Monetary Evolution, Free Banking, And Economic Order

Steven Horwitz

• Publisher: Routledge
• Year: 2019
• Print ISBN: 9780367007874
• eBook ISBN: 9780429722882
• Topic: Business
• Subtopic: Business General
• Index: Business

This book deals with the origin and functions of money and banking, emphasizing the role both play in the promotion of economic order. Developing the insights of Hayek and others of the Austrian tradition, Professor Horwitz argues that an appreciation of the spontaneous evolutionary processes that produce and maintain our monetary institutions shou

1
Problems with Formal Models of Monetary Exchange

Despite its pervasive presence in the economic and social landscape, money has posed numerous problems for modern economic theory.¹ What to the outsider must seem like the most obvious, most common, and most fundamental aspect of modern economies, the use of money, is to the economist a source of much discomfort and disruption. In short, the properties of real-world moneys throw a monkey wrench into the neoclassical theory of economic exchange. Neoclassidsm’s most refined set of tools, general equilibrium theory, allows little place for money as a generally accepted medium of exchange, arguably its most important function. In spite of this, theorists have continued to try to use the mathematical and geometrical formalizations of equilibrium theory to construct explanations of the real-world origin and use of money.

Such attempts have met with very limited success, and continued efforts in this direction are likely to be constrained by the inherent limits of the formalistic conception of theory held by neoclassical economists. After overviewing the main problem involved in constructing a general equilibrium theory of money, we will proceed to examine critically several previous attempts to do so. This critical analysis will also indicate the broad outlines of an alternative approach to monetary exchange, one that will be explored in succeeding chapters.

The Fundamental Problem

At the bottom of the problem of explaining money is the nature of the tools that the standard economist uses. General equilibrium theory is accepted by a majority of economists as their scientific approach to the study of human behavior.² When such economists analyze a person, institution, or situation it is the constructs of general equilibrium theory that classify their observations and results. To understand why general equilibrium (GE) theory and money are so incompatible, a brief summary of the theory is in order.

Many GE theorists would argue that they are attempting to give a scientific rendering of Adam Smith’s “invisible hand.” Smith argued that if left to their own devices, individuals would be led “by an invisible hand” to produce beneficial outcomes they neither designed nor intended. GE theory can be thought of as describing the end result of such a process; that is, what would the world the invisible hand leads to look like? It would be a world where all possible benefits from exchange have been exhausted. People would be unable to improve their current outcomes without harming another’s. This implies that all markets would need to be cleared. No excess supplies or demands of any commodities could exist. No one would face frustration trying to buy or sell at prevailing prices. Put another way, we can imagine a vector of prices such that the market for every good would be cleared. Is it possible to discover or describe such a vector of prices, and what kinds of assumptions would have to be true in order to arrive at it? These kinds of questions have motivated the development of GE theory.

For general equilibrium to exist, several other things must be true. First, agents must both have perfect knowledge of their own tastes and preferences and know and take as given all of the opportunities available to them in the market. Second, utility functions must obey certain regularity conditions. Third, no trading can take place until the GE price vector is found. These assumptions are necessary to ensure that no false trading takes place. Incomplete knowledge, that is, false trading, would allow for nonmaximizing situations, incompatible with the overall view of a perfectly equilibrated market. Agents in this model match their own preferences (through curves of indifference generated from self-knowledge of their utility functions) with the trade-offs available in the market. Each agent maximizes his utility subject to this budget constraint. At such a point of maximum utility, the agent by definition has equalized the ratio of the marginal utility of each of his goods to its price: (MU_x/P_x) = (MU_y/P_y) = (MU_n/P_n), and so on. This simply indicates that the agent cannot gain any utility by shifting a dollar’s worth of consumption to any other good given the array of prices.

It should be noted that the assumption of perfect knowledge by definition implies that there is no uncertainty of any kind.³ Agents know what the future will bring, they know what they will buy and sell and when they will buy and sell it. Like a frictionless surface in physics, GE theory puts no barriers in the way of action and reaction. The existence of economic “frictions” would imply that potentially mutually beneficial exchanges fall by the wayside because the costs of exchanging prevent them from taking place.

