Books / Digital Text

V. The Demand for Money

Let’s analyze the various elements that constitute the public’s demand for money. We have already seen that the demand curve for money will be falling in relation to the purchasing power of money; what we want to look at now is the cause of upward or downward shifts in that demand curve.


Before money can be held in one’s cash balance, it must be obtained in exchange. That is, we must sell goods and services we produce in order to “buy” money. However, if the supply of goods and services increases in the economy (i.e., supply curves shift to the right), the demand for money in exchange will also increase. An increased supply of goods produced will raise the demand for money and also therefore lower the overall level of prices. As we can see in Figure 3.6, as the demand for money rises, a shortage of cash balances develops at the old equilibrium price level, and prices fall until a new equilibrium, PPM, is achieved.

Historically, the supply of goods and services has usually increased every year. To the extent it does so, this increase in the demand for money will tend to lower prices over a period of time. Indeed, so powerful has this force been for lowering prices, that they fell from the mid-eighteenth century until 1940, with the only exception being during periods of major wars: the Napoleonic Wars, War of 1812, the Civil War, and World War I. Paper money was increasing the money supply during this era, but increases in M were more than offset by the enormous increases in the supply of goods produced during the Industrial Revolution in an unprecedented period of economic growth. Only during wartime, when the governments ran the printing presses at full blast to pay for the war effort, did the money supply overcome the effects of increasing production and cause price levels to zoom upward.1


The demand for money is also affected by the frequency with which people are paid their wages or salaries. Suppose, for example, that Mr. Jones and Mr. Smith are each paid an income of $12,000 a year, or $1,000 per month. But there is a difference: Jones is paid every week, and Smith every month. Does this make any difference to their economic situation? Let us first take Smith, and find out what his cash balance is on each day. Let us assume, to keep things simple, that each man is paid on the first day of the wage period, and then spends money at an even rate until the last day, when his money is exhausted (and we assume that each man’s income equals his expenditures for the relevant time periods).

Smith receives $1,000 on the first of the month, and then draws down his $1,000 cash balance at an even rate until the end of the month by a bit more than $33 a day.


What is Smith’s average cash balance for the month? We can find out by simply adding $1,000 on Day 1, and 0 on Day 30, and dividing by 2: the answer is $500.

Let us now compare Smith to Jones, who has the same total income, but receives his paycheck once a week. Figuring four weeks to the month to simplify matters, this means that Jones gets a check of $250 at the beginning of each week and then draws it down steadily until he reaches a cash balance of zero at the end of the week. His monetary picture will be as follows:


Jones gets a check for $250 at the beginning of each week, and then draws down his cash balance each day by approximately $35 until he reaches zero at the end of the week. His income is the same as Smith’s; but what is his average cash balance? Again, we can arrive at this figure by adding $250, at the beginning of each week, and 0 and dividing by 2: the result is $125.

In short, even though their incomes are identical, Smith, who gets paid less frequently, has to keep an average cash balance four times that of Jones. Jones is paid four times as frequently as Smith, and hence has to keep a cash balance of only 1/4 the amount.

Cash balances, therefore, do not only do work in relation to the level of prices. They also perform work in relation to the frequency of income. The less frequent the payment, the higher the average cash balance, and therefore the greater the demand for money, the greater the amount, at any price level, that a person will seek to keep in his cash balance. The same cash balances can do more money work the greater the frequency of payment.

In my salad days, I experienced the problem of frequency of payment firsthand. I was working on a foundation grant. My income was fairly high, but I was getting paid only twice a year. The result was that the benefits of my respectable income were partially offset by the necessity of keeping an enormous cash balance in the bank, just to finance my daily expenditures. In many painful ways, I was far worse off than I would have been with the same income with more frequent checks coming in.

What effect might this have on the price level? If the general frequency of payment changes in a society, this will shift the demand for money and raise or lower the price level. Thus, if people suddenly stop being paid once a month, and instead get paid twice a month, this will lower everyone’s demand for money. They will keep a lower cash balance for their existing income and price level, and so the demand for money will shift downward, as in Figure 3.7. People will try to get rid of their surplus cash balances and, as they do so by spending money on goods and services, prices will be driven upward until a new higher equilibrium price level will clear the market and equilibrate the existing supply to the decreased demand.

