The Social Institution of Money

Geoffrey Ingham. The Sage Handbook of Sociology. Editor: Craig Calhoun, Chris Rojek, Bryan Turner. Sage Publication. 2005.

Introduction

Money is one of the modern world’s essential ‘social technologies.’ Sociology, however, which is claimed to be the distinctive intellectual framework for understanding modernity,’ seems to have ignored money because it is not ‘sociological enough’ (Collins, 1979). A recent revival of interest in the subject only serves to highlight the longer-term neglect (Dodd, 1994; Zelizer, 1994; Leyshon and Thrift, 1997; Ingham, 1996, 1999, 2000a,b, 2001, 2002; Hart, 2000). Aside from reiterating the obvious importance of ‘trust,’ sociology has not addressed the problem of the actual social production of money as an institution. Rather, sociology is concerned with very general descriptions of the consequences of money for ‘modern’ society (Giddens, 1990), its ‘social meanings’ (Zelizer, 1994) and, more indirectly, with the Marxist problem of ‘finance capital.’ This one-sided treatment would not matter if economics provided an adequate explanation of money’s existence and functions, but it does not (Ingham, 1996).

Mainstream economics contains what appear to be contradictory conceptions of money. On the one hand, as in common sense, money is a quantifiable commodity that ‘circulates’ with a ‘velocity.’ In fact, this notion was already anachronistic at the time of its classical statement in Fisher’s ‘quantity theory’ (1907). By the early twentieth century, almost all capitalist transactions were carried out by a stroke of the pen in the book clearing of debits and credits in the banking giro, not by the circulation of ‘money-stuff.’ None the less, a hundred years on, in the era of so-called ‘virtual’ e-money, the analytical structure of ‘quantity theory’ continues to inform orthodox economics (Smithin, 2000; Issing, 2001). On the other hand, neoclassical economic ‘high theory asserts that money is relatively unimportant.’ Money is no more than a ‘neutral veil’ over transactions in the ‘real economy.’ Neoclassical economics’ most prestigious paradigm (general equilibrium theory) acknowledges that it has no place for money in its mathematical models (Hahn, 1982: 1).

This state of affairs is a result of the division of intellectual labour between economics and sociology that followed the methodological dispute (Methodenstreit) in the social sciences at the beginning of the twentieth century (Hodgson, 2001). Money was placed under the jurisdiction of economics; but it is the conception of money held by economic ‘theorists’ that accounts for the inadequate understanding of money in both disciplines. Sociology should reclaim the analysis of money. Not only is money socially produced, it is actually constituted by structures of social relations (Ingham, 1996,2000a).

During the Methodenstreit, an alternative to the economic commodity-exchange theory of money was advanced by the broad ‘Historical School’ and influenced contemporary sociological and heterodox economic thinking. Here, money is neither a ‘commodity’ nor mere ‘neutral veil’; rather, it is a ‘claim’ or a ‘promise’ of payment—that is, a social relation. Banished from mainstream economics and, as a result of economic theory’s hegemony, neglected by modern sociology, this analysis almost was lost to mainstream social science. It is claimed that Marxism avoided these errors (Fine and Lapavitsas, 2000), but this analysis has also been weakened by a conception of money as a ‘mask’ or ‘veil’ over an underlying ‘reality’ (Ingham, 2001).

Differences between the two conceptions are evident in the emphases that they give to money’s ‘functions.’ By the late nineteenth century, the question of what money is had given way, in economic analysis, to an evasive functionalist definition, which remains the standard textbook approach. Money is what money does, and it is said to do three things. It is (1) a measure of value (unit, or money of account); (2) a medium of exchange and means of payment; (3) a store of value. From this deceptively simple starting point problems soon become apparent. Do all the functions have to be performed? Are they all of equal importance? If not, which is definitive? Mainstream economics has focused on money as a medium of exchange; the other functions are assumed to follow from it. The ‘historical’ alternative stresses the importance of money of account as an abstract measure of value, which is stored and transported through time. The issues concern the nature of the relationship between the realm of commodities and the realm of money. Are they one or two realms? Is money any more than an expression of the realm of commodities? Can it have a value outside this realm?

Money in Economics

Economic Orthodoxy: Money in the ‘Real Economy’

Modelling itself on the natural sciences, economics sought to establish deductive generalizations based on the axioms of individual rational choice maximization of utility and the associated equilibrium model of the perfectly competitive market (Machlup, 1978). The ideal type of the ‘economy’ comprises exchange ratios between commodities (object-object relations) expressed in money prices, established by individual acts of utility calculation (individual-object relations). These object-object and agent-object relations constitute what is known as the ‘real’ or ‘natural’ economy. Agent-agent or social relationsform no part of the model (Weber, 1978: 63-4; Ganssmann, 1988).

