholesale trade as 17.79 times greater.

These velocity figures, in my opinion, are absurd and impossible. There are several ways of showing why they must be.

Let us begin with the truism, so astonishingly overlooked, that each individual or family income can be spent only once. This means that in a society with a given economic organization and division of labor the annual velocity of circulation from year to year cannot change very much.

Bresciani-Turroni nowhere mentions this. He thinks he can explain the huge increases in velocity of circulation that he assumes took place from month to month. He refers, for example, to the fact that some salaried employees received their pay only once every three months. Suppose, then, he argues, that at the height of the inflation, instead of spending their quarterly pay checks over each quarter, these employees spent the entire amount in the first few days after the checks were received? Would not this explain the increased money velocity?

There are several things wrong with such an explanation. First, those who were paid quarterly in the Germany of the early 1920s must have been a very small portion of the population. Second, it would not be easy to buy three months’ supplies of everything in the first day or two. A three-months’ family food supply, for example, could not be stored at home or kept fresh there. And if most of these quarterly payments or attempted expenditures fell on the same day, merchants would simply not have the goods in stock to sell.

Third, even if this kind of speed-up happened, it would not lead to a quarterly increase in velocity of circulation or even to much of a monthly increase. If a man spends his whole ninety-days’ income on the first day, he has nothing to spend on any of the next ninety days. The average quarterly rate of spending does not change. So if, at the height of the inflation, every family in Germany was paid daily, and spent the whole of each day’s income on the day it was received, then it spent one-365th of it every day instead of one-52nd of it every week. The monthly rate did not change much.

But there is still another and much more fundamental reason why Bresciani-Turroni’s velocity-of-money conclusions are unacceptable. The very phrase “velocity of circulation,” as I pointed out in the last chapter, embodies a false concept. Money does not literally circulate. This is a metaphor. Money is exchanged for goods and services. It is hardly possible to spend money without, by the same action, buying goods. (The borrowing or repayment of money loans constitutes a relatively small part of the total transfer of money, and so long as it does not increase or decrease the outstanding money stock it does not necessarily have much effect on the exchange value of the money unit.) Therefore it is hardly possible to speed up the velocity of circulation of money without speeding up to an approximately equal extent the velocity of circulation of goods. And if one does this (as Bresciani-Turroni himself admits) the exchange value of the money unit is not thereby depressed.

But in fact the sale of goods cannot be increased for any prolonged period beyond a very limited amount. (By a “prolonged period” I refer to anything beyond a couple of months.) This is true for the simple reason that the volume of goods for sale just cannot be increased by much in a short time. Bresciani-Turroni’s tables show the average velocity of circulation of money to have increased in the first nine months of 1923 to an average of 8.25 times that of 1913. But this would practically have to mean that the quantity of goods sold in those nine months—and therefore, in effect, the quantity of goods produced in those nine months—must have been 8.25 times as great as the quantity produced in the corresponding nine months of 1913.

This is not only incredible on its face; it is known to have been untrue in Germany in 1923. For by Bresciani-Turroni’s own account, production was disorganized by the inflation in 1923, and fell substantially.

There is still another factor that the assumption of a hugely increased velocity of money in a hyperinflation overlooks. In order for such an increase to occur, it is not merely necessary that the holders of money should be eager to get rid of it as quickly as possible, but that the sellers of goods should be correspondingly ready to part with their goods for money. But Bresciani-Turroni himself tells us, “The risk of transactions effected by payment in paper marks became so great in the summer of 1923 that many producers and merchants preferred not to sell at all, rather than accept in exchange a money subject to rapid depreciation.”1

It is instructive to notice that Bresciani-Turroni in the end distrusts his own figures and his own explanation. He carries his own calculations only up to October 1923, when, as we have seen, he estimates that the average velocity of circulation of money must have been some fourteen times as great as in 1913. But he tells us that “in August 1923 the value of the paper money in circulation amounted on some days to scarcely 80 million gold marks”2 (compared with 6,000 million in 1913). But on his own basis of calculation, as presented in his annual and monthly tables, this would require us to assume that on these days velocity of circulation must have been seventy-five times as great as the 1913 rates. Moreover, he also tells us that “on November 15th [1923]—on the eve of the cessation of the discount of Treasury bills by the Reichsbank—based on the official value of the gold mark (six hundred billion paper marks), the total value of the notes of the Reichsbank in circulation was 154.7 million gold marks. But based on the exchange rate of the paper mark in New York the total value was as low as 97.4 million gold marks.”

