Chapter 3 SAY’S LAW

The question of which monetary regime is optimal has several facets. Ethically, the question may be about which regime is consistent with a “rule of law,” procedural justice, and individual rights. Socio-politically, one may ask which regime minimizes rent-seeking behavior on the part of special interest groups. In a technical economic sense, one may inquire as to which institutional structure minimizes the destructive effects of business cycles. This chapter will examine the relationships between free banking and cyclical stability.

BUSINESS CYCLES

Just what are business cycles? At the most basic level, cycles are evidence of market discoordination.¹ In the expansionary phase (the “boom”), there exists, in the aggregate, an excess demand for goods. That is, resource owners, for example, laborers, demand goods whose market value exceeds the value of the resources the individuals supply. During the contractionary phase (the “recession”), there is, in the aggregate, an excess supply of goods. Resource owners supply resources whose value exceeds the market value of the goods demanded. The value of resources supplied may be taken to mean those goods one could potentially purchase with the income generated by the sale or rental of one’s resources. This Steven Horwitz calls “notional demand”; he terms the value of goods actually demanded “effective demand” (1990, 8). In a barter economy, there can exist no difference between effective demand and notional demand. However, when there is one good that is more marketable than any other and, therefore, is more or less universally accepted as an exchange medium (money), then a rift can develop between effective and notional demands. Notional demand is translated into effective demand via money. If the money supply is deficient, notional demand exceeds effective demand. If the money supply is excessive, effective demand exceeds notional demand.

What, then, is necessary in order for effective demand to equal notional demand? In other words, what is necessary in order to avoid business cycles? The necessary—but not sufficient—condition is monetary equilibrium. Real factors may still upset individual markets, but as long as monetary equilibrium prevails, the effects of these real shocks will be minimized. “What monetary equilibrium does is to allow for the most coordination possible at the microeconomic level by ensuring that the supply of inside money is correct” (Horwitz 1990, 9).

Two questions could be raised at this juncture. First, how should one define monetary equilibrium? George Selgin (1988a), Horwitz (1990), Leland Yeager (1986), and Yeager and Robert Greenfield (1989), for instance, all suggest that monetary equilibrium exists when the supply of money equals the demand for money at the prevailing level of prices. For Selgin, at least, this is rather problematic. To use such a definition seems contrary to his analysis of the effects of a pervasive productivity gain.² Such a gain in productivity decreased goods’ production costs and, in turn, the price level. Money supply and money demand were not equated at the prevailing level of prices, but at a new, lower level.³

Perhaps a better definition would be that monetary equilibrium occurs when the supply of money equals the demand for money, given the underlying state of general productivity and the concomitant price level. Improvements in productivity would warrant a decline in the price level, and retrogressions in productivity would require a rise in the price level. If, on the other hand, either (1) the income velocity of money or (2) the composition of money demand were to change (with no change in productivity), then no alteration in the price level would be necessary. Aggregate nominal money balances would remain constant in the face of either a change in the composition of money demand or a change in productivity. Money balances would vary inversely with changes in money’s income velocity.

Second, most economists perceive issues regarding money supply and money demand as fundamentally and ineluctably macroeconomic in nature. Yet, as cited above, Horwitz refers to these as problems of microeconomic coordination. There is a profound clash of perspectives at work here. Those who perceive an economy as blocks of aggregates and the task of the economist as the manipulation of those aggregates will think of money in a macrocontext. Those who perceive an economy as a multitude of interacting individuals possessed of imperfect knowledge and the task of the economist as the understanding of that interactive discovery process, will think of money in a microcontext. The perspective adopted here is that money is a microissue that has profound ma-croimplications.⁴ To pursue this point, one should consider Horwitz’s comments that

it is relative prices that guide actors in their allocative decisions in the market. What prices do is allow actors to coordinate their actions . . . prices reflect and convey knowledge . . . competition allows this knowledge to be passed in the market through acts of buying and selling. . . . What economies “do” is coordinate actions through prices. With this view, it is easy to see why macroeconomic problems are really microeconomic problems. (1988, 32)

Macroeconomic discoordination is the result of microeconomic discoordination. Many factors can bring this about: price controls, trade barriers, licensing requirements, antitrust statutes, taxes, subsidies, and so forth. However, the one occurrence that most frequently causes the most severe microeconomic (and macroeconomic) disruptions is monetary disequilibrium. This brings one back to an earlier point. To sustain monetary equilibrium is to minimize market discoordination, minimize the departures of effective demand from notional demand, and minimize both the frequency and severity of business cycles.

