The equilibrium real wage defines full employment of the labor force, and full employment of the labor force (with a given production function) defines the full-employment level of output. Classical theory found no obstacle to the attainment of these positions as long as the money wage was flexible - that is, as long as it would fall in the face of unemployment. The possibility that this level of output once produced would not find a market was dismissed; Say’s Law ruled out any deficiency of aggregate demand.

Say’s Law, named for the French economist Jean Baptiste Say (1767-1832), most simply states supply creates its own demand. More precisely, it states whatever the level of output, the income created in the course of producing the output will necessarily lead to an equal amount of spending and thus and amount of spending sufficient to purchase the goods and services produced.

342 Index

Thus, if output is below that which can be produced with a fully employed labor force, inadequate demand cannot stand in the way of an expansion of output. As long as there are idle resources that can be put to work, the very expansion of output resulting from the utilization of such resources will create a proportionate rise in income that will be used to purchase the expanded output. In this way, Say’s Law denied that involuntary unemployment could be caused by a deficiency of aggregate demand. ⁵

Say’s Law is a theory that so closely resembles a familiar national income identity that there is some danger of finding the theory true because the identity is necessarily true. From the national income identities, it is clear that for every dollar of product there is a dollar of income. In whichever way we choose to define product, whether in gross (GNP) or net (NNP) terms, the product itself may be measured by the sum of incomes generated in producing it. This identity between product and income holds definitionally at any level of product, full-employment or anywhere below. The identity also says that any increase in product will be matched by an equal increase in income. But Say’s Law says this plus something more that is not definitionally true: that any increase in income will be matched by an equal increase in spending. The heart of the Keynesian theory is found in the argument that this last step, which is required to validate Say’s Law, does not necessarily follow. The equal rise in spending may not appear.

What is the basis for this unvarying equality between income and spending presumed by Say’s Law? In simplest terms, it is the argument that anyone (for example, a cobbler) who produces more product (shoes) than he needs for his personal use does so only to exchange this excess for the products of others. In the case of barter-exchange, and Say’s Law was originally set forth for a barter economy, this is necessarily the case. To “supply” one good in barter is unavoidably to “demand” another. A long line of classical economists believed that the law was equally true in a money economy. Although one’s excess production in a money economy is exchanged in the market for money and not for other goods, it may still be argued that the purpose of production is not to secure money as such but to secure money with which to buy the products of others. Though the interposition of money converts the direct or barter exchange of “goods for goods” into the indirect exchange of “goods for money for goods,” the mere introduction of money was thought to make no difference. Money, it was thought, functions only as a ‘medium of exchange.” Nobody other than a miser wants money for itself rather than for what it will buy. As soon as each person receives money for the sale of the goods or services he has supplied, he spends that money to buy goods and services supplied by others. This does not mean that every person spends whatever money he receives in a matter of minutes after receiving it. At any given time every person holds some

5. For a concise summary of Say’s Law, see W. S. Vickrey, Metastatics and Macroeconomics, Harcourt, Brace & World, 1964, pp. 168-70. A fuller exposition may be found in a textbook that was popular during the 1920s: F. M. Taylor, Principles of Economics, 9th ed., Ronald, 1921, pp. 196-205. For a detailed analysis see J. A. Schumpeter, History of Economic Analysis, London: Oxford Univ. Press, 1954, pp. 615-25.

343

money because of the unevenness between his receipts and his expenditures. However, so long as people hold no more money than is required for this, money is being used only as a medium of exchange.

As soon as we recognize that people other than misers may at times have reasons to hold some money in idle cash balances - that is, to use money as a “store of value” and not only as a “medium of exchange” - a possible break appears in the chain between the receipt of aggregate income in the form of money and the spending of that money income for the aggregate output whose production generated it. Such a break in the chain invalidates Say’s Law when applied to an economy using money. However, in order to proceed with the construction of the classical system, let us for the moment accept Say’s Law as valid in a money-using economy. There is then no break between the receipt of money income and the spending thereof, and there is accordingly no such thing as a deficiency of aggregate demand.