General equilibrium is a model where agents with perfect knowledge maximize known and given utility functions subject to a known and given budget constraint made up of known and given wealth and prices. There are no costs to transacting, no costs to acquiring information, no uncertainty about the future, none of a number of real-world factors that are part and parcel of a real-world economy. More important, because of these assumptions, the world of GE also has no place for firms (all “production” is a functional response to prices of inputs and outputs), no historical time (all maximization takes place at a point along a discrete time vector, and perfect knowledge precludes the mental changes that are our way of recognizing the passage of time), and no place for money.⁴

Fundamentally, there is no need for money in general equilibrium. The GE world is a barter world because trading is in a significant sense predetermined by the assumption of given knowledge. Agents know what they want and they know what they have. This perfection of knowledge permits barter exchanges to occur. As Hahn (1970, p. 3) has noted, “The Walrasian economy that we have been considering … is essentially one of barter.” Because the GE price vector is found prior to any exchanges, there is no need for a medium of exchange to enter the picture. Goods would not differ in their liquidity, obviating the need for money, which is defined by its superior liquidity. Each good has its own rate of return, and by hypothesis each good is perfectly liquid. This implies the view that money is “barren” (it has no yield of its own),⁵ and sense can be made of Keynes’s (1937, pp. 115–16) quip “Why should anyone outside a lunatic asylum wish to use money as a store of wealth [in that model]?” The very model that economists consider to be the foundation of their analysis of prices produces a world where the use of money, the object with which real-world prices are formed, is meaningless. Indeed, then, it should not be a surprise that attempts to find a role for money in GE models have met with only marginal success.

Recognition of the paradox of a general equilibrium theory of money is not absent in the literature. Paul Davidson (1978, p. 140), for one, argues that “in a neoclassical world of perfect certainty and perfect markets … it would of course be irrational to hold money.” Axel Leijonhufvud (1968, pp. 79–80, n. 25) offers the following observation: “The current status of ‘pure’ monetary theory is a curious one: The preferred analytical tool of many of its most distinguished practitioners is the general equilibrium model. But money cannot be ‘important’ in theories which devote attention only to equilibrium situations.” Karl Brunner and Allan Meltzer (1971, p. 785) note that “the standard theory of exchange, or price, however, provides no hint as to why dominant mediums of exchange emerge.” The basic problem is best summarized by Ross Starr (1972, p. 608): “The standard theory [is] a model of exchange in which money does not appear. This poses a dilemma. How do we make money appear without making the standard theory disappear?”

Neoclassical economists seem to be caught on the horns of a dilemma. Their theory cannot explain one of the obvious facts of the real world. So do they keep the theory and ignore the world or restructure their theory to explain the world? At least to this point the vast majority of neoclassical monetary theorists seem to have taken the former way out. The latest advances in monetary theory, especially overlapping-generations and legal restrictions models, seem to be heading off into ever more esoteric territory and are deriving “moneys” that bear little resemblance to anything seen in any existing economy.⁶

Is Partial Equilibrium a Way Out?

The focus of the previous section was on general equilibrium theory’s difficulties in explaining money. Not all of the economics profession works with GE’s tools. Many prefer so-called partial equilibrium analysis, which concentrates on particular markets, firms, or industries while holding constant the systemwide effects that interest GE theorists.

Although this approach may be more fruitful than GE for some microeconomic analyses, it is not an improvement when we try to deal with macroeconomic problems. By its very nature, money touches everything, and trying to isolate its effects in individual markets ignores fundamental aspects of what money is all about Impounding systemic effects in a ceteris paribus clause throws the baby out with the bathwater.

Ultimately, partial equilibrium analysis must either collapse to sectoral GE analysis or be inconsistent. To the extent that partial equilibrium deals with money, firms, and uncertainty (i.e., all the things absent from GE), partial equilibrium analysts can no longer assume that they are dealing with equilibrium situations and so cannot reach welfare conclusions that require equilibrium. If instead partial equilibrium analysts choose to assume away those things, then partial equilibrium analysis is reduced to sectoral general equilibrium and provides no additional or interesting insights. In either case, it is not a way out.⁷

Classical Foundations of Neoclassical Monetary Theory

Previous attempts to understand monetary exchange are numerous and varied and date back to the beginnings of scientific economics. Authors such as Adam Smith and J. B. Say recognized the advantages of money over barter, and most realized that those advantages cannot be captured in oversimplified comparative statics experiments. The benefits of money can best be seen from a dynamic or evolutionary perspective on economic exchange. In general, money allows social actors to more easily achieve their ultimate ends by eliminating the need for a double coincidence of wants in order to exchange.⁸ Under barter, an exchange partner must “have what you want” and “want what you have.” Money reduces this necessity to a simple double coincidence of “desired exchanges” rather than wants (Jones 1976, p. 761). Money allows us to exchange for a good(s) that we do not directly want but will use to exchange for the goods we ultimately want.