On the other hand, if frequency of payment of salaries should shift generally from once a week to twice a month, the reverse will happen. People will now need to carry a higher average cash balance for their given incomes. In their scramble for higher cash balances, their demand for money rises, as in Figure 3.6 above. They can only raise their cash balances by cutting back their spending, which in turn will lower prices of goods and thereby relieve the “shortage” of cash balances.

Realistically, however, frequency of payment does not change very often, if at all. Any marked change, furthermore, will only be one-shot, and certainly will not be continuous. Frequency of payment is not going to go up or down every year. Changes in frequency, therefore, could scarcely account for our contemporary problems of chronic inflation. If anything, the general shift from blue-collar to white-collar jobs in recent decades has probably reduced the frequency of payment a bit, and therefore had a slight price-lowering effect. But we can safely ignore this factor if we are looking for important causal factors.


On the other hand, there is another causal factor which can only lower the demand for money over time: new methods of economizing the need for cash balances. These are technological innovations like any other, and will result in a lower demand for money for each successful innovation.

An example is the development of more efficient “clearing systems,” that is, institutions for the clearing of debt. My eighth-grade teacher, perhaps unwittingly, once illustrated the effect of clearing systems on reducing average cash balances. In effect he said to the class: Suppose that each of you owes $10 that will be due on the first day of next month. If there are, say, 30 kids in each class, each will need to come up with $10 to pay their debt, so that a total cash balance of $300 will be demanded by the class in order to pay their various debts.

But now, my teacher pointed out, suppose that each of you still owes $10 due on the first day of next month. But each of you owes $10 to the boy or girl on your left. More precisely: The teacher owes $10 to the first kid in the front of the class, then each kid in turn owes $10 to the kid on his left, until finally the last person at the end of the line, in turn, owes $10 to the teacher. Each of these debts is due on the first of the month. But in that case, each of us can wipe out his or her debt all at once, at a single blow, without using any cash balance at all. Presto chango! The class’s demand for cash balance for repaying debt has been reduced as if by magic, from $300 to zero. If there were an institutional mechanism for finding and clearing these debts, we could dramatically and drastically reduce our need for accumulating and keeping cash balances, at least for the payment of debt.

Any devices for economizing cash balances will do as well as clearing systems in reducing the public’s demand for money. Credit cards are an excellent current example. Contrary to some views, credit cards are not in themselves money and therefore do not add to the money supply. Suppose, for example, that I eat dinner in a restaurant, run up a $20 bill, and pay by American Express card rather than by cash. The American Express card is not money. One way to see that is to note whether using the card constitutes final payment for the dinner. One crucial feature of money is that using it constitutes final payment; there is no need for any more. If I pay for the dinner with a $20 bill, for example, that’s it; my debt has been canceled finally and completely. Hence the $20 was truly money. But handing the restaurant my American Express card hardly completes the matter; on the contrary, I then have to pay American Express $20, plus interest at some later date.

In fact, when a credit card is used, two credit transactions are taking place at once. In the above example, American Express lends me the money by paying the restaurant on my behalf; at the same time, I pledge to pay American Express $20 plus interest. In other words, American Express picks up my tab and then I owe it money.

Credit cards, then, are not part of the money supply. But carrying them enables me to walk around with a far lower cash balance, for they provide me with the ability to borrow instantly from the credit card companies. Credit cards permit me to economize on cash.

The development of credit cards, clearing systems, and other devices to economize cash, will therefore cause the demand for money to be reduced, and prices to increase. Again, however, these effects are one-shot, as the new device is invented and spreads throughout the economy, and its impact is probably not very important quantitatively. These new devices cannot begin to account for the chronic, let alone the accelerating, inflation that plagues the modern world.


An intangible, but highly important determinant of the demand for money, is the basic confidence that the public or market has in the money itself. Thus, an attempt by the Mongols to introduce paper money in Persia in the twelfth and thirteenth centuries flopped, because no one would accept it. The public had no confidence in the paper money, despite the awesomely coercive decrees that always marked Mongol rule. Hence, the public’s demand for the money was zero. It takes many years—in China it took two to three centuries—for the public to gain enough confidence in the money, so that its demand for the money will rise from near zero to a degree great enough to circulate throughout the kingdom.