The ‘real’ economy is essentially a model of a simple ‘natural’ (moneyless) barter economy (for the classic description, see Schumpeter, 1994 [1954]: 277; NB its Aristotelian origins). It describes exchange ratios between commodities, determined by individual calculations of their utilities in bilateral barter exchanges. The key assertion is that barter transforms myriad exchange ratios into a single price for a uniform good. Money is introduced into the model as a commodity that acts as a medium of exchange to facilitate the process—for example, cigarettes in prison. Money is only a medium of exchange. As a commodity, the medium of exchange can have an exchange ratio with other commodities. Or as a symbol, it can directly represent ‘real’ commodities. It is in this sense that money is a ‘neutral veil’ that has no efficacy other than to overcome the ‘inconveniences of barter’ which, in the late-nineteenth-century formulation, result from the absence of a ‘double coincidence of wants’ (see Ingham, 2000a). Money is more efficient than barter, but analytically they are structurally identical.

Menger’s (1892) rational choice analysis of the evolution of money remains the basis for all neoclassical explanations of money’s existence (Dowd, 2000; Klein and Selgin, 2000). Money is the unintended consequence of individual economic rationality. In order to maximize their barter options, it is argued, traders hold stocks of the most tradable commodities which, consequently, become media of exchange—beans, iron tools, etc. Coinage is explained with the further conjecture that precious metals have additional advantageous properties—such as durability, divisibility, portability etc. Metal is weighed and minted into uniform pieces and the commodity becomes money. In short, orthodox economic accounts of money are commodity-exchange theories. Both money’s ‘historical’ origins and its ‘logical’ conditions of existence are explained as the outcome of a natural process of economic exchange (Ingham, 2000a).

The ‘dematerialization’ of money broke this explanatory link between individual rationality and system benefits. Paradoxically, for Menger, ‘institutions such as money make for the common interest, and yet… conflict with the nearest and immediate interests of contracting individuals.’ Why should the ‘individual be ready to exchange his goods for little metal disks apparently useless as such, or for documents representing the latter?’ (quoted in Jones, 1976: 757). Today, neoclassical economics tries to resolve the problem by showing that holding (non-commodity) money reduces transaction costs for the individual (Dowd, 2000; Klein and Selgin, 2000), but only succeeds in exposing the logical circularity of neoclassical economics’ methodological individualism. It can establish only that it is ‘advantageous for any given agent to mediate his transactions by money provided that all other agents do likewise’ (Hahn, 1987: 26). Of course, it is not so much a question of whether it is advantageous to use money if others do, but rather that agents cannot use money unless others do likewise. To state the sociologically obvious: the advantages of money for the individual presuppose the existence of money as an institution.

There are other problems. First, the ‘barter → commodity → money’ transition is not supported by the historical record (Ingham, 2000a; Wray, 2000). Second, the model of the natural barter economy with its ‘neutral veil’ of money is singularly inappropriate for the capitalist monetary system. In the Commodity → Money → Commodity (C-M-C1) sequence of the ‘real’ economy, money exists only as a medium for the gaining of utility through the exchange of commodities. The financing of production does not take place in the model. In the early twentieth century, attempts were made to explain the fact of bank credit, within the framework of ‘real’ analysis. For example, Wicksell’s ‘natural’ rate of interest is a measure of the ‘natural’ propensities and productivity in the ‘real’ economy and not, for example, the power of bankers to set a ‘money rate.’ The ‘natural rate of interest’ is an extension of the ‘neutral veil’ concept insofar as money, in the last instance, can only reflect or express the ‘real.’ In contrast, as we shall see, Weber and Keynes saw that capitalism involves a Money → Commodity → Money sequence (M-C-M1) in which the money side is relatively autonomous. The act of bank lending creates money-capital to finance the future production of commodities (on Keynes, see Smithin, 1994:2). The bank loan—that is, the capitalist’s debt—pays wages which are spent as money.

After the mid-twentieth-century’s Keynesian interlude, orthodox economic theory was restored in Friedman’s ‘monetarism,’ and further problems soon became apparent (Smithin, 1994). In ‘monetarism,’ money is a ‘thing’ whose supply is quantifiable and controllable. (Ceteris paribus, an increase in the supply of money will increase prices.) However, it soon became apparent that it was not clear what should be counted as money- notes, coins, current bank accounts, savings accounts etc. The issue is complex, but the concept of money as a quantifiable and controllable ‘stock’ produced policy incoherence in the continuous proliferation of measures of money- M0, M1,… M10 and so on. Moreover, by the 1990s, monetary aggregates increased as inflation fell, in contradiction of the theory.

