250 words reflection paper
75aki131
ContentServer3.pdf
REVIEW OF SOCIAL ECONOMY
Money is a Social Relation Geoffrey Ingham
University of Cambridge
Abstract Money continues to present theoretical and conceptual difficulties for mainstream neoclassical micro- and macroeconomics. In most treatments it is "neutral," as either: a symbolic "veil" over an underlying "real" economy; a special "commodity" with utility for the rational maximizing individual; or as part of a model of aggregate econotnic "variables." The methodological assumptions of these approa- ches cannot account for the fundamental characteristics of money as a measure of value and acceptable means of payment. Many of these diffculties are resolved if money is conceptualized as a structure of social relations.
Keywords: money, credit, methodology, economics and social theory
INTRODUCTION There is no denying that views on money are as difficult to describe as are shifting clouds.
(Schumpeter 1994 (1954): 289)'
Money is a puzzle. The standard answer to the question of what is money derives from the late nineteenth-century functionalist account:
1 In general, writers on money tend to express perplexity. Marx admitted his difficulties, but apparently was mistaken in the attribution to Gladstone of the view that "even love has not tumed more men into fools than has meditation on the nature of money" (see Ganssman 1988). Kindleberger begins a chapter on the nature of money with a long quote from Dombey and Son in which the former is defeated by the latter's question: "Papa, what's money?" (Kindleberger 1984: 19). Robertson thought it appropriate to use quotations from Through the Looking Glass at the head of each chapter of his Money. In concluding a recent book on money, the sociologist Dodd was reminded of St. Augustine's remark that "time" became more and not less elusive with close examination (Dodd 1994: 153).
Review of Social Economy Vol LIV No. 4 Winter 1996 ISSN 0034-6764 © 1996 The Association for Social Economics
REVIEW OF SOCIAL ECONOMY
money is what money does. Conventionally, it is a measure of value (or unit of account); a medium of exchange; a standard of deferred payment; and a store of value. However, there is, in fact, wide and at times quite vigorous disagreement as to what money does and which are the most important of the functions. Does the "thing" have to perform all the functions in order to be money? Is "credit" money? And so on. Further investigation reveals a quite surprising paradox: Money is one of the most important pieces of "social technology" ever developed, but as an object of study in its own right it is neglected by the dominant or mainstream traditions not only in modem economics but also in sociology.
Surveying the economic literature on the nature and functions of money and looking at the more prominent recent disputes on the "behavior" of money, one is struck by the extent to which modem neoclassical economics remains dominated by the conceptual apparatus of the "theoretical" school of economics that triumphed in the Methodenstreit (conflict over methods) around the tum of the century. At this juncture, historical and sociological perspectives on the "economic" were expunged from the core of the discipline (see for example: Schumpeter 1994 (1954); Machlup 1978; Swedberg 1987). Indeed, the most mathematically sophisticated and therefore most revered theories in neoclassical economics—broadly speaking, neo- Walrasian general equilibrium theories—have had great difficulty in actually finding a place for money in their schemes. "The most serious challenge that the existence of money poses to the theorist is this: the best model of the economy cannot find room for it." (Hahn 1982: 1). In these abstract models, money can, at best, only have the role of a pregiven numeraire—an accounting device in what is, essentially, a simple barter economy. A closely related approach in the broad neoclassical school has been to look on money as a rather special "commodity," but one which, like all other commodities, must possess some utility for the rational maximizing individual. (As we shall see, this basis for money has proved difficult to demonstrate with this methodological framework.) In general, money in mainstream neo- classicism money is, analytically, merely a "veil" or a neutral "lubri- cant" in a model of the economy, which is seen as comprising "real" factors (Schumpeterl994 (1954): 227).
