Competition and Finance

The principal purpose of this book is to explore financial and monetary laissez-faire. In so doing, I also hope to provide some insight into real-world financial and monetary problems, and, in particular, into the issue of the appropriate role of the state. The issue of financial laissez-faire — or free banking — is of course highly controversial, and there is little doubt that free banking is still a minority view.¹ Nonetheless, I believe that free banking is both feasible and desirable, and that the main criticisms levelled against it are essentially unsound. My main theme, therefore, is that free trade is just as desirable in the financial sector of the economy as it is in any other sector. However, financial free trade in its broadest sense means much more than just, say, the deregulated banking on a given monetary standard (e.g., a gold standard) on which much of the free banking literature has hitherto focused. To examine financial laissez-faire in the context of a given monetary standard is useful, but limited, and we really need to go further and examine laissez-faire in a broader framework in which the monetary standard itself — that is, the set of rules governing the issue of currency, which indirectly determine the price level and other nominal variables — is endogenised. Financial laissez-faire in its broadest sense therefore includes the various monetary issues related to the determination and operation of the monetary standard, as well as the financial issues on which free bankers have hitherto tended to focus.

Economists have long been interested in how firms and individuals finance their activities, and there is a vast literature dealing with the relative supplies of different types of financial instruments. Two types of financial instrument are particularly important: debt contracts that promise investors a specified return in ‘normal’, non-default states and first claim to the issuer’s assets in default states; and equity contracts that give holders a claim on an issuer’s residual income (i.e., its profit) and ultimate control over its assets provided the issuer does not default. Traditionally, the finance literature has tended to focus on the supply of and demand for these financial instruments, but had relatively little to say on why those particular contract forms are used in the first place.¹ This latter question — the ‘security design’ problem, as Harris and Raviv (1991, 2) call it — was neglected, and yet it has become increasingly apparent in recent years that the answers to some of the most basic issues in financial economics actually depend on it. If we do not understand why agents use the contract forms they do, then we can only have, at most, a limited understanding of firm capital structure and financing decisions, and, therefore, of everything that depends on them (e.g., the activities of financial intermediaries). A clear understanding of contract form is thus essential if we are to understand the more sophisticated issues to be considered later. Our first task must therefore be to establish what ‘optimal’ contracts look like, and how they relate to the contracts we observe in the ‘real’ world.

Chapter 2 looked at the basics of contract theory in a setting with two classes of agent. This chapter builds upon that discussion by extending the treatment of contracting theory to encompass contracts involving three or more types of agent. Instead of there being just equity-holders and a single class of undifferentiated debtholders, we now consider cases where there might be equity-holders and two classes of debt-holder (i.e., one with a senior repayment claim on the firm and one with a junior claim), or where the manager of a debt-issuing firm is no longer sole owner, and so we have two types of equity-holder — the inside holder (i.e., the manager) and the outside equity-holder (i.e., the other equity-holder(s)) — as well as the debt-holder. Indeed, we often have situations where there is inside and outside equity as well as junior and senior debt, and we frequently get cases where the firm issues other types of financial instrument as well (e.g., subordinated debt).

The essence of financial intermediation is the use of a third party to facilitate the transfer of information or wealth between two others. Financial intermediaries exist because they improve on unintermediated markets in which the ‘ultimate’ parties (such as borrowers and savers, or firms and investors) deal directly with each other without the use of any intermediary. An intermediary might be useful in a situation where investors want information about prospective investment projects, but are placed or or lack the expertise to collect that information themselves. The intermediary might then be able to provide that information at cost lower than that at which the investors could provide it for themselves. This type of intermediary is known as a broker, and there are many different varieties of brokers, including investment analysts, credit-rating agencies, auditing firms, and various other specialist information services. However, many intermediaries also go beyond the mere provision of information and actually invest on behalf of their clients whom they issue with claims against themselves. These intermediaries can be broadly divided into two types — banks and mutual funds — which are distinguishable from each other by the types of liability they issue. These latter intermediaries might also go beyond providing a conduit for borrowers and lenders to trade with each other, and provide the means of payment or a payments system that agents use to make payments to each other (e.g., by issuing currency).