Public confidence in the country’s money can be lost as well as gained. Thus, suppose that a money is King Henry’s paper, and King Henry has entered a war with another state which he seems about to lose. King Henry’s money is going to drop in public esteem and its demand can suddenly collapse.

It should be clear then, that the demand for paper money, in contrast to gold, is potentially highly volatile. Gold and silver are always in demand, regardless of clime, century, or government in power. But public confidence in, and hence demand for, paper money depends on the ultimate confidence—or lack thereof—of the public in the viability of the issuing government. Admittedly, however, this influence on the demand for money will only take effect in moments of severe crisis for the ruling regime. In the usual course of events, the public’s demand for the government’s money will likely be sustained.


We have dealt so far with influences on the demand for money that have been either one-shot (frequency of payment and clearing systems), remote (confidence in the money), or gradual (supply of good and services). We come now to the most important single influence on the demand for money: This is the public’s expectation of what will happen to prices in the near, or foreseeable, future. Public expectation of future price levels is far and away the most important determinant of the demand for money.

But expectations do not arise out of thin air; generally, they are related to the immediate past record of the economy. If prices, for example, have been more or less stable for decades, it is very likely that the public will expect prices to continue on a similar path. There is no absolute necessity for this, of course; if conditions are changing swiftly, or are expected to change quickly, then people will take the changes into account.

If prices are generally expected to remain the same, then the demand for money, at least from the point of view of expectations, will remain constant, and the demand for money curve will remain in place. But suppose that, as was the case during the relatively free-market and hard-money nineteenth century, prices fell gradually from year to year. In that case, when people knew in their hearts that prices would be, say, 3 percent lower next year, the tendency would be to hold on to their money and to postpone purchase of the house or washing machine, or whatever, until next year, when prices would be lower. Because of these deflationary expectations, then, the demand for money will rise, since people will hold on to more of their money at any given price level, as they are expecting prices to fall shortly. This rise in the demand for money (shown in Figure 3.6) would cause prices to fall immediately. In a sense, the market, by expecting a fall in prices, discounts that fall, and makes it happen right away instead of later. Expectations speed up future price reactions.

On the other hand, suppose that people anticipate a large increase in the money supply and hence a large future increase in prices. Their deflationary expectations have now been replaced by inflationary expectations. People now know in their hearts that prices will rise substantially in the near future. As a result, they decide to buy now—to buy the car, the house, or the washing machine—instead of waiting for a year or two when they know full well that prices will be higher. In response to inflationary expectations, then, people will draw down their cash balances, and their demand for money curve will shift downward (shown in Figure 3.7). But as people act on their expectations of rising prices, their lowered demand for cash pushes up the prices now rather than later. The more people anticipate future price increases, the faster will those increases occur.

Deflationary price expectations, then, will lower prices, and inflationary expectations will raise them. It should also be clear that the greater the spread and the intensity of these expectations, the bigger the shift in the public’s demand for money, and the greater the effect in changing prices.

While important, however, the expectations component of the demand for money is speculative and reactive rather than an independent force. Generally, the public does not change its expectations suddenly or arbitrarily; they are usually based on the record of the immediate past. Generally, too, expectations are sluggish in revising themselves to adapt to new conditions; expectations, in short, tend to be conservative and dependent on the record of the recent past. The independent force is changes in the money supply; the demand for money reacts sluggishly and reactively to the money supply factor, which in turn is largely determined by government, that is, by forces and institutions outside the market economy.

During the 1920s, Ludwig von Mises outlined a typical inflation process from his analysis of the catastrophic hyperinflation in Germany in 1923—the first runaway inflation in a modern, industrialized country. The German inflation had begun during World War I, when the Germans, like most of the warring nations, inflated their money supply to pay for the war effort, and found themselves forced to go off the gold standard and to make their paper currency irredeemable. The money supply in the warring countries would double or triple. But in what Mises saw to be Phase I of a typical inflation, prices did not rise nearly proportionately to the money supply. If M in a country triples, why would prices go up by much less? Because of the psychology of the average German, who thought to himself as follows: “I know that prices are much higher now than they were in the good old days before 1914. But that’s because of wartime, and because all goods are scarce due to diversion of resources to the war effort. When the war is over, things will get back to normal, and prices will fall back to 1914 levels.” In other words, the German public originally had strong deflationary expectations. Much of the new money was therefore added to cash balances and the Germans’ demand for money rose. In short, while M increased a great deal, the demand for money also rose and thereby offset some of the inflationary impact on prices. This process can be seen in Figure 5.3.