This, however, must be a concept of monetary equilibrium that takes into account changes in productivity. Otherwise, such “equilibrium” can actually be disruptive. For instance, if productivity rises, then a price-stabilization approach to monetary equilibrium would require an increase in the money supply to prevent a decline in the price level. This would pose at least two problems for the economy. First, since unit production costs would be falling, this would manifest itself as an (apparent) increase in firms’ profits. Those extra profits will be eliminated, however, as marginal factor costs are bid up because of the rise in marginal productivity. Firms would face a “signal extraction” problem. The “hidden inflation” caused by the policy of maintaining prices at their former level would lead producers first to overexpand and then sharply contract production. Second, a productivity norm (which would be characteristic of free banking) requires fewer relative price changes than does price level stabilization. Selgin constructs a simple example of three factors of production and 1,000 final goods (1990, 275–76). Given certain plausible conditions, the productivity norm would require only one adjustment: a decline in the price of that good regarding which productivity had increased. In contrast, to keep the price level stable under the same circumstances would necessitate changes in the prices of 999 final goods as well as the three factors of production. Moreover, to maintain monetary equilibrium in the context of productivity changes will tend to keep effective demand equal to notional demand. It is this approach—and this only—which is fully consistent with Say’s Law.

CLASSICAL ECONOMICS

Surely the single unifying theme of classical economics was that set of propositions subsumed under the title of “Say’s Law.” Named after the French economist Jean-Baptiste Say, who did much to popularize the work of Adam Smith in Europe, this principle has twice been the focal point of intense, even acrimonious, debates in economics.⁵ First in the 1820s, early proponents of free markets, such as Say, James Mill, and David Ricardo, defended the validity of Say’s Law against critics, such as Thomas Malthus, Jean Simonde de Sismondi, and James Maitland (Earl of Lauderdale).⁶ A century later, in the 1930s, the critics included John Maynard Keynes, John Hicks, Alvin Hansen, and Joan Robinson. The defenders of Say’s Law at that time included Ludwig von Mises, F. A. Hayek, Lionel Robbins, D. H. Robertson, and William H. Hutt.

Thomas Sowell offers a lucid summary of this principle:

The basic idea behind Say’s Law is both simple and important. The production of goods (including services) causes incomes to be paid to suppliers of the factors (labor, capital, land, etc.) used in producing the goods. The total price of the goods is the sum of these payments for wages, profits, rent, etc.—which is to say that the income generated during the production of a given output is equal to the value of that output. An increased supply of output means an increase in the income necessary to create a demand for that output. Supply creates its own demand. (1972, 4)

Of course, here the terms supply and demand do not refer to the supply of or demand for a specific good, but to aggregate supply and aggregate demand. That is, Say’s Law suggests that the supply of goods in general creates demand for goods in general, or to revert to terminology used earlier, notional demand creates its own effective demand. There has been little argument over the proposition that notional demand can potentially create its own effective demand (Sowell 1972, 36). The question has long been: Will notional demand actually create its own effective demand?⁷ If so, under what conditions will this be true?

The contrast between actual and potential results is manifest in the various forms (or subpropositions) of Say’s Law. Sowell identifies these as Walras’ Law, Say’s Identity, and Say’s Equality (1972, 34–35). Say’s Equality states that, if there is equilibrium in all markets (goods markets, factor markets, and the market for money), then the nominal value (money prices times quantities) of goods supplied will equal the value of goods demanded. This reveals the classical concern with disproportionality in specific markets, for example, the labor market (Hutt 1979, 135–77). Say’s Equality establishes that notional demand will equal effective demand if a matrix of equilibria prevails. Say’s Identity adopts a long-run perspective in which there is neither excess supply of nor excess demand for money. This proposition asserts, therefore, that the nominal value of goods supplied is identical to the nominal value of goods demanded. This, then, is merely a general long-run tendency. Walras’ Law states that the sum of the value of goods supplied plus money supplied equals the sum of the value of goods demanded plus money demanded. “It implies that an excess quantity of goods supplied is the same as an excess demand for money” (Sowell 1972, 34).