However, the fundamental problem in economic orthodoxy, from which all the other difficulties stem, is the misunderstanding and neglect of money of account. Medium of exchange is the key function and it is assumed that all the others follow from it. The ‘natural’ market produces a transactions-cost efficient medium of exchange that becomes the standard of value and numerical money of account. Coins evolved from weighing pieces of precious metal that were cut from bars and, after standardization, counted. However, there are a priori and empirical grounds for reversing the sequence. Money of account is logically anterior and historically prior to the market (Keynes, 1930; Hicks, 1989; Hoover, 1996; Ingham, 1996, 2000a,b; Orléan, 1998; Wray, 2000).

Without further assumptions, it is difficult to envisage how a money of account could emerge from myriad bilateral barter exchange ratios based upon subjective preferences. One hundred goods could yield 4950 exchange rates (Davies, 1994: 15). How could discrete barter exchange ratios of, say, 3 chickens to 1 duck, or 6 ducks to 1 chicken, and so on, produce a single unit of account? The conventional economic answer that a ‘duck standard’ emerges ‘spontaneously’ involves a circular argument. A single ‘duck standard’ cannot be the equilibrium price of ducks established by supply and demand because, in the absence of a money of account, ducks would continue to have multiple and variable exchange values. A genuine ‘market’ which produces a single price for ducks requires a money of account—that is, a stable yardstick for measuring value. As opposed to the commodity duck, the monetary duck in any duck standard would be an abstract duck. If the process of exchange could not have produced the abstract concept of money of account, how did it originate? The question is actually at the very heart of a problem that distinguishes economics from sociology. Can an inter-subjective scale of value (money of account) emerge from myriad subjective preferences? Posed in this way, the question of money is at the centre of the general question in Talcott Parsons’s sociological critique of economic theory—although it has not been seen in this light. From its starting point of individual subjective preferences, utilitarian theory cannot explain social order (Parsons, 1937).

Economic Heterodoxy: Money as Abstract Value and Token Credit

Heterodox monetary analysis has two sources. On the one hand, it can be traced to analyses of the credit-money that appeared in Western Europe in the sixteenth century. The new forms of money were not simply credit in the sense of deferred payment; rather, these ‘credits’ were ‘money,’ in that mere ‘promises to pay’ (IOUs), issued outside the sovereign mints, began to circulate as means of payment. Only later were they backed by metal in a hybrid bank credit/gold standard. The general use of transferable debt is specific to capitalism. ‘Depersonalized’ and hence transferable debt is used as means of payment to a third party: As IOU held by B is used to pay C (Ingham, 1999). After over two thousand years during which coin and money were synonymous, this new money-form posed intellectual puzzles (Sherman, 1997). Some of the answers gave rise to an analysis of money that departed from the Aristotelian theories of commodity money to which all orthodox economic theories may be traced. They led to the idea that all money was constituted by social relations of credit and debt (Ingham, 2000a: 23).

A second source of heterodox analysis accompanied the construction of the nineteenth-century German state. Money’s role in taxation and as the expression of national integrity and power were emphasized (Schumpeter, 1994 [1954]). Knapp’s State Theory of Money (1973 [1924]) challenged economic explanations of money’s properties in terms of the exchange value of its commodity form. By declaring what it will accept for the discharge of tax debt, denominated in its own unit of account, the state creates and establishes the ‘validity’ of money. Private bank notes become money when they are denominated in the state’s money of account and accepted as payment of tax debts owed to the state and reissued in payment to the state’s creditors (Knapp, 1973 [1924]: 95, 143, 196). Money consists in a reciprocal relationship: states issue ‘credits’ to pay for their goods and services which, in turn, must be acquired for payment of taxes. New money cannot be created without the creation of a complementary debt (Gardiner, 1993). Money is a social relation, not a thing.

Money is a ‘token’ that ‘bears’ units of abstract value. ‘State theory’ is also known as ‘chartalism’ (from charta, the Latin for token) and sometimes as monetary ‘nominalism’ (Ellis, 1934). Regardless of its specific form, money is, generically, a credit—that is, it is a claim on goods. ‘State theory’ was anathema to the early-twentieth-century proponents of the commodity-exchange theory of value—that value canonly be established in exchange determined by the forces of supply and demand. In objecting that states could not establish the purchasing power of money, the economic theorists misunderstood Knapp. In fact, his argument helps to resolve commodity theory’s difficulty in trying to identify the ‘moneyness’ by its utility or exchange value alone. Economic theory cannot uniquely specify money—that is, distinguish money from other commodities. Following Knapp, money becomes a commodity with an exchange value only after it has been constituted as money by a social and political process. States establish the validity of money by the proclamation of the nominal unit of abstract value and the acceptation of the tokens that correspond to it.

