Murray N. Rothbard
Money in a Free Society
8. The "Proper" Supply of Money
Now we may ask: what is the supply of money in society and how is that supply used? In particular, we may raise the perennial question, how much money "do we need"? Must the money supply be regulated by some sort of "criterion," or can it be left alone to the free market?
First, the total stock, or supply, of money in society at any one time, is the total weight of the existing money-stuff. Let us assume, for the time being, that only one commodity is established on the free market as money. Let us further assume that gold is that commodity (although we could have taken silver, or even iron; it is up to the market, and not to us, to decide the best commodity to use as money). Since money is gold, the total supply of money is the total weight of gold existing in society. The shape of gold does not matter?except if the cost of changing shapes in certain ways is greater than in others (e.g., minting coins costing more than melting them). In that case, one of the shapes will be chosen by the market as the money-of-account, and the other shapes will have a premium or discount in accordance with their relative costs on the market.
Changes in the total gold stock will be governed by the same causes as changes in other goods. Increases will stem from greater production from mines; decreases from being used up in wear and tear, in industry, etc. Because the market will choose a durable commodity as money, and because money is not used up at the rate of other commodities--but is employed as a medium of exchange--the proportion of new annual production to its total stock will tend to be quite small. Changes in total gold stock, then, generally take place very slowly.
What "should" the supply of money be? All sorts of criteria have been put forward: that money should move in accordance with population, with the "volume of trade," with the "amounts of goods produced," so as to keep the "price level" constant, etc. Few indeed have suggested leaving the decision to the market. But money differs from other commodities in one essential fact. And grasping this difference furnishes a key to understanding monetary matters. When the supply of any other good increases, this increase confers a social benefit; it is a matter for general rejoicing. More consumer goods mean a higher standard of living for the public; more capital goods mean sustained and increased living standards in the future. The discovery of new, fertile land or natural resources also promises to add to living standards, present and future. But what about money? Does an addition to the money supply also benefit the public at large?
Consumer goods are used up by consumers; capital goods and natural resources are used up in the process of producing consumer goods. But money is not used up; its function is to act as a medium of exchanges--to enable goods and services to travel more expeditiously from one person to another. These exchanges 3%3 are all made in terms of money prices. Thus, if a television set exchanges for three gold ounces, we say that the "price" of the television set is three ounces. At any one time, all goods in the economy will exchange at certain gold--ratios or prices. As we have said, money, or gold, is the common denominator of all prices. But what of money itself? Does it have a "price"? Since a price is simply an exchange-ratio, it clearly does. But, in this case, the "price of money" is an array of the infinite number of exchange-ratios for all the various goods on the market.
Thus, suppose that a television set costs three gold ounces, an auto sixty ounces, a loaf of bread 1/100 of an ounce, and an hour of Mr. Jones' legal services one ounce. The "price of money" will then be an array of alternative exchanges. One ounce of gold will be "worth" either 1/3 of a television set, 1/60 of an auto, 100 loaves of bread, or one hour of Jones' legal service. And so on down the line. The price of money, then, is the "purchasing power" of the monetary unit--in this case, of the gold ounce. It tells what that ounce can purchase in exchange, just as the money-price of a television set tells how much money a television set can bring in exchange. What determines the price of money? The same forces that determine all prices on the market?that venerable but eternally true law: "supply and demand." We all know that if the supply of eggs increases, the price will tend to fall; if the buyers' demand for eggs increases, the price will tend to rise. The same is true for money. An increase in the supply of money will tend to lower its "price"; an increase in the demand for money will raise it. But what is the demand for money? In the case of eggs, we know what "demand" means; it is the amount of money consumers are willing to spend on eggs, plus eggs retained and not sold by suppliers. Similarly, in the case of money, "demand" means the various goods offered in exchange for money, plus the money retained in cash and not spent over a certain time period. In both cases, "supply" may refer to the total stock of the good on the market.
What happens, then, if the supply of gold increases, demand for money remaining the same? The "price of money" falls, i.e., the purchasing power of the money-unit will fall all along the line. An ounce of gold will now be worth less than 100 loaves of bread, 1/3 of a television set, etc. Conversely, if the supply of gold falls, the purchasing power of the gold-ounce rises.
What is the effect of a change in the money supply? Following the example of David Hume, one of the first economists, we may ask ourselves what would happen if, overnight, some good fairy slipped into pockets, purses, and bank vaults, and doubled our supply of money. In our example, she magically doubled our supply of gold. Would we be twice as rich? Obviously not. What makes us rich is an abundance of goods, and what limits that abundance is a scarcity of resources: namely land, labor and capital. Multiplying coin will not whisk these resources into being. We may feel twice as rich for the moment, but clearly all we are doing is diluting the money supply. As the public rushes out to spend its new-found wealth, prices will, very roughly, double--or at least rise until the demand is satisfied, and money no longer bids against itself for the existing goods.
Thus, we see that while an increase in the money supply, like an increase in the supply of any good, lowers its price, the change does not--unlike other goods--confer a social benefit. The public at large is not made richer. Whereas new consumer or capital goods add to standards of living, new money only raises prices--i.e., dilutes its own purchasing power. The reason for this puzzle is that money is only useful for its exchange value. Other goods have various "real" utilities, so than an increase in their supply satisfies more consumer wants. Money has only utility for prospective exchange; its utility lies in its exchange value, or "purchasing power." Our law--that an increase in money does not confer a social benefit--stems from its unique use as a medium of exchange.
An increase in the money supply, then, only dilutes the effectiveness of each gold ounce; on the other hand, a fall in the supply of money raises the power of each gold ounce to do its work. We come to the startling truth that it doesn't matter what the supply of money is. Any supply will do as well as any other supply. The free market will simply adjust by changing the purchasing power, or effectiveness of the gold-unit. There is no need to tamper with the market in order to alter the money supply that it determines.
At this point, the monetary planner might object: "All right, granting that it is pointless to increase the money supply, isn't gold mining a waste of resources? Shouldn't the government keep the money supply constant, and prohibit new mining?" This argument might be plausible to those who hold no principled objections to government meddling, thought it would not convince the determined advocate of liberty. But the objection overlooks an important point: that gold is not only money, but is also, inevitably, a commodity. An increased supply of gold may not confer any monetary benefit, but it does confer a non-monetary benefit--i.e., it does increase the supply of gold used in consumption (ornaments, dental work, and the like) and in production (industrial work). Gold mining, therefore, is not a social waste at all.
We conclude, therefore, that determining the supply of money, like all other goods, is best left to the free market. Aside from the general moral and economic advantages of freedom over coercion, no dictated quantity of money will do the work better, and the free market will set the production of gold in accordance with its relative ability to satisfy the needs of consumers, as compared with all other productive goods. 
 Gold mining is, of course, no more profitable than any other business; in the long-run, its rate of return will be equal to the net rate of return in any other industry.