Walras’ Law cuts to the heart of the matter. If monetary equilibrium (as denned earlier) holds, then there can be no monetary disturbances that might fuel a business cycle. The only possible disruptive influence will be real shocks that cause temporary disequilibria in specific markets. If monetary equilibrium is maintained more or less continuously, then such real shocks will have neither pervasive nor lasting effects. Effective demand will tend to equal notional demand (micro- and macroeconomic coordination will be maximized) as long as the market for money is in equilibrium. Therefore, properly understood, Say’s Law is not (and never was) an unconditional proposition, but a conditional one. Given monetary equilibrium, the expected value of the difference between effective demand and notional demand equals zero.

The microeconomic aspects of Say’s Law seem to be largely forgotten or ignored these days, but that was not always the case. Benjamin Anderson, writing in 1949, commented on Say’s Law in the following manner:

The doctrine that supply creates its own demand . . . assumes a proper equilibrium among the different kinds of production, assumes proper terms of exchange (that is, price relationships) among different kinds of products, assumes proper relations between prices and costs. . . . Moreover, the money and capital markets must be in a state of balance. When there is an excess of bank credit used as a substitute for savings . . . or when the total volume of money and credit is expanded far beyond the growth of production and trade, disequilibria arise. (1979, 385–86)

Anderson touches on several key points. He stresses the microfoundations upon which Say’s Law was constructed.⁸ He notes that there must be a “proper relation” between prices and costs. The reader will recall that the productivity norm approach to monetary equilibrium (as under free banking) will maintain such a proper relation, whereas a stable price level rule will not necessarily do so. Finally, he emphasizes the importance of “balance” (equilibrium) in the money and credit markets, and expresses particular concern for the case in which credit “substitutes” for savings.

TIME AND MONEY

Horwitz suggests that sound macroeconomic analysis must focus on the coordination process that occurs in specific markets (1990). In other words, the microfoundations must precede the macroconclusions. It is difficult to challenge such an assertion with any seriousness. Expanding on Roger Garrison’s reference to time and money as “the universals of macroeconomics,”⁹ Horwitz further argues that “such analysis will have to consider the two most pervasive economic goods—time and money. A coordination-oriented macroeconomics would focus on movements in the markets for time . . . and money” (1990, 1). In more conventional terms, of what do the market for money and the market for time consist? The money market involves the supply of and demand for money, the purchasing power of money (its price, the inverse of the price level), and banking. The time market involves savings and investment, the natural rate of interest, the supply of and demand for credit, market rates of interest, and banking.¹⁰

What is unique and important about free banking on a convertible specie standard is the relationship between the money and time markets.¹¹ As both Selgin (1988a, 54–55) and Horwitz (1990, 4–6) argue, the demand for inside money represents voluntary (not forced) savings on the part of consumers. Consumers let their cash balances rise by choosing to refrain from acts of consumption. This is unavoidable as long as consumers must give up real goods in order to acquire money. In particular, by not redeeming the deposit accounts or banknotes they hold, consumers leave more specie reserves in the hands of banks, “which is equivalent to a very short term act of savings” (Horwitz 1990, 4). Furthermore, the demand for inside money is equivalent to the supply of loanable funds. This stems from the fact that banks make loans by crediting a customer’s deposit account. Unless the customer is (at least briefly) willing to hold that additional inside money, then no new loan can be created. Thus, the demand for inside money, voluntary savings, and the supply of loanable funds all move together.

The supply of inside money “represents investment” (Horwitz 1990, 5), because the “way that banks add to the money supply is through deposit/loan creation” (Horwitz 1990, 4). Furthermore, those deposits are created in response to investors seeking funds. So new inside money will only be created when investment demand rises. The inside money supply also represents the demand for loanable funds, since such funds take a monetary form in any nonbarter economy, and little or no outside money is in circulation (Selgin 1988a, 31). Thus, the supply of inside money, investment, and the demand for loanable funds all move together.

One may conceive of two markets: (1) the interaction of ex ante savings and ex ante investment, and (2) the interaction of the supply of and demand for credit (or loanable funds). The latter will be perceived by banks in the form of an interaction between deposits and loans. This follows from the realization that deposits represent the supply of credit and loans represent the demand for credit. The natural rate of interest is that which equates ex ante savings and ex ante investment, with savings assumed to be positively related, and investment negatively related, to that natural rate.¹² The market rate of interest is that which equates the supply of credit (deposits) with the demand for credit (loans). It is assumed that deposits are positively related and loans negatively related to the market rate.