Smith (1976, pp. 26–27) noted the double-coincidence problem and argued that people held certain quantities of desirable goods as a means of overcoming it. Say (1971) anticipated many of the insights of modern monetary theory in his discussion of the essential properties of money. Say (p. 218) had a similar description of the process of overcoming the double-coincidence problem and gave a particularly lucid account of the relative importance of individuals and governments in the development of monetary exchange. William Stanley Jevons’s Money and the Mechanism of Exchange (1969) contains what many consider to be the seminal discussion of many of the problems of modern money theory.

The starting place for most of GE monetary theory is with the work of Leon Walras, the founder of GE approaches. Most Walrasian models simply introduce money as an (n)th good in a general equilibrium system. Added on as an additional good, money can be neatly fit into GE models. Contrary to GE, however, the real-world problem is that money is not simply a good that shares the same qualities as all other economic goods.⁹ The basic point is that money, by definition, is the most liquid good in the economy. GE models, by hypothesis, contain goods with equal liquidity. How then can a theory of equally salable goods make room for a good of extremely high salability?

Problems with the Forgone-Interest View of Money Demand

The lineage of monetary theory, like much of economics in general, begins to diverge in the twentieth century. The focal point of these diverse paths is in the work of Alfred Marshall (1842–1924). One lineage runs from Marshall to Hicks and much of the neoclassical synthesis and monetarism. The other half of Marshall leads to Keynes and the various post-Keynesian approaches to money. Because most of the Keynesian literature is based on criticisms of the neoclassical approach, the distinctly Keynesian arguments will be saved for the end of this chapter.

The next major contribution in neoclassical monetary theory is from Sir John Hicks. In his 1935 article “A Suggestion for Simplifying the Theory of Money,” Hicks tried to offer a synthesis of value theory and monetary theory. Attempts at such a synthesis have been at the root of much of modern monetary theory and attempts to explain the origin of money. Hicks (1935, p. 18) poses the basic question that pervades modern theory:

What has to be explained is the decision to hold assets in the form of barren money, rather than interest- or profit-yielding securities…. So long as rates of interest are positive, the decision to hold money rather than lend it, or use it to pay off old debts, is apparently an unprofitable one.

To explain this apparent problem, Hicks invokes a broad concept of transactions costs. Because investing in interest-earning assets is costly, it is conceivable that the interest minus the transaction cost might be negative, making money a better “investment.” Hicks can continue to assume that money yields no positive return by arguing that no return can be better (in opportunity-cost terms) than the negative return possible from some costly interest-earning assets. He explains the holding of money without recourse to any explanation of what money does; he shows only that other assets are worse choices due to the costs involved. What follows for Hicks is an exhaustive analysis of these costs and the roles they play.

The problem with this approach, and the bulk of monetary theory descending from it, is that it starts off on the wrong foot by assuming that money is “barren,” that is, that it yields no return. The barrenness view has a long history tracing back to Aristotle but runs into difficulties when one reconsiders exactly what is meant by yield or return. Although it is undoubtedly true that money does not yield an objective, pecuniary return as do interest-earning assets, it does yield a subjective, nonpecuniary return.¹⁰

The yield provided by money is subjectively evaluated “availability services,” that is, money provides services to its possessor by being readily available. Because it is perfectly liquid, and acceptable anywhere, money is held so that its purchasing power is available when needed. The need for liquidity arises because exact knowledge of the timing and amount of income and expenditures is impossible. Money is a bridge over this uncertainty, a “stock of wealth” available in an accessible form. One could imagine the problems involved with trying to both know and time income and expenditures exactly so that one could simply liquidate nonmoney assets at the precise instant that their value is required for expenditures. Money removes this necessity. Even if it turns out that not all of one’s money balances are needed this month, the fact that money is available to be spent if needed gives it a positive yield.

Money is not the only good that provides these availability services. For example, one would not say that a fire truck standing in the garage, or an umbrella under one’s arm, is wasteful or barren. Obviously when it is needed, a fire truck renders firefighting services, but when standing it is available, waiting to be needed. Imagine that standing fire trucks were considered to be barren, leading to fire companies lending the trucks for other uses and intending to get them back in...

Table of contents

• Cover
• Half Title
• Series Page
• Title
• Copyright
• Dedication
• Contents
• Acknowledgments
• Introduction
• 1 Problems with Formal Models of Monetary Exchange
• 2 Rules, Institutions, and the Evolution of Economic Order
• 3 Money as the Language of the Market Process
• 4 The Evolution of Monetary Order
• 5 Regulatory Chaos and Spontaneous Order Under the National Banking System
• 6 Conclusion: Money in a Nonrationalist Approach to Economic Systems
• References
• About the Book and Author
• Index