THE QUANTITY THEORY OF MONEY

Classical theory relied on Say’s Law to assure that aggregate demand would always be equal to aggregate supply; any increase in output automatically created an equal increase in spending that removed the increase in output from the market. Classical theory relied primarily on a flexible money wage to assure that the actual level of output at any time would be that produced by a fully employed labor force. Full employment, we remember, calls for a real wage that equates the supply of with the demand for labor; for any given price level, a flexible wage will adjust as required to produce the required real wage. All this leaves the price level unspecified, and to cover this classical theory relied on the quantity theory of money.

Index

The Quantity Theory as a Theory of the Price Level

In the classical system money’s function is essentially limited to that of a medium of exchange: Money is a device designed simply to overcome the difficulties unavoidable in barter exchange. But, even with money’s role thus limited, a question remains: Does a bushel of wheat exchange for $2 and a ton of coal for $20, or does a bushel of wheat exchange for $5 and a ton of coal for $50? The answer given by classical theory is that the absolute level of prices, $2 or $5 for wheat and $20 or $50 for coal, depends on the quantity of money in the economy. This relationship in which the price level is made a function of the money supply is known as the quantity theory of money. Furthermore, the relationship between changes in the money supply and changes in the price level was held to be strictly proportional. This conclusion depended on several assumptions that may most simply be brought out by examining the identity, in which M is the supply of money, V is its velocity, or the number

344 Index

of times it turns over per time period in the purchase of final output O, and P is the price level of this output. ⁶

It should be emphasized that is nothing more than an identity and as such stands completely apart from the quantity theory or any other theory. It is analogous to the identity between product and income in the national income accounts, which stands completely apart from Say’s Law. says simply that the quantity of money multiplied by the number of times each unit of money on the average is spent for final output in any time period equals the quantity of final goods and services sold during that period multiplied by the price level of those goods and services. If O is the physical volume of goods and services represented by any period’s GNP, P is the price level of these goods, and V is the number of times the money supply is used to purchase goods whose value is P0, the familiar GNP identity (neglecting net exports), may be expanded to read . Each part of this expression set off by an identity sign is identical in value to every other part; each part of this expression is merely a different way of describing the same dollar amount.

The identity is converted into the quantity theory of money under the assumptions that O and V are constant or stable in the short run and that P is passive. With O and V constant, the assumption that P is passive means that P depends on changes in M rather than that changes in M depend on changes in P. Given these assumptions, any short-run increase (or decrease) in M must lead to a proportionate rise (or fall) in P .⁷Without these assumptions, however, it is just as inevitable that any increase (or decrease) in M will not lead to a proportionate rise (or fall) in P (barring the unlikely case in which changes in V and O are exactly offsetting).

The classical view that the level of output is stable in the short run is based on the argument that the normal level of output is that produced by a fully employed labor force working with a fixed stock of capital and given production techniques. In terms of Figure 17-1, the production function can shift upward with technological advances and growth in the capital stock, resulting in a rise in output with a given labor input, but these changes occur gradually over the long run. The labor supply curve can also shift to the right with a resultant increase in output, but this too is a change that occurs gradually with the growth in population over the long run. Short-run variations in output could appear as a result of departures from the normal position of a fully employed labor force, but such departures were regarded as infrequent and

6. M is a stock variable and O a flow variable. If O is defined for one quarter, M is the average stock of money in the economy during that quarter, V is the number of times that average stock of money is used to purchase final output during that quarter, and P is the average price level of output for that quarter.

7. For example, begin with 100 x 4 = 2 x 200. Increase M by 10 percent to 110. On the assumptions of the constancy of V and O, we have 110 x 4 = 2.2 x 200, or a 10 percent increase in P, a rise proportionate with the increase in M. Or, alternatively, rewrite the identity as in which the assumed constancy of V and O makes V/O a positive proportionality constant, here equal to 4/200, or 0.02. Then, whatever the value for M, P is always 0.02 times that value.