Given the above relationships, one can illustrate the interest rate effects of those parametric changes dealt with in Chapter 2. Specifically, Figures 12, 13, and 14 reveal the impacts on the natural rate of interest (i_n) and the market rate of interest (i_m) of, respectively, an increase in k, an increase in productivity (decrease in goods’ production costs), and a preference shift toward outside money and away from inside money. As k rises (with productivity and real income constant), both the supply of and demand for inside money rise.¹³ Therefore, savings, investment, deposits, and loans all increase (Figure 12). As long as monetary equilibrium is maintained, the shifts of each pair of schedules will be equiproportional, since the time markets reflect the money market. The natural and market rates of interest remain constant and equal to one another. Also, these rates of interest keep pace with the price level, which also does not change in this instance. If productivity increases (with k constant), then (1) real income and nominal money demand rise and (2) the nominal money supply and the price level fall. As a result, in the markets for time, savings and deposits rise, while investment and loans decline (Figure 13). Both the natural and market rates decline and remain equal. If consumer preferences shift toward outside money (a redemption run), then there is a net decline in the inside money supply and inside money demand, with the price level exhibiting no net change. Thus, savings, deposits, investment, and loans all decline, and the natural and market rates remain, once again, constant and equal (Figure 14).

Figure 12
Interest Rate Effects of an Increase in k

Figure 13
Interest Rate Effects of a Decrease in Goods ’ Production Costs

Figure 14
Interest Rate Effects of an Increase in k_o/Decrease in k_i

In short, under free banking, the time and money markets will remain in equilibrium individually and in relation to one another. Interest rates will track changes in the price level, and neither excess money nor excess credit will be created. There will be neither inflation nor unjustified deflation. Producers’ profits will be neither artifically expanded nor artificially contracted. Perhaps most important of all, the market rate of interest will keep pace with the natural rate.

These conclusions bring one to confront the issue of business cycle theories. Reflecting upon free banking, two very provocative theses emerge. The first suggests that the dominant approach to business cycles is not instructive or even meaningful in a free-banking context. The second proposes that certain alternative approaches are, at root, either different facets of the same theory or, at least, complementary theories.

KEYNESIANS, MONETARISTS, AND AUSTRIANS

The set of propositions and policies known as Keynesianism completely dominated macroeconomics from the late 1930s to the mid-1970s. It took the appearance of inflationary recessions (“stagflation”) during the 1970s to impel the economics profession to question the validity of those propositions.¹⁴ Even now, most economists seem to take for granted much of that which comprised Keynesian thought.¹⁵ It is impossible in a short space to present Keynesianism in a fashion that does it complete justice (the same, of course, is true of the other schools of thought discussed below). Nevertheless, in order to grasp why the Keynesian critique of capitalism is irrelevant to free banking, one must have some idea of the fundamentals of that critique.

Keynesianism proposes that consumers may exhibit a sudden increase in their demand for money (“hoarding”) as a result of, say, a lack of confidence regarding future economic conditions.¹⁶ Thus, consumption expenditures decline since consumers are saving more in the form of cash balances. Ex ante savings exceed ex ante investment. This has effects in both goods markets and factor markets. Consumer demand has fallen off, so producers experience an unwanted, ex post increase in inventory investment. That is, excess inventory accumulates. Assuming that nominal wage rates and goods’ prices are generally “rigid downward,” two things result: (1) producers cannot sell their surplus, so (2) they cut back on production by laying off workers. The free market has failed, and Say’s Law is invalid. The government must come to the rescue via either fiscal or monetary policy. It may increase its expenditures so as to stimulate product demand and, thereby, the demand for labor,¹⁷ or it may increase the money supply such that the price level rises. This will reduce the real wage rate and increase the demand for labor.¹⁸

With free banking, however, an increase in the demand for inside money brought about by an increase in k will be matched by an increase in its supply. Equilibrium in the markets for money and time will be maintained. The Keynesian story of excess savings or underconsumption seems not to apply. As Horwitz puts it:

The only way general underconsumption could occur under free banking is if a significant loss of confidence in the banking system led to an unexpected demand to redeem liabilities in the reserve money. Such runs have been extremely rare in competitive systems. . . . Keynes’s essential complaint is actually against not an oversupply of goods but an undersupply of money. . . . Keynes’s complaints should not have been directed against Say’s Law and the market per se, but against central banks. . . . Keynesian concerns are, at best, valid only in the absence of monetary competition. (1989, 429)

Two schools of economic thought that have presented powerful challenges to the Keynesian orthodoxy are the monetarists and the Austrians. Both espouse free markets, and both insist that disruptions in the market for money are the primary cause of business cycles. However, they differ—often bitterly—over certain specifics of policy as well as over fundamental methodological issues.¹⁹ Nevertheless, it will be seen that monetarists and Austrians tell business cycle stories that are either parallel or complementary, rather than contradictory.