Together with early English ‘credit theory,’ ‘state theory’ influenced Keynes’s A Treatise on Money (1930). During his ‘Babylonian madness’ in the 1920s, Keynes studied the German Historical School’s work on ancient Near Eastern money. During the third and second millennia BC, their economies were organized with a money of account, but payments were made in commodities, labour service, silver by weight etc. (Ingham, 2000a). ‘Money’ existed for several thousand years before the first use of coinage around 700 BC: ‘Money of Account, namely that in which Debts and Prices and General Purchasing Power are expressed is the primary concept of a Theory of Money.’ Forms of money such as coins ‘can only exist in relation to a Money of Account’ (Keynes, 1930: 3; emphasis added). In other words, the quality of ‘moneyness’ is conferred by the abstract measure of value that is imposed by the state when it writes the monetary ‘dictionary’ (Keynes, 1930:4-5, 11-15).

Keynes also identified ‘Acknowledgements of Debt’ as forms of money (Keynes, 1930: 6-9). The chapter ‘The “Creation” of Bank Money’ provides a description of the creation of new deposits of money by the act of lending in a way that is relatively independent of the level of incoming deposits of savings: ‘There is no limit to the amount of bank-money that banks can safely create provided that they move forward in step’ (emphasis in original). Bank chairmen believe that ‘outside forces,’ over which they have no control, determine their decisions. ‘[Y]et the “outside forces” may be nothing but himself and his fellow chairmen, and certainly not his depositors’ (Keynes, 1930: 26-7). The analysis points to the socially constructed reality of the norms of banking practice. Bank money is the result of the act of lending—that is to say, the social relation of debt constitutes money.

Keynes’s analysis is continued in the heterodox post-Keynesian theory of ‘endogenous’ money (Wray, 1990; Smithin, 1994; Rochon, 1999). French and Italian ‘monetary circuit’ analysis also has Keynesian roots (Parguez and Seccareccia, 2000). The idea that all money is debt is also found in the work of French interdisciplinary social scientists (Aglietta and Orléan, 1998). Other post-Keynesians have returned to Keynes’s inspiration in Knapp to build a distinctive ‘neo-chartalist school’ (Wray, 1998; Bell, 2000, 2001).

Money in Sociological Theory

Both economic traditions have influenced sociology, but orthodoxy has had much the greater impact. (Social anthropology has been similarly affected, see Hart, 2000). Weber and Simmel were influenced by the heterodox ‘Historical School,’ but it is precisely these parts of their work that have been seriously neglected.

Money as a Symbolic Medium

Parsons’s early work played a part in confirming the terms of the division of intellectual labour between economics and sociology. They are distinct, but complementary as he was assured of ‘the essential soundness, from a sociological view, of the main core tradition in economics’ (Parsons, 1991 [1953]). From a sociological standpoint, money is a symbolic generalized medium of communication and interaction (Parsons and Smelser, 1956; Dodd, 1994). It facilitates the integration of the functionally differentiated parts of the social system—in an analogous way to integration through prices in economic theory. But as a ‘symbol’ money is ‘neutral’ insofar as it does not affect the underlying constitution of either the ‘real’ economy or social system. Parsons followed economics’ axiom that value is only realizable in exchange and that money is only a symbol of value—that is, money, as symbolic medium, is without value (Ganssmann, 1988: 308). Like its economic parent, this notion does not grasp the obvious fact of money as a store of abstract value that maybe appropriated. Furthermore, Parsonian sociology not only failed to take into account that domination derives from the possession of money, but also that it derives from control of the actual process of money’s production by states and banks. Apart from a description of money’s integrative functions, all other questions could be left to economics.

This general orientation has persisted in sociology. Habermas, Luhmann and Giddens, for example, have all followed this concept of money as a ‘symbolic token’ or ‘media of interchange [sic]’ (Giddens, 1990:22; see Dodd’s(1994) secondary analysis of Habermas and Luhmann). Money promotes ‘systemic complexity’ and the ‘time space distanciation’ of modernity; but its existence is taken for granted. The importance of ‘trust’ is repeated, but this has ‘as much explanatory value as saying that credit comes from credere’ (Ganssmann, 1988). Explaining money involves the historical explanation of a specific form of ‘social technology’ that accounts for abstract value and transports it through time. To be sure, money has the consequences that Giddens and others outline; but only if the social relations of its production remain intact. In the absence of this analysis, sociology implies a functionalist explanation of money’s existence that parallels the teleological theorems to be found in mainstream economics. Like economics, much modern sociology has lost sight of the obvious. Money is not merely a symbolic token that integrates ‘disembedded’ social systems (Giddens, 1990); it is also value in itself. Control of money’s production is a pivotal social institution.