A mutual fund can be defined as a financial intermediary that issues its own liabilities and uses the proceeds to buy income-earning assets, but issues only one kind of liability (see, e.g., Glasner, 1987; Cowen and Kroszner, 1989; and Wicker, 1988). A mutual fund therefore differs from a broker in that it holds assets and issues liabilities as part of its intermediary function; it also differs from a bank in that it issues only a single class of liability — which, since it is a claim to the fund’s residual income, can usually be regarded as a form of equity — while a bank issues at least two types of liability, namely, debt and equity. Alternatively, a mutual fund can be defined as an investment company that invests deposited funds in a portfolio of investments, but is set up in such a way that shareholders (or residual claimants) and depositors are identical (see, e.g., Ippolito, 1992, 831).

A bank is a financial intermediary that holds assets and issues liabilities, but issues at least two classes of liability — typically, forms of debt and equity — with differing claims to its assets in the event it defaults.’ A bank is like a mutual fund in so far as it is a financial intermediary that invests on its own account, and issues claims against itself to the ‘ultimate’ investors who provide the funds it invests. However, a bank is unlike a mutual fund and like most other large firms in capitalist economies and in so far as it has a capital structure consisting of two or more distinct classes of liability. Indeed, one can think of a bank simply as a debt-andequity-issuing firm that is engaged in the business of financial intermediation, in a manner analogous to the way in which, say, a typical steel-producing company is a debt-and-equity-issuing company engaged in the business of making steel.

This chapter considers the scale of operation of the banking firm. We focus on two key issues. The first is whether the banking industry shows sufficiently large economies of scale that only one firm survives in the competitive equilibrium, i.e., we consider whether banking is a natural monopoly. This is an important question because arguments that banking is a natural monopoly have sometimes been used to justify government intervention into the banking industry, or else to make the case that there is no point deregulating banking because the industry — or, more specifically, some aspect of it, such as the note issue — would be a monopoly anyway even under laissez-faire. Our discussion suggests that the claim that banking is a natural monopoly is theoretically weak: theory suggests that banking is subject to scalar economies, but these economies are not sufficiently pronounced to make banking a natural monopoly. The theory is also confirmed by the empirical evidence, and there is in fact no evidence at all that banking is a natural monopoly.

We end our present treatment of banking by considering some recent theoretical work on banking instability. This work has attracted considerable attention, and purports to provide rigorous theoretical justifications for government deposit insurance and other features of modern central banking. The key paper in this new literature is Diamond and Dybvig (hereafter DD, 1983). This paper posed two questions that have dominated the subsequent literature. First, why are financial institutions subject to instability and, as a subsidiary question, why is this instability damaging? Second, what, if anything, should the government or central banking authorities do about this instability? DD suggested that the instability arises because depositors’ liquidity demands are uncertain and banks’ assets are less liquid than their liabilities, and they suggest that the instability is harmful because it ruins risk-sharing arrangements and damages production. They also argued that this instability creates a need for government deposit insurance or a lender of last resort to provide emergency loans to banks. Later literature in the DD tradition develops these answers further and has been used to justify additional policies such as interest-rate ceilings (e.g., Anderlini, 1986b; Smith, 1984) and reserve requirements (e.g., Freeman, 1988).

This chapter addresses various issues related to exchange media and payment systems. We begin with the media of exchange issues. A medium of exchange (MOE) may be defined as any object which is taken in exchange, not to be consumed or used up by the receiver, but with the intention of being exchanged for something else at some point in the future. The first issue is therefore to explain the development of a commonly or generally accepted MOE. Economists broadly agree over how a commonly accepted commodity MOE can arise spontaneously from initial conditions of barter, but what happens next is more controversial. The controversy centres on the claim made by Kiyotaki and Wright (1989), and others, that commodity currency would then directly give way to fiat or inconvertible paper currency. However, such a development is in fact most unlikely. The commodity currency actually would give way to convertible paper currency and that the currency would remain convertible thereafter. Convertibility arises and then persists because the public demand it, and they demand it because of the reassurance it provides them regarding the value of their currency. An implication of this position is that real-world fiat currencies did not arise spontaneously, but because of government intervention to suppress the convertibility guarantee.