Monetarism proposes that the demand for money is a stable and predictable function of certain variables: permanent income, the ratio of human to nonhuman wealth, the expected rate of inflation, and expected interest rates (Friedman 1971, 11–14). Furthermore, real output is, in the long run, determined by “real factors” (supplies of material resources, size of the labor force, per-capita productivity, etc.). However, in the short run, real output may be affected by monetary conditions.

Since the demand for money is stable, any significant fluctuations in the money supply are likely to result in monetary disequilibrium and bring about macroeconomic disruption. Business cycles begin with an increase in the money supply such that the money supply exceeds money demand at the existing level of prices. As moneyholders spend their excess cash balances, prices rise. If the resulting actual rate of inflation exceeds the expected rate of inflation forecast by workers, then real wage rates fall, the demand for labor rises, and unemployment declines below the natural rate (assuming nominal wage rates are bid up by an amount less than the actual rate of inflation).²⁰ In order to maintain the low rate of unemployment, the rate of money growth and, thus, the rate of inflation must not only continue, but accelerate. Eventually the monetary authorities will become concerned about the rapidly rising inflation and begin to reduce the rate of growth in money (if they do not, then hyperinflation is inevitable). As the growth in money slows down, the rate of inflation falls. This fools workers in the opposite direction. The actual rate of inflation is less than the expected rate, real wage rates rise, and unemployment increases. In order temporarily to enjoy low unemployment, the society must first endure rising inflation and then, later, high unemployment.²¹ The expansion necessitates the contraction. Stability is regained when the actual rate of unemployment once again equals the natural rate of of unemployment.

The monetarist focus is primarily on the markets for labor, goods, and money. In what seems to be a contrasting approach, Austrians concentrate on the markets for time, goods, and money. The basic Austrian analysis is as follows.

Money enters the economy not as “helicopter money,” that is, not as equiproportional increases in everyone’s nominal cash balances, but in specific markets such that it brings about changes in relative prices. These relative price changes can be disruptive even if the overall price level does not change. The most important relative price is the price of credit—the market interest rate—which affects the extent to which there is intertemporal coordination.²² If the central bank increases the supply of bank reserves and, thus, the monetary base, and assuming that an increase in money demand did not precede the increase in reserves, then the result will be both an excess supply of money and an excess supply of credit.²³ Since goods markets often adjust more slowly than do financial markets, the market interest rate will tend to fall quickly, while prices of goods will rise, but more slowly.

The decline in the market interest rate is not a reflection of a decline in consumer time preferences, that is, consumers’ desired rate of consumption versus savings. Such a decline in the rate of time preference would have been manifested by an increase in the demand for money (increase in voluntary savings). What this means is that the natural rate of interest has not changed. Here is the crucial point for Austrians: Business cycles are the result of differences between the market rate of interest and the natural rate.

The lower market rate induces entrepreneurs to undertake additional investment projects.²⁴ Since voluntary savings have not increased, ex ante investment must exceed ex ante savings, but ex post, the two must be equal. This necessitates the appearance of “forced savings.” “The forced savers are the existing holders of money. Their ability to consume is impaired by the influx of new purchasing power represented by the excess supply of money” (Horwitz 1990, 12). The demand for capital goods has increased, but the demand for consumer goods has not declined, because the boom in capital goods leads to greater incomes for the resource owners in that sector. This increase in incomes is spent primarily on consumer goods. The economy appears to be “booming.” Sooner or later, however, entrepreneurs discover that the real resources necessary for their capital projects will not be forthcoming. Also, the prices of inputs rise as firms bid for factors of production. Profits, which at first appeared robust, begin to dwindle. The supply of credit declines, and interest rates rise, further reducing profitability. Projects are cancelled, workers are laid off, losses are incurred, and the recession follows. According to this Austrian malinvestment approach, the seeds of the recession may be found in the expansionary period, the fluctuations in the prices and production levels of capital goods will be greater than the fluctuations in the consumer goods sector, and stability returns when the market rate of interest once again equals the natural rate.