Marx and Marxist Analysis

Parsons’s dismissal of Marx as a minor classical economist is a gross exaggeration, but it contains a grain of truth. The labour theory of value committed Marx, and his successors, to a version of the commodity theory of money, with all its attendant errors. To this extent, Marx’s general theory of money was mistaken. Most importantly, this attachment to the labour theory of value of commodity money prevented Marx from realizing that his theory of capital as a social relation applied also to money. In particular, he did not fully understand capitalist credit money (Cutler et al., 1978: 24-6). Later Marxist and sociological analyses of ‘finance capital’ have perpetuated the misunderstanding. (on Hilferding’s errors, see Henwood, 1997.)

Like Adam Smith, Marx held that ‘[g]old confronts other commodities as money only because it previously confronted them as a commodity … It acts as a universal measure of value, and only through performing this function does gold … become money’ (Marx, 1976: 162, 188). Precious metal can become a measure of value because mining and minting embody labour which can be expressed in ‘the quantity of any other commodity in which the same amount of labour-time is congealed’ (p. 186). Forms of ‘credit’ are derivative: bank notes and bills of exchange are money insofar as they directly represent both precious metals and/or commodities in exchange.

But Marx’s distinctive departure from classical economics is to show that monetary relationships do not merely represent a natural economic reality, but also mask the latter’s underlying reality of the social relations of production. These constitute the reality that appears in a monetized alienated form. For Marx there are two ‘veils.’ Behind money lie ‘real’ economic forces, as they do in somewhat different manner in the orthodox economics. In turn, behind these economic forces lie the ‘real’ social relations, which also appear as monetary relations. Tearing away these monetary ‘masks’ or ‘veils’ will demystify capitalism and its money, which will become ‘visible and dazzling to our eyes’ (p. 187).

This kind of reasoning is why Marx is considered as a classical sociologist; but it also implies that money can be analytically ‘bracketed.’ Notwithstanding the two ‘veils,’ Marx’s analytical position is similar to classical economics. Emphasis on the social relations of production of commodities and the labour theory of value prevented Marx (and almost all his contemporaries) from recognizing the relative autonomy of the production of abstract value in the form of credit-money, or the more radical position that all money is token credit. Marx dismissed as ‘professorial twaddle’ Roscher’s complaint that economists ‘do not bear sufficiently in mind the peculiarities that distinguish money from other commodities’ (n. 13, p. 187). In company with all commodity theorists, Marx failed to consider money as abstract value, defined by a money of account and sustained by its own social relations of production.

At times, Marx appeared to have grasped that capitalist credit-money can be created autonomously outside the sphere of the production and circulation of commodities; but, then it plays an essentially dysfunctional role. Bank credit could expand beyond ‘its necessary proportions’ and become ‘the most potent means of driving capitalist production beyond its own limits, and this has become one of the most effective vehicles of crises and swindle’ (Marx, 1981: 735-9). Marx was conventional in this view that credit instruments—bills of exchange, promissory notes etc.—were, or rather should be in a rationally organized system, no more than functional substitutes for hard cash.

The anachronistic and misleading commodity-exchange theory of money is evident in Hilferding’s Finance Capital (1981 [1910]), which despite the apparent critique, was entirely consistent with orthodox economic theory of the time. He dismissed Knapp’s ‘state theory’ for ‘eschewing all economic explanation.’ Rather, ‘money … originates in the exchange process and requires no other condition’ (Hilferding, 1981 [1910]: 36; see also 376). Credit creation is anchored in the ‘real’ economy of production and, therefore ‘the quantity of credit money is limited by the level of production and circulation’ (pp. 64-5). Banks aid the capitalist process by garnering the bourgeoisie’s ‘idle capital’ together with the ‘idle money of all other classes for use in production’ (p. 90). All this is perfectly true. But, as Schumpeter and Keynes argued, the differentia specifica of capitalism lies in banks’ ‘endogenous’ creation of new deposits of credit-money ex nihilo—or, more accurately, out of the social relation of debt. This lending isnew money and not merely the collection of pre-existing ‘little pools’ into larger reservoirs (Schumpeter, 1994 [1954]: 1113). In failing to see this essentially capitalist process, Hilferding and generations of Marxists and sociologists have actually underestimated the power of ‘finance capital’!

Like orthodox economics, the Marxist analysis of money has been disabled by the search for the value of money in the commodity (Fine and Lapavitsas, 2000; and see the critique in Ingham, 2001). It has been unable to consider the proposition that all money consists in symbolic ‘tokens’ of abstract value that signify, and are constituted by, social relations of credit-debt. From a sociological standpoint, these social relations must be considered as the ‘reality’ of money. We may now turn to those aspects of Simmel’s and Weber’s work into which these and other ‘Historical School’ arguments were incorporated.