So based on the offical value of the gold mark, Bresciani-Turroni would have had to conclude that the velocity of circulation must have increased about thirty-nine times over 1913, and based on the paper mark exchange rate in New York, sixty-two times over 1913.

He draws no such conclusion and cites neither figure. Instead, he completely shifts his explanation of the decline in value of the paper mark. He then decides that “the increase in the velocity of circulation ... does not completely explain the very great reduction of the total real value of the paper money, ... for the place of the paper mark was taken by foreign exchange” and the return of metallic money to circulation.3

The “Cash Balance” Approach

I should like to add here that I not only regard an increase in the velocity of circulation as a totally false explanation of a rise in prices more rapid than the rise in the quantity of outstanding money in a hyperinflation, but that I consider an alternative explanation adopted by a number of economists—the “cash holdings,” or “cash balance,” approach—as also quite inadequate, especially in certain formulations.

Some economists formulate the cash holdings approach as follows: At the beginning of an inflation, prices generally do not rise as fast as the quantity of money is increased, because people think that prices have risen to unsustainable levels and will soon fall back to “normal.” They hold off many purchases and add to their cash holdings. This in itself keeps prices from rising as much as the quantity of money has been increased. But when people finally come to fear that the inflation is going to be prolonged, and that the rise of prices may go on indefinitely, they begin to buy in advance. They pull down their cash holdings. It is this action that increases the rate at which prices begin to rise.

There are two major defects in this explanation. One is that, even if otherwise correct, it would account only for a relatively small change in prices compared with the rate of monetary increase. Suppose people normally kept as an average cash balance the equivalent of 10 percent of their annual incomes, or roughly enough to spend over the next thirty-six days. If, in an inflation, they were willing to let their cash balances fall even to zero, this would only add some 10 or 11 percent to the total “active” money use. It could not account for the almost incredible fall in the purchasing power of the monetary unit, when compared even with the increase in the money stock, that occurs in a hyperinflation.

The other major defect in the cash holdings approach is that, no matter how much or often individuals decide to spend, the average cash holdings of all individuals in the country cannot be reduced! If a country has a population of approximately 200 million, and the total money supply is $800 billion (counting currency in the hands of the public, plus both demand and time bank deposits), then the average cash holding of each individual must be $4,000. The money must always be held by someone. What Peter spends, Paul receives. If half the people in the country, by increasing their spending, reduce their cash holdings by an average of $1,000 each, the other half must increase their cash holdings by the same average amount.

Subjective Value

What, then, is the basic explanation for the value of money, and for changes in that value?

It is the same as the explanation for the value of anything else. It is the subjective valuation that each of us puts on it. The objective purchasing power, or exchange value, of the monetary unit is derived from the composite of these subjective valuations. It is not, however, merely a physical or arithmetical composite of these individual subjective valuations. Individual valuations are themselves greatly influenced by what each of us finds to be the market, or “social,” value. Just as hydrogen and oxygen may combine to form a substance—water—that seems to bear little resemblance to either, so the social, or market, valuation of money as well as other things is akin to a sort of chemical rather than an arithmetical combination of individual valuations.

All valuation begins in the minds of individuals. We are accustomed to saying that market value is determined by supply and demand, and this is as true of money as of other commodities. But we should be careful not to interpret either supply or demand in purely physical terms, but rather in psychological terms. Demand rises when people want something more than they did before. It falls when they want it less. Supply is more often thought of in a purely physical sense, but as an economic term it also refers to psychic factors. It may vary with price. At a higher price producers may make more of a commodity, or be ready to offer more of the existing stock for sale.

When it comes to money, economists have been too prone to explain value in purely physical or mathematical terms. Hence the strange vogue of the rigid proportional quantity theory of money, of the algebraic “equation of exchange,” and of the alleged determining role played by the velocity of circulation of money.