Both monetarists and Austrians place the primary blame for business cycles on central banks and their predilection for expansionary monetary policy. Monetarists emphasize the relations between the money supply and inflation, real wage rates, and unemployment. Austrians emphasize the effects of the money supply on credit markets, interest rates, and the structure of production. Both build their theories around unobservable variables: monetarists are concerned about departures from the natural rate of unemployment, and Austrians about departures from the natural rate of interest. Are these not two sides of the same coin? If the market rate of interest is less (greater) than the natural rate of interest, then surely it is also true that the actual rate of unemployment is less (greater) than the natural rate of unemployment. As unanticipated monetary expansion produces inflation, lower real wage rates, and increased employment, it must also bring about lower interest rates and greater production of both capital and consumer goods. Furthermore, when economic agents eventually perceive conditions correctly, the resulting increase in the expected rate of inflation will both (1) raise real wage rates and reduce employment, and (2) raise the market interest rate and reduce production. These may be argued to be two halves of the same basic theory. At the very least, they are complementary.

Any significant commonality between these two theories is usually denied vigorously by both sides. Nevertheless, this is not the first work in which parallels have been drawn between Austrians and monetarists. Horwitz (1990) deals sympathetically with both, and suggests that Austrians offer the more compelling analysis of inflation, but monetarists the better analysis of deflation. Thomas M. Humphrey goes so far as to declare that “monetarist and Austrian theories of the business cycle share many of the same or similar characteristics. Because of this, the two approaches should be seen as complementary rather than as competing” (1984, 19). More recently, Humphrey (1990) has argued that Knut Wicksell’s interest rate model (which forms the basis of much Austrian thinking) is quite similar to the monetary model of Irving Fisher (one of the most influential monetarists). Finally, it is interesting to notice that David Laidler, a well-known monetarist, describes “new-classical” economists²⁵ both as economists who wished “to restate monetarist analysis with greater rigour than its pioneers” (1990, 57) and as “neo-Austrians” (1982, 77–83).

SUMMARY

This chapter has discussed several related issues: Say’s Law, business cycles, and free banking. Business cycles were identified as fundamentally a problem of microeconomic discoordination, which resulted in a disparity between effective demand and notional demand. Say’s Law, or more precisely that version of it known as Walras’ Law, identified the likely source of the discoordination: money. The nature of free banking makes it manifestly clear that in a free market context, the markets for money and time are interwoven. That is, money affects both the markets for various present goods and services, and the intertemporal market for present goods versus future goods. It was argued that free banking not only will tend to keep money in equilibrium, but also to keep the market interest rate equal to the natural rate. Finally, certain strong parallels were drawn between the monetarist and Austrian business cycle theories. It was suggested that these two theories are more nearly reinforcing than conflicting, despite the common belief to the contrary.

Chapters 2 and 3 have tried to present a theoretical foundation for the proposition that free banking tends to maintain monetary equilibrium, or alternatively, one might say that free banking minimizes business cycles (the expected value of the difference between effective demand and notional demand is zero). Can central banking do likewise? That is the topic addressed in Chapter 4.

NOTES

1. The discussion that follows owes much to the work of Steven Horwitz (1988, 1989, 1990). Horwitz has very skillfully articulated several ideas on which the author had himself been working independently.

2. See Chapter 2.

3. Of course, one might ask when this price level prevails. If money supply and money demand were to be equated at the price level that existed when the productivity gains were first experienced, then money holdings would be excessive, and the spending of the excess would prevent the price level decline. Money supply and money demand must be equated at that price level that reflects the decrease in production costs caused by the gains in productivity. Selgin’s is a “productivity norm” policy, not a policy of strict price level stabilization. His operative definition of “monetary equilibrium” should take account of the distinction. Those who opt for a stable price level as the preferred policy goal will want to retain a definition that makes no mention of productivity fluctuations.

4. One should note that the formal model of the previous chapter dealt with both micro- and macroaspects of the various parametric effects. Nevertheless, considerable aggregation was, admittedly, involved in the model. Much of the beauty of the complex matrix of individual interactions was traded for the crude simplicity of a mathematical model. Such a procedure seems fully justified pedagogically, at least as long as one does not fail to make it clear that the aggregates are the abstraction and the individuals are the reality.