Simmel and Weber on Money

The Philosophy of Money

Unfortunately, sociology has taken Simmel at his misleading word that The Philosophy of Money is not really about money, but rather about how money expresses the essence of modern life (Dodd, 1994: 175). The modern spirit of discontinuous, fragmented, increasingly abstract impersonal relations finds its most perfect expression in money: ‘The more the life of society becomes dominated by monetary relationships, the more the relativistic character of existence finds its expression in conscious life’ (Simmel, 1978 [1907]: 512). This form of ‘sociation’ generates individuality, personal freedom and intellectualism (Dodd, 1994; Turner, 1999). However, in addition to the analysis of the effects of money, The Philosophy of Money contains important, but fragmented, accounts of money’s origins, its essential qualities and how these are produced. Two aspects of The Philosophy of Money have received less attention than they deserve: the analysis of money as abstract value, and as a form of ‘sociation’in itself, that is to say, as constituted by social relations.

Simmel rejects all economic theory, including Marxism, which locates money’s value in the specific substance or content of the ‘money stuff.’ Rather, money is the pure form of abstract value. The value of money does not derive from either the costs of its production, or supply and demand, or labour-value. Rather, ‘[m]oney is the representative of abstract value’ (p. 120). Money is ‘the value of things without the things themselves’ (p. 121). Money is the abstraction of the ‘distilled exchangeability of objects … the relation between things, a relation that persists in spite of the changes in the things themselves’ (p. 124; emphasis in original). Simmel’s critique of commodity theories of money is developed with a dismissal of the implication that measures must have the same quality as the object to be measured—measures of length are long and, therefore, a measure of value must be valuable (p. 131). Some measures of length are long; but as Simmel argued, this is because measure and measured object share the same quality of length. ‘To establish a proportion between two quantities, not by direct comparison, but in terms of the fact that each of them relates to a third quantity and that these relations are equal or unequal’ is one of society’s great accomplishments (p. 146). Thus, following the ‘nominalists’ of the ‘Historical School,’ Simmel asserts the logical primacy of the abstraction of money of account. Money is ‘one of those normative ideas that obey the norms that they themselves represent’ (p. 122) (see Orléan, 1998: money is autoréférentielle; see also Searle, 1995).

Writing at the apogee of the gold standard, Simmel conceded that ‘[m]oney performs its services best when it is not simply money, that is when it does not merely represent the value of things in pure abstraction’ (p. 165). But he does not lose sight of his essential and prescient point: ‘It is not technically feasible,’ Simmel continues, ‘to accomplish what is technically correct,’ namely to transform the money function into a pure token money, and to detach it completely from every substantial value that limits the quantity of money, even though the actual development of money suggests that this will be the final outcome (p. 165; emphasis added). With the breaking of the link between gold and the dollar in 1971, commodity money ceased to exist.

In contrast to orthodox economics, Simmel understands that exchange by money is structurally different from barter. Money is a form of ‘sociation’ and not a ‘thing’: ‘When barter is replaced by money transactions a third factor is introduced between the two parties … the direct line of contact between them … moves to the relationship which each of them … has with the economic community that accepts the money.’ Simmel then endorses the credit theory of money: ‘[t]his is the core of the truth in the theory that money is only a claim upon society’ (p. 177). Indeed, ‘[metallic money, which is usually regarded as the absolute opposite of credit money, contains in fact two presuppositions of credit which are particularly intertwined’ (p. 178). First, the metallic substance cannot be normally tested in cash transactions and is, rather, verified by the secondary characteristics stamped on coins by the issuing authority. Second, people must trust that the tokens of value will retain their value. This may be based on objective probabilities; but this ‘kind of trust is only a weak form of inductive knowledge’ (p. 179). There can never be sufficient information for it to be the only basis for holding money. Additionally, money requires an element of ‘supra-theoretical belief’ or ‘social-psychological quasi-religious faith’ (p. 179). ‘Money is the purest reification of means, a concrete instrument which is absolutely identical with its abstract concept; it is a pure instrument’ (p. 211). And the qualities of this pure abstract value reside in ‘social organisation and … supra-subjective norms’ (p. 210).

Modern sociology’s exclusive emphasis on ‘trust’ tends to trivialize Simmel’s analysis. Like Weber, he saw that the development of modern states and non-metallic, ‘dematerialized’ money were intimately connected. Modern states built themselves, in large part on the basis of credible metallic standards and coinage. Money led to the dissolution of the personalized bonds of feudal relations and the‘enforcement of money transactions meant an extension of royal power into areas in which private and personal modes of exchange had existed’ (p. 185; emphasis added). However, in a dialectical process, the ‘value of money is based on a guarantee represented by the central political power, which eventually replaces the significance of the metal’ (p. 184). In this historical process, coercion, as always, preceded any ‘trust’ in the establishment of a currency. Modern sociological analysis tends to forget that monetary sovereignty was established to a large extent by extreme physical coercion—such as branding on the forehead with coins, and execution for counterfeiting.