In Phase I of inflation, the government pumps a great deal of new money into the system, so that M increases sharply to M’. Ordinarily, prices would have risen greatly (or PPM fallen sharply) from 0A to 0C. But deflationary expectations by the public have intervened and have increased the demand for money from D to D’, so that prices will rise and PPM falls much less substantially, from 0A to 0B.

Unfortunately, the relatively small price rise often acts as heady wine to government. Suddenly, the government officials see a new Santa Claus, a cornucopia, a magic elixir. They can increase the money supply to a fare-thee-well, finance their deficits and subsidize favored political groups with cheap credit, and prices will rise only by a little bit!


It is human nature that when you see something work well, you do more of it. If, in its ceaseless quest for revenue, government sees a seemingly harmless method of raising funds without causing much inflation, it will grab on to it. It will continue to pump new money into the system, and, given a high or increasing demand for money, prices, at first, might rise by only a little.

But let the process continue for a length of time, and the public’s response will gradually, but inevitably, change. In Germany, after the war was over, prices still kept rising; and then the postwar years went by, and inflation continued in force. Slowly, but surely, the public began to realize: “We have been waiting for a return to the good old days and a fall of prices back to 1914. But prices have been steadily increasing. So it looks as if there will be no return to the good old days. Prices will not fall; in fact, they will probably keep going up.” As this psychology takes hold, the public’s thinking in Phase I changes into that of Phase II: “Prices will keep going up, instead of going down. Therefore, I know in my heart that prices will be higher next year.” The public’s deflationary expectations have been superseded by inflationary ones. Rather than hold on to its money to wait for price declines, the public will spend its money faster, will draw down cash balances to make purchases ahead of price increases. In Phase II of inflation, instead of a rising demand for money moderating price increases, a falling demand for money will intensify the inflation (Figure 5.4).


Here, in Phase II of the inflation, the money supply increases again, from M’ to M’’. But now the psychology of the public changes, from deflationary to inflationary expectations. And so, instead of prices rising (PPM falling) from 0B to 0D, the falling demand for money, from D’ to D’’, raises prices from 0D to 0E. Expectations, having caught up with the inflationary reality, now accelerate the inflation instead of moderating it.

Both these phases of a typical inflation can be combined as shown in Figure 5.5.


There is no scientific way to predict at what point in any inflation expectations will reverse from deflationary to inflationary. The answer will differ from one country to another, and from one epoch to another, and will depend on many subtle cultural factors, such as trust in government, speed of communication, and many others. In Germany, this transition took four wartime years and one or two postwar years. In the United States, after World War II, it took about two decades for the message to slowly seep in that inflation was going to be a permanent fact of the American way of life.

When expectations tip decisively over from deflationary, or steady, to inflationary, the economy enters a danger zone. The crucial question is how the government and its monetary authorities are going to react to the new situation. When prices are going up faster than the money supply, the people begin to experience a severe shortage of money, for they now face a shortage of cash balances relative to the much higher price levels. Total cash balances are no longer sufficient to carry transactions at the higher price. The people will then clamor for the government to issue more money to catch up to the higher price. If the government tightens its own belt and stops printing (or otherwise creating) new money, then inflationary expectations will eventually be reversed, and prices will fall once more—thus relieving the money shortage by lowering prices. But if government follows its own inherent inclination to counterfeit and appeases the clamor by printing more money so as to allow the public’s cash balances to “catch up” to prices, then the country is off to the races. Money and prices will follow each other upward in an ever-accelerating spiral, until finally prices “run away,” doing something like tripling every hour. Chaos ensues, for now the psychology of the public is not merely inflationary, but hyperinflationary, and Phase III’s runaway psychology is as follows: “The value of money is disappearing even as I sit here and contemplate it. I must get rid of money right away, and buy anything, it matters not what, so long as it isn’t money.” A frantic rush ensues to get rid of money at all costs and to buy anything else. In Germany, this was called a “flight into real values.” The demand for money falls precipitously almost to zero, and prices skyrocket upward virtually to infinity. The money collapses in a wild “crack-up boom.” In the German hyperinflation of 1923, workers were paid twice a day, and the housewife would stand at the factory gate and rush with wheelbarrows full of million mark notes to buy anything at all for money. Production fell, as people became more interested in speculating than in real production or in working for wages. Germans began to use foreign currencies or to barter in commodities. The once-proud mark collapsed.