5. Thomas Sowell (1972) presents a detailed, but very readable analysis of (1) the various interpretations of Say’s Law and (2) the controversies surrounding it, particularly those of the early nineteenth century.

6. Later in that century, one of the most famous critics of Say’s Law appeared—Karl Marx.

7. Alternatively, one might phrase this as: Will ex ante production actually equal ex post demand? As will be seen, there is a temporal aspect to this crucial question.

8. As Sowell puts it, “Classical economists referred to the restoration of equilibrium in terms of internal proportions, not aggregate quality (or value) of output” (1972, 36). Their primary concern was with microeconomic coordination rather than with the interactions of aggregates.

9. The author would be sorely tempted to add knowledge as the third universal. Is not economics fundamentally about the generation and use of knowledge, and is not time the dimension along which one searches for knowledge, and money the medium that conveys knowledge?

10. It should not be supposed that the discussion that follows reflects in every particular Horwitz’s presentation of these issues.

11. It is assumed that there exist no legal constraints that artifically segment depository institutions into “commercial banks,” “thrifts,” and “credit unions.” Therefore, the term “bank” here may be taken to represent any financial intermediary that accepts deposits. One should see Nicholas Lash (1987) for a succinct survey of restrictions on the financial services industry.

12. This is the conventional definition of the natural rate. An alternative, but equivalent, definition states that the natural (or “pure”) rate is that which equates the supply of present goods (demand for future goods) with the demand for present goods (supply of future goods) (Rothbard 1970, 319–23).

13. See the discussions in Chapter 2 regarding this as well as the other two parametric effects dealt with here.

14. Some economists never found Keynes’ thinking to be persuasive. These men—Mises, Hayek, Hutt, Robertson, et al.—defended Say’s Law and free markets from the first appearance of Keynes’ General Theory of Employment, Interest, and Money in 1936. In the last twenty years, criticism of Keynes has become both more common and more diverse. The recent critics of Keynes include Milton Friedman, Murray Rothbard, and Robert Lucas, among many others.

15. The past tense is employed here because the appellation “Keynesian” has fallen out of favor. Those who subscribe to (modified versions of) Keynes’ ideas now call themselves “neo-Keynesians” or “post-Keynesians.”

16. The same general scenario can, per Keynes, be precipitated by a sudden decline in ex ante investment because of a loss of confidence on the part of entrepreneurs.

17. One may legitimately question whether the demand for goods is necessarily and proportionately a demand for labor.

18. This assumes a stable relationship between inflation and unemployment, that is, a stable Phillips curve tradeoff.

19. For example, one may say that, generally, Austrians favor free banking on a gold standard, and monetarists favor central banking with fiat currency but subject to some sort of strict monetary growth rule. As for methodology, Austrians are very skeptical of the use of mathematical models and econometric testing of hypotheses, whereas monetarists routinely construct formal models and undertake statistical tests.

20. This will likely be the case if either (1) workers form their inflationary expectations adaptively, that is, based purely on past trends, or (2) labor contracts are typically negotiated for long time periods, for example, for a two-year period.

21. In other words, the negative relation between inflation and unemployment applies to the short run, but becomes a positive relation in the long run.

22. Significant portions of the Austrian theory can be traced back to the work of Knut Wicksell.

23. As seen earlier, in an all-inside-money regime, an excess supply of money would represent an excess demand for credit, and such excesses would only be temporary phenomena.

24. Non-Austrians seem often to misunderstand this point. They interpret this to mean that business cycles are caused only by “low” interest rates (thought of in absolute terms). Austrians insist that what they mean is simply interest rates that are lower than they would have been in the absence of the monetary expansion (in relative terms). From an empirical standpoint, the problem is that the Austrians’ benchmark of stability, the natural rate of interest, is unobservable, but then, is not the natural rate of unemployment also unobservable?

25. New-classical economists assume that individuals’ expectations are formed “rationally” (people cannot be systematically deceived regarding the actual rate of inflation) and that markets clear continuously (economic agents never fail to take advantage of opportunities to increase their utility). These assumptions lead such economists to propose that government policies will seldom have any impact on real variables, such as the rate of unemployment or real output.