345

subject to prompt correction in a system of competitive markets. Thus, given the assumption that full employment of the labor force is normal, the assumption of a stable level of output for any short-run period follows logically.

The classical view that the velocity or turnover rate of money is constant is based on the argument that the institutional, structural, and customary conditions that determine velocity usually change very gradually. Among these conditions are the frequency with which economic units receive and make payments, the regularity of these receipts and payments, and the portion of such receipts and payments that are on a money or barter basis. ⁸Though these and all other conditions affecting the size of V are subject to change, quantity theory asserts the gradualness of such change in support of its conclusion that V is constant in the short run. ⁹

The Quantity Theory as a Theory of Aggregate Demand

The quantity theory carries with it a crude theory of aggregate demand. If V is constant, the supply of money determines the total amount of spending for final output in any specified time period. Whatever the money supply may be, classical theory held that it was all in active circulation under normal conditions. There was no money in idle cash holdings. Here we encounter the same assumption that underlies Say’s Law in its application to a money-using economy: People use money as nothing more than a medium of exchange. An absence of idle money is a necessary part of Say’s Law when applied to a money-using economy, and it is an equally necessary part of the rigid quantity theory. Classical theory could see no reason why rational people would choose to hold any portion of their money receipts in idle cash form. As people came into the possession of money, there was in this view only one disposition for it: spending. Spending was either for consumption or capital goods. As we will see

8. Eyen though no one may choose to hold idle cash, everyone holds some cash to even out the difference between receipts and payments. For example, if an employee is paid $200 every other Friday, he does not typically spend the whole $200 on payday (and end up “broke” for the next thirteen days). If he spends the $200 evenly over the two-week period, his average cash balance will turn out to be $100. Since his spending biweekly is $200, this $100 average balance has a V of 2 biweekly; since his spending annually is $5,200, this $100 average balance has a V of 52 annually. If he were instead paid $100 every Friday, by the same line of argument his average cash balance would be $50. This $50 average balance has a V of 4 biweekly and a V of 104 annually. Generalizing for the economy as a whole, with a given supply of money, a change in which everyone were paid half as much twice as often would mean the existing supply of money could handle a much greater volume of final purchases. This rise in V with constant M would mean a rise in PO proportionate with the rise in V.

This illustration covers frequency of receipts, only one of the many conditions that were believed to change very slowly and thus to make for stability in V. For a discussion of the determinants of V, see L. V. Chandler, The Economics of Money and Banking, 4th ed., Harper & Row, 1964, pp. 315-21.

9. Even in its crude form, the quantity theory did not argue that short-run V and O were as stable or P as passive as is here assumed. We will retain these assumptions in the extreme form in which they have been given so that we may proceed with the construction of the simplified classical model.

346

Index

below, the act of saving, or not spending for consumption goods, was automatically transformed into an act of spending for capital goods. Money that was held back from consumption spending would be loaned to firms that would in turn spend the money, dollar for dollar, for capital goods. Thus, although persons do save, classical theory held that they would not hold any amounts saved in the form of cash and that therefore no money would escape from active circulation.

As long as money was used exclusively as a medium of exchange and thus remained completely in active circulation, any change in the supply of money would lead to a change in spending that was equal to the change in M times the constant multiple V. This brings us to Figure 17-2C. ¹⁰ Price level is measured along the horizontal axis and output along the vertical axis. The greater M is, the greater is MV; the greater MV is, the greater PO must be by the nature of the identity . lf V is constant, changes in aggregate demand result only from changes in M, and any change in aggregate demand may be measured by the equal change in the product of P x 0 that accompanies the change in aggregate demand. If O₁ represents the full-employment output, meaning that O₁ is constant in the short run, the rise in P from P₁to P₂must be proportional to the increase in M, represented by the shift of the curve from M₁V to M₂ V.

10. Parts A and B merely repeat Figure 17-1; Part D will be discussed shortly.

347

Any increase in M shifts the curve to the right and, with O constant, raises P proportionately; any decrease in M shifts the curve to the left and, again with 0 constant, reduces P proportionately.

Index

A&E Homepage

next page