I shall argue that the difficulty in accounting for the existence of money and its role in actually existing economic systems is the result
508
MONEY IS A SOCIAL RELATION
of the absence of anything resembling an adequate specification of its social structural conditions of existence. In "critical realism's" terms, mainstream economics does not possess an adequate ontological theory of money (Lawson 1994). These comments apply mainly to neoclas- sical microeconomics. But modem macroeconomics fares little better, largely because it too proceeds from neoclassical rational choice assumptions, and, furthermore, conceptualizes money simply as one of the statistically related aggregate economic "variables." (See Lawson's "critical realism" perpective on economic methodology [Lawson 1995]). However, the initial partial break with economic "classicism" in the first part of the century—most notably by Keynes—was, to some extent at least, an attempt to make sense of new, or more widespread forms of bank and state credit money in which, to some at any rate, the "social" basis of money was blatantly obvious. Over the past few years, a distinctive "post-Keynesian" macroeconomic school of monetary economics has been developed wbicb, in opposition to neoclassicism, bas taken up tbis aspect of Keynes's work and focuses on "credit- money" (Wray 1990; Smithin 1994; Rogers 1989; Moore, 1988). This work cannot be discussed in detail here; but, altbough it is undoubtedly an advance on orthodox thinking, post-Keynesian monetary theory is understandably embroiled in the problematics of academic economics in which the terms of discourse are partly dictated by tbe neoclassical and/or "monetarist" conceptualizations it opposes. Hence, the pre- occupation with "endogeneity" and debates between "horizontalists" and "verticalists" conducted in terms of the strength of statistical relations between "variables" (Cottrell 1994).^
Furthermore, there is a second aspect to the paradox in that although money should be seen as having "social" conditions of existence, sociology, with the notable exceptions of Weber and Simmel, has contributed very little directly to the study of the actual social production of money as a system of social relations, sui generis. This contentious argument will be pursued in detail elsewhere; here we may briefiy note the basis for this judgment. In the division of intellectual labor between economics and sociology that developed in the first part
2 See Ingham (1996a) for a brief discussion of the distinction between the economic "aggregates" and "variables" of macroeconomics and social structure seen as an "emergent" property. I hope to address modem macroeconomic and post-Keynesian concepts of money, in more detail, in the future.
509
REVIEW OF SOCIAL ECONOMY
of the century money was placed under the jurisdiction of economics (Swedberg 1987; Ingham 1996a). Moreover, partly as a consequence of this division, the major schools of sociology have tacitly incorporated the orthodox conception of money from "real" neoclassical analysis. In Parsons' social theory, for example, money is taken to symbolize underlying social relations; it is a generalized medium of social interaction (Parsons 1937; 1951). (See the excellent discussions by Ganssman [1988] and Dodd [1994].)Although Marx stressed the fundamental monetary structure of capitalism (M-C-M), he was primarily concemed with showing that money was a "mask" (or "veil") over the underlying "real" social relations of the production of commodities.^ Institutions for the social production of money, such as banks, are not seen as the essence of capitalism; rather, this is sought in the reality of the capital-labour relations of production expressed in an alienated monetary form. (To [mis]use Marx's terminology, I would suggest, rather, that money should be seen as a "force of production" which, as the post-Keynesians correctly imply, has its own particular "relations of production.'"*)
Obviously, money is socially produced in the sense that it does not occur naturally, and it also mediates and symbolizes social relations— for example, capital-wage labor. However, I wish to go further and argue that money itself is a social relation. By this I mean that "money" can only be sensibly seen as being constituted by social relations. I have already hinted that this claim is most obviously sustained in the case of credit-money as "promises" to pay; but I shall argue that all forms of money are social relations and consequently, for example, the conven- tional textbook distinction between "money" and "credit" is not merely anachronistic, but is based on a conceptual confusion.
3 For Marx there are two veils: behind money lie "real" economic "forces" and, in tum, behind these lie the "real" social relations that again, in tum, appear as monetary relations.
4 The tendency to analyze economies almost exclusively in terms of "production relations" is pervasive within sociology. See, for example, Mann's attempt to write a "total" sociological history from the standpoint of the gradual extension of "infra- structural" and "despotic" "social power." He devotes considerable attention to state finance, but nowhere is money as "social power" sui generis given any specific treatment. "Economic power" and "economic organisation" are conceptualized entirely in terms of production and exchange (Mann 1986: 25).
510
MONEY IS A SOCIAL RELATION
MONEY IN ORTHODOX "REAL" ECONOMIC ANALYSIS
Economic thought is marked by a long continuity of interrelated disputes which involve surprisingly divergent conceptions of the nature of money. The major differences have been between "metalists" and anti-metalists during the sixteenth and seventeenth centuries (Schump- eter 1994 [1954]); the "Currency" and "Banking" schools and more generally between "materialists and "nominalists" in the first half of the nineteenth century; and the seesaw battle between "monetarism" and various forms of Keynesian economics in the middle of the twentieth.