However, having rejected essentialist theories of intrinsic precious metallic value and the classical labour theory of value, Simmel is left with the very same problem that the Austrian ‘subjectivists’ had to face—how can myriad individual preferences produce a scale of inter-subjective value. ‘Money as abstract value expresses nothing but the relativity of things that constitute value’ (p. 121); but at the same time, it transcends the relativity of exchangeable values and ‘as the stable pole, contrasts with the eternal movements, fluctuations of the objects with all others’ (p. 121; emphasis added). But how does it do it?

Simmel answers the question with a historical analysis of money’s transformation from substance to pure abstractionIt is full of insights gleaned from the Historical School, but is no more than a description of the process of becoming the non-material abstraction he correctly identified as money. Moreover, it is confused. He failed to see that, if all money is credit, Hildebrand’s barter → commodity → money → credit evolution is a contradiction.

Two fundamental questions remained unanswered in The Philosophy of Money. First, what are the origins of the concept of money as value? Simmel agrees with the Austrian economists that money expresses exchangeability, but sees that it cannot have been the result of the process of exchange. Rather, it ‘can have developed only out of previously existing values …’ (p. 119; emphasis added). But which might these have been? Simmel left no more than scattered clues. Second, how is the abstract value of modern dematerialized money established and maintained? Precious metal is a means of maintaining confidence, but in an ‘ideal world’ money would be no more than ‘its essential function,’ as a symbol of abstract value. Here, Simmel reverts to a thoroughly positivist economic conception of money: ‘[M]oney would then reach a neutral position which would be as little affected by the fluctuations in commodities as is the yardstick by the different lengths that it measures’ (p. 191; emphasis added). In other words, Simmel accepts economists’ ‘ideal world’ in which the value of commodities is the result of the interplay of subjective preferences, mediated by the neutral symbol of money. But this ‘ideal world’ is not explained; it does not have a social structure. We need to turn to Weber for a sociological formulation in which the value of money expresses the social conflict that lies behind subjective preferences.

Weber on Money

The enormous secondary sociological literature on Weber’s analysis of capitalism scarcely refers to his analysis of money. The chapters on money and banking in General Economic History have been almost completely ignored (Weber, 1981 [1927]). Emphasis on religion has led to a distorted view of his work: for example, the underdevelopment of capitalism in China was not so much the result of a religious ethic as the fact that its money was ‘scarcely as developed as Ptolemaic Egypt’ (Weber, 1951: 3). This neglect is more puzzling in light of his lavish praise for Knapp’s The State Theory of Money (1973 [1924]). One would have expected scholars to have followed Weber’s lead in exploring the ‘permanently fundamental importance’ of this ‘magnificent work’ (Weber, 1978: 184, 169; also 78-9).

Money expands market, or ‘indirect,’ exchange by which it is ‘possible to obtain goods which are separated from those offered in exchange for them in space, in time, in respect of the persons involved, and, what is very important, in respect to the quantity in each side of the transaction’ (p. 80). The most important element is not the existence of a commodity-money as a medium of exchange, but the possibility of monetary calculation—‘assigning money values to all goods and services’ (p. 81). Money of account—the ‘continuity of the nominal unit of money, even though the monetary material may have changed’—makes this calculation possible. It is as an abstraction that ‘the individual values the nominal unit of money as a certain proportional part of his income, and not as a chartal piece of metal or note’ (p. 168). Following Knapp, Weber refers to the state’s definition of money in terms of a unit of account for the legal payment of debts, as its formal validity (p. 169).

Weber upheld Knapp’s distinction between the ‘valuableness’ and ‘value’ (p. 193; see also 78-9). But, in addition to money’s ‘formal validity’ (‘valuableness’), there must also exist the ‘probability that it will be at some future time acceptable in exchange for specified or unspecified goods in price relationships which are capable of approximate estimate’ (p. 169). In this emendation of ‘state theory,’ Weber followed economic orthodoxy, and his critique of Knapp’s analysis of inflation is based, to some extent, upon the commodity and quantity theories of money (p. 192, see also 180-4). Weber deplored the kind of disciplinary segregation that eventually came about after the Methodenstreit, but he believed that the analysis of the price of goods—including the purchasing power of money—was more properly part of economics (p. 79). None the less, he was unable to resist, mainly in footnotes, making incisive comments on the nature of economic theorizing. Economy and Society contains the germs of a sociological recasting of a ‘substantive theory of money’ (p. 190) which implies a further departure from orthodox economic thought.

Typically, Weber confronts both economic orthodoxy and its socialist critics (pp. 78-80, 107-9). Prices, which in conventional theory are the result of the interplay of supply and demand, are seen as the ‘product of conflicts of interest (that) result from power constellations’ in ‘the struggle for economic existence.’ Consequently, money is not economic theory’s ‘neutral veil’ draped over exchange ratios of commodities. Rather, money ‘is primarily a weapon in this struggle, and prices are expressions of this struggle; they are instruments in this struggle only as estimated quantifications of relative chances in this struggle’ (p. 108).