The absurd and disastrous way in which the Reichsbank—the German Central Bank—met the crucial clamor for more money to spend immediately in the hyperinflation of the early 1920s is revealed in a notorious speech delivered by Rudolf Havenstein, the head of the Reichsbank, in August 1923. The Reichsbank was the sole source of paper money, and Havenstein made clear that the bank would meet its responsibilities by fulfilling the increased demand for paper money. Denominations of the notes would be multiplied, and the Reichsbank would stand ready to keep its printing presses open all night to fill the demand. As Havenstein put it:

The wholly extraordinary depreciation of the mark has naturally created a rapidly increasing demand for additional currency, which the Reichsbank has not always been able fully to satisfy. A simplified production of notes of large denominations enabled us to bring ever greater amounts into circulation. But these enormous sums are barely adequate to cover the vastly increased demand for the means of payment, which has just recently attained an absolutely fantastic level, especially as a result of the extraordinary increases in wages and salaries.

The running of the Reichsbank’s note-printing organization, which has become absolutely enormous, is making the most extreme demands on our personnel.2

During the later months of 1923, the German mark suffered from an accelerating spiral of hyperinflation: the German government (Reichsbank) poured out ever-greater quantifies of paper money which the public got rid of as fast as possible. In July 1914, the German mark had been worth approximately 25 cents. By November 1923, the mark had depreciated so terrifyingly that it took 4.2 trillion marks to purchase one dollar (in contrast to 25.3 billion marks to the dollar only the month before).

And yet, despite the chaos and devastation, which wiped out the middle class, pensioners and fixed-income groups, and the emergence of a form of barter (often employing foreign currency as money), the mark continued to be used. How did Germany get out of its runaway inflation? only when the government resolved to stop monetary inflation, and to take steps dramatic enough to convince the inflation-wracked German public that it was serious about it. The German government brought an end to the crack-up boom by the “miracle of the Rentenmark” The mark was scrapped, or rather, a new currency, the Rentenmark, was issued, valued at 1 trillion old marks, which were convertible into the new currency. The government pledged that the quantity of Rentenmarks issued would be strictly limited to a fixed amount (a pledge that was kept for some time), and the Reichsbank was prohibited from printing any further notes to finance the formerly enormous government deficit. Once these stern measures had been put into effect, the hyperinflation was brought to an end. The German economy rapidly recovered. Yet, it must be pointed out that the German economy did not escape a posthyperinflation recession, called a “stabilization crisis,” in which the swollen and unsound investments of the inflationary period were rapidly liquidated. No one complained bitterly; the lessons of the monstrous inflation were burned into everyone’s heart.3

Only a clear and dramatic cessation of the spiraling expansion of the money supply can turn off the money tap and thereby reverse the accelerating inflationary expectations of the public. Only such a dramatic end to monetary inflation can induce the public to start holding cash balances once again.

Thus we see that price levels are determined by the supply and the demand for money, and that expansion of the money sup-ply—a function solely of government—is the prime active force in inflation.

  • 1. The only exception was the period 1896-1914, when new gold discoveries caused moderate increases in the price level.
  • 2. In Fritz K. Ringer, ed., The German Inflation of 1923 (New York: Oxford University Press, 1969), p. 96.
  • 3. For a good overview of the German economy, see Gustav Stolper, The German Economy, 1870 to the Present (New York: Harcourt, Brace & World, 1967); for an excellent history and analysis of the German hyperinflation, see Costantino Bresciani-Turroni, The Economics of Inflation (London: George Allen & Unwin, 1937).