It is time now to examine the unit of account (UA) in more detail. We are concerned here with three principal issues. The first is the nature and uses of a unit of account. The UA is a unit of that commodity or asset — the medium of account (MOA) — in terms of whose units ‘prices are quoted and accounts are kept’ (Niehans, 1978, 118). The UA is that to which the numbers refer in price tags. The medium of account has a name — the dollar, the pound, or whatever — and refers to a specific commodity or asset, but prices are actually expressed in terms of dollar-units. The MOA is the dollar (or pound, or whatever), but the UA is the dollar-unit. This distinction is important because what matters is not so much the choice of unit per se, but the choice of whatever it is whose units are used to express prices. To illustrate, it is hard to believe that there would have been any significant economic effects if the USA had adopted the cent-unit as its unit of account instead of the dollar-unit. All nominal values would have been 100 times greater than they now are, but real values and economic outcomes would have been much the same. However, the monetary history of the USA would have been very different indeed had the USA adopted the pound sterling as it’s MOA instead of the dollar — American inflation would have been determined by the policy of the Bank of England, and so on. It is the choice of MOA that matters, and not the choice of UA as such.

Chapters 12–14 consider various mechanical issues relating to the operation of the monetary standard, and we begin in this chapter with the mechanics of a convertible monetary system.¹ We can talk of the convertibility of a monetary system in two different senses. In a weaker and less familiar sense, convertibility involves a commitment by one or more banks of issue to maintain continuously or every so often a pre-announced exchange rate between the MOE they issue and something else (i.e., they peg the latter’s nominal price). For example, they might undertake to peg the nominal price of an ounce of gold throughout each trading day, or on the first day of every month or year. We can also talk about convertibility in a stronger and more familiar sense. ‘Strong’ convertibility involves a similar price-pegging mle, but also includes a supplementary requirement that banks continuously offer an ‘over the counter’ (OTC) service to the general public by which they commit themselves to redeem and sell their liabilities for some medium (or media) of redemption (MOR) at an exchange rate determined by a preannounced formula.² There may be certain conditions attached to the banks’ obligation to buy and sell their liabilities (e.g., members of the public might have to give advance notice to withdraw certain deposits), but in what follows it is easier to abstract from such conditions and implicitly presume that the banks agree to redeem and sell their liabilities on demand without notice.

This chapter and Chapter 14 consider commodity-based monetary standards, i.e., monetary standards in which the value of the unit of account is tied by some rule to specific quantities of one or more ‘real’ commodities. However, before we can discuss such systems, we need first to clarify what the term monetary standard actually means. Too many writers launch off into detailed discussions of particular monetary standards without ever specifying what a monetary standard as such actually is. As Mason wrote in his Clarification of the Monetary Standard (1963), the concept of the monetary standard

has been almost entirely neglected in modern literature. Dictionaries and encyclopedias completely ignore [it] … Even writers who prefaced their treatment of monetary standards with a purported delineation of basic concepts typically ignore the basic concept. (Mason, 1963, 3–4, his emphasis)

This chapter addresses price level optimality, but our focus of interest is with price-level changes over time rather than with the absolute level of prices as such. The absolute level of prices is of little or no interest in itself, but price-level changes can have profound and complex consequences. Most economists believe that a continually rising price level (i.e., inflation) damages the economy and, indeed, the social order more generally, but there is much controversy over how it does so and how damaging it is. This chapter tries to address these issues and come to a judgement on what the optimal price-level path might be.

We have so far focused on hypothetical laissez-faire systems, but what about modern real-world systems with their ubiquitous state intervention? What bearing does laissez-faire have on the real world? The answer is that it provides an ideal benchmark — not only a benchmark, but also an ideal to be aimed for — against which we can measure real-world systems. As we saw repeatedly in earlier chapters, and despite many claims to the contrary, there are no convincing reasons to suppose that laissez-faire ‘fails’ in any reasonable sense — most particularly, laissez-faire provides a safe, stable and efficient financial system, and ensures that the currency has a guaranteed stable value. Laissez-faire is highly efficient, but it is efficient not only in the narrow sense most familiar to economists, but also in a broader institutional sense — the institutions that arise under laissez-faire serve socially useful roles, they are regarded as legitimate, and so forth. The key to its efficiency is very simple: laissez-faire harmonises the disparate interests of different individuals, interests which we cannot assume to be naturally harmonious, and which, under alternative arrangements, could easily produce all sorts of undesirable outcomes. In the words of Adam Smith, a man will be more likely to obtain what he needs from others