The market may be a power struggle, but Weber offers no comfort to the socialists, who, following Marx, wished to remedy the inequality by issuing vouchers for an agreed ‘quantity of socially useful labour.’ But, in order to produce rational calculability, money has to be a weapon in the struggle for economic existence between ‘the play of interests oriented only to profitability’ (p. 183). The exchange of a socially agreed quantity of labour for specific goods would ‘follow the rules of barter’ (p. 80), and could not produce a measure of abstract value. Weber agreed with the Austrian theoretical economists in the ‘socialist calculation’ debate that money can never be a ‘harmless “voucher”’ as its valuation is ‘always in very complex ways dependent on its scarcity’ (p. 79). Any equilibrium or price stability in an equation of quantities of money and goods, in particular the interest rate, will be the expression of a predictable balance of power. Conversely, in this admittedly incomplete formulation, price instability in general is as much the result of the ‘economic struggle for existence’ as it is the product of an overabundance (inflation) or scarcity (deflation) of money. In short, socialism could not produce rational monetary calculation. Bureaucratic administration could never produce the ‘“right” volume or the “right” type of money’ because state bureaucracies are ‘primarily oriented to the creation of purchasing power for certain interest groups’ (including the state itself)—which would cause inflation (p. 183).

Simmel and Weber saw clearly the merits of the ‘nominalist,’ ‘state’ and ‘credit’ theories of money. These provide a foundation for a more comprehensive sociological theory of money as a social institution.

Fundamentals of a Sociology of Money

Attention should focus on three questions. What is money? How is money produced? How does money obtain, retain, or lose its value?

What is Money?

Economic theory’s focus on money as an actual medium of exchange entails a ‘category error’ in which specific forms of money have been mistaken for the generic quality of ‘moneyness.’ This has resulted in long-standing confusion over closely related issues—for example, the distinction between money and credit, the so-called ‘dematerialization’ of money, the advent of virtual postmodern’ money (Leyshon and Thrift, 1997), and electronic money and the ‘end of money.’ (For a discussion of these and related issues, see Ingham, 2002.) The unique specification of money is as a measure of abstract value and ameans of storing and transporting this abstract value.

Monetary exchange consists in the calculation and exchange and transfer of debits and credits according to a money of account. Money cannot be created without the simultaneous creation of debt. For money to be money, it presupposes the existence of a debt measured in money of account elsewhere in the social system. The holder of money is owed goods.

[M] oney is only a claim upon society … The liquidation of every private obligation by money means that the community now assumes this obligation to the creditor … [M] etallic money is also a promise to pay and … differs from the cheque with respect to the size of the group which vouches for its being accepted. The common relationship that the owner of money and the seller have to a social group—the claim of the former to a service and the trust of the latter that this claim will be honoured—provides the sociological constellation in which money transactions, as distinct from barter are accomplished. (Simmel, 1978 [1907]: 177, 174-9)

Money is circulating debt, but perhaps the traditional metaphor of a ‘circulation’ is inappropriate. Rather, vast dense networks of overlapping and interconnected bilateral credit-debit relationships constitute money. This is more obvious in the case of the ‘clearing’ of debits and credits in a bank giro, where money-stuff does not actually flow from one account to another. But it applies equally to coins and notes, which might be referred to as ‘portable debt’ (Gardiner, 1993: 224). The essential point is that the debt is either transferable (bank giro) or portable (coin) because it is denominated in money of account. Money is constituted by the continuation of relations of credit-debit. In Marc Bloch’s counterintuitive observation, money would disappear if all debts were paid (Bloch, 1954).

This conceptualization becomes clearer with consideration of the multiplicity and dissociation of money ‘things’ in relation to the abstraction of money. The measure (money of account), means of payment for the unilateral discharge of debt and any media of exchange need not be integrated in single form, as in coinage: Cash, plastic cards, cheques, magnetic traces in computer disks, and so on. The point is clearly expressed in a study of money and national identity in early capitalism:

By the 1830s, then, Britons could at different times and places have understood gold sovereigns, banknotes, or bills of exchange as the privileged local representatives of the pound … the pound as an abstraction was constituted precisely by its capacity to assume these heterogeneous forms, since its existence as a national currency was determined by the mediations between them. (Rowlinson, 1999: 64-5)

How is Money Produced?

Different modes of the production of money may be identified. These consist in social relations between issuers, issuer and users, and the technological means available for the storage and transportation of abstract value—from clay tablets, to coins, to pen and paper, to magnetic traces and so on. However, the fundamental question concerns the ‘origins’ of money of account; that is to say, the abstract ‘idea’of money.’

Money of account