The two sides' differences with regard to the nature of money and how to "manage" it are grounded in different conceptions of the scope and method of economics, which Schumpeter has referred to as "real" and "monetary" analysis. This was referred to earlier, but we now need to examine this distinction in a little more detail.
Real analysis proceeds from the principle that all the essential phenom- ena of economic life are capable of being described in terms of goods and services, of decisions about them, and of relations between them. Money enters the picture only in the modest role of a technical device that has been adopted in order to facilitate transactions . . . so long as it functions normally, it does not affect the economic process, which behaves in the same way as it would in a barter economy: this is essentially what the concept of Neutral Money implies. Thus, money has been called a "garb" or "veil" of the things that really matter Not only can it be discarded whenever we are analyzing the fundamental features of the economic process but it must be discarded just as a veil must be drawn aside if we are to see the face behind it. Accordingly, money prices must give way to the exchange ratios between the commodities that are the really important thing "behind" money prices . . . saving and investment must be interpreted to mean saving of some real factors of production . . . such as buildings, machines, raw materials; and, though "in the form of money," it is these physical capital goods that are "really" lent when an industrial borrower arranges for a loan.
(Schumpeter 1994: 277)^
From the present standpoint two closely related features of this "real" approach to money are most important. First, the almost exclusive concem is with money's medium of exchange function as a neutral
5 The term "real," as Schumpeter notes, is not very felicitous. Apart from the confusion, say, with "real" and "nominal" prices, "real" analysis is distinctly "unreal," as critical "realism" would argue, from both historical and sociological standpoints!
511
REVIEW OF SOCIAL ECONOMY
"lubricant" of exchanges between "real" "goods" or economic "forces" or "variables." Money as a measure of value/unit of account is taken as given and means of payment is assumed to be covered by the medium of exchange function. Second, "real" analysis is radically ahistorical and asocial in the sense that particular forms of economic organization are deemed to be epiphenomenal or merely "contextual"; complex social structure—banks, productive enterprise, etc.—is reduced to purely abstract exchange relations between rational maximizing agents.^ For example, one of this century's most influential economists would have us believe that:
even the most advanced industrial economies, if we strip exchange down to its bare essentials and peel off the obscuring layer of money, we find trade between individuals and nations largely boils down to barter.
(Samueison 1973: 55. Quoted in Wray 1990: 3)''
The Search for the "Micro-foundations" of Money
"Real" monetary analysis derives from "metallist" or "commodity" theory, which argued that the function of the medium of exchange was dependent on its being a commodity with an exchange value independ- ent of its form as currency. Thus, there could be an exchange ratio between the "real" values of precious metals (or wheat or beans) in the form of money and other commodities. With the increasing use of "worthless" token and credit-money in modem economies (i.e. non- commodity money), the basic logical structure of "real" analysis was preserved in two ways. First, paper and base-metal could stand for and be convertible into a "standard" of gold or silver. This formulation largely govemed monetary policy to the early twentieth century and basically involved schemes which tried to make paper behave as if it were gold.
A second and more narrowly academic response in the wake of
6 This position on the relation between the "economic" and the "social" or "historical" has recently been expressed succinctly by Oliver Williamson in a contribution to The Handbook of Sociology (Smelser and Swedberg, eds., 1994). He approvingly cites legal philosophy's distinction between the "essentials" of the "rational core" and the "tosh"—that is, "superfluous rituals, rules of procedure without clear purpose [and] needless precautions preserved through habit" (97-98).
7 Note also that Samueison is not saying that it is heuristically useful to look on advanced industrial economies in this way, but that they actually "boil down" to simple barter.
512
MONEY IS A SOCIAL RELATION
"worthless" paper money argued that money was really no more than a "token" or "symbol" of "real" goods. This line of reasoning found its most brilliant expression in Walras's system in which money is reduced to pure number—the numeraire—which enables the multilateral exchange of goods in instantaneously clearing markets in a world of virtual barter. As has frequently been noted, the presence of money makes no difference to the logical structure of such general equilibrium models. Abstract neo-Walrasian models are intentionally "timeless" and as such do not imply or infer a history of monetary evolution. Money is not accounted for, rather it is simply introduced along with the "auctioneer" in order to render the system operational.^
However, neoclassical economics in the English marginalist tradi- tion did provide a rudimentary theory of money's "origins." Money evolved to overcome the inconveniences of barter which arise from the absence of a "double coincidence of wants."^ The crudest formulations imply a teleological functionalism which Menger, for one, tried to avoid by arguing that rational transactors in a barter economy would realize the individual advantage in holding stocks of the most saleable commodity as a medium of exchange (Menger 1892).'° Thus, as an
8 For example, one of Clower's early essays attempted to deal with this lacunae from within the neoclassical tradition and provide a pure theory of a money economy. However, the approach only serves to illustrate the impasse created by the neglect of history. By means of a purely formal analysis, Clower can only conclude that: "A commodity is regarded as money for our purposes if and only if it can be traded for all other commodities in the economy. Correspondingly, a money economy is one in which not all commodities are money . . . money buys goods and goods buy money; but goods do not buy goods" (Clower, 1967, in 1984: 86). In short, the approach can proceed no further than providing a purely logical description of money as an "object."