What is overlooked is that the equation of exchange is a mathematical delusion. It is not an equation, as imagined, with money on the left side and goods on the right. There is no meaningful way in which all goods and services can be added to each other except in terms of their money prices. There is no meaningful way, for example, in which a pound of gold watches, a dozen square yards of cotton, a ten-room house, and a ton of sand can be added together except in terms of their individual prices in money. What we are adding is the amount of money required to buy them. Therefore the product of the equation of exchange, on each side, is a sum of money. These sums are equal because they are identical. The equation merely asserts that what is paid is equal to what is received. Neither the quantity theory nor the equation of exchange contains any proof of causation.4 And the number of times that a unit of money changes hands has no necessary connection with the “level” of prices.

What is called the cash balance approach is less fallacious than the mechanical quantity theory of money. It does contain an element of truth, but in some formulations it confuses cause and effect. It is true that when people think that the value of money is going to rise—in other words, when they think commodity prices are going to decline—they tend to spend less money immediately. And when they think the value of money is going to fall—that is, that commodity prices are going to rise—they tend to spend more money immediately. But the cash balance approach puts too much emphasis on a physical act and too little on the subjective change of valuation that prompts the act. The value of money does not decline because people try to speed up their spending; they speed up their spending because they think the purchasing power of their money is going to decline.

We can understand this better if we consider the purchase and sale of shares on the stock exchange. Suppose during a day’s session American Steel publishes an unexpectedly favorable quarterly earnings report, that 10,000 shares are traded in, and that the price rises from 30 to 40. The price has not risen because Jones, Smith, and Watson have bought 10,000 shares from Brown, Green, and Doakes. After all, as many shares have been sold as bought. The price rises because both buyers and sellers now estimate the value of American Steel shares higher than they did before. Suppose, again, that National Motors closed at 35 on Monday, that after the close the directors unexpectedly fail to declare the regular dividend, and that the stock opens Tuesday morning at 25. This sort of thing happens frequently. There have been meanwhile no sales on which to blame the decline. The stock has fallen in price simply because both buyers and sellers now put a lower estimate on it. This is precisely what happens with the value of money. It is changes in value estimates that count, not changes in cash balances.

And this is why, in the late stages of a hyperinflation, prices start to soar far faster than the supply of money is increased and even far faster than it can be increased. Nearly everybody is convinced that the inflation is going to go on, that the printing of paper money will be more and more accelerated, that prices will rise at a faster and faster rate. They want to exchange their money for anything they can get. Finally, however, holders of goods refuse to accept that money on any terms.

Thus every inflation must eventually be ended by government or it must “self-destruct”—but not until after it has done untold harm.

5 Economics of Inflation, p. 174.

6 Ibid.

7 Ibid., pp. 173–74.

8 For an elaboration of this analysis, see Benjamin M. Anderson, Jr., The Value of Money (New York: Richard R. Smith, 1917, 1936), ch. 13.

• 17*
• 2Karl Helfferich, Das Geld (sixth edition, Leipzig, 1923).
• 3Economics of Inflation, p. 197.
• 4Lecture, 1942; published in Encounter, 1975.
• 5* Estimated from data through June.
• 6The increase of the circulation has not preceded the rise of prices and the depreciation of the exchange, but it followed slowly and at great distance. The circulation increased from May 1921 to the end of January 1923 by 23 times; it is not possible that this increase had caused the rise in the prices of imported goods and of the dollar, which in that period increased by 344 times.
• 7There was a great decline in labor efficiency. Part of this was the result of malnutrition brought about by high food prices. Bresciani-Turroni tells us: “In the acutest phase of the inflation Germany offered the grotesque, and at the same time tragic, spectacle of a people which, rather than produce food, clothes, shoes, and milk for its own babies, was exhausting its energies in the manufacture of machines or the building of factories.”
• 8For instance, you might drop in at the tobacconist’s for a cigar. Alarmed by the price, you’d rush to a competitor, find that his price was still higher, and race back to the first shop, which may have doubled or tripled its price in the meantime. There was no help for it, you had to dig into your pocketbook and take out a huge bundle of millions, or even billions, depending on the date.