9 Ever since Jevons's account in the late nineteenth century nearly all economic textbooks point out that barter requires a double coincidence of wants—that is, A has ducks but wants chickens at the same time that B has chickens but wants ducks. Not only does this approach tend to confuse, in Schumpeter's terms, "logical" and "historical" conception of "origins'; it is also empirically suspect on two counts. First, there is considerable evidence to suggest that such barter economies never existed in the form assumed in "real" analysis. (Heinsohn and Steiger 1989). Second, barter has occupied a wider role than is implied in evolutionary-efficency model. Historically, both barter and monetary exchange increased with the production of "surpluses"; and, moreover, barter need not be as inefficient as is commonly thought. (Braudel 1985; Melitz 1974).
10 Functionalism is bedeviled by logical problems. Teleogy involves treating "effects" (e.g. the systemic benefit of a multilateral exchange of heterogeneous goods) as "causes" of origin; but there is also the difficulty of functional alternatives. Functionalism is unable logically to specify which items might fulfill any specific
513
REVIEW OF SOCIAL ECONOMY
unintended consequence of individual rationality, precious metals, which in addition possess the convenient attributes of portability, divisibility, durability etc., become money.
Unfortunately, the demise of commodity-money broke this explana- tory link between individual rationality and collective or system benefits. Hence Menger's paradox that institutions such as money "make for the common interest, and y e t . . . conflict with the nearest and immediate interests of contracting individuals" in that an "individual should 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). Modern neoclassicism's attempted resolution, which is consistent with the axiom of rational maximization, has been to update Menger and to assert that money reduces transactions costs for the individual (Jones 1976; Ostroy and Starr 1974; Clower 1984). As we shall see in a moment from the implications of Hahn's analysis, the very best that this axiomatic-deductive method can demonstrate is that, once in existence and widely accepted, money can be an "individual" as well as a "public" good.
"Money may slip through our fingers," Frank Hahn laments, "unless its role in transactions is made essential" (Hahn 1987: 42)." His approach is instructive for two reasons. In the first place, the clarity of the exposition throws a sharp light on the disabling consequences of the "primitive" terms of microeconomics. Second, the rigor of the analysis leads Hahn to consider abandoning, or at least seriously modifying, these assumptions.
By modeling a money economy as a Nash equilibrium exchange game, Hahn shows that it is indeed "advantageous for any given agent to mediate his transactions by money provided that all other agents do likewise" (Hahn 1987: 26). Hardly a startling conclusion as he acknowledges, but at least it presents a formal proof that money's liquidity exists as a social convention. Surely, however, it is not so
function; for example, Samuelson's analysis of the "function" of money as a intergenerational store of value does not explain why money rather than another "store" or "asset" performs in this way (Samuelson 1966 [1958]). In short, the theory cannot adequately account for money's conditions of existence.
11 Hahn decisively dismisses the "theoretical history" which explains money's appearance by its superior efficiency over barter: "This Panglossian teleology lacks all merit until a process is described which brings about the superior outcome" (Hahn 1987: 24).
514
MONEY IS A SOCIAL RELATION
much a question of whether it is advantageous to use money if others do, hut rather that agents cannot use money unless others do likewise. To state the ohvious: the advantage of money presupposes a monetary system. We have here a typical example of the way in which the rational choice explanations of neoclassical economics soon become locked into slightly ahsurd circularities. Money is an advantage to the individual only if others use it; but, according to the theory, they can only rationally use it if it can be shown to he an individual advantage.
Significantly, Hahn admits that his purely deductive approach cannot identify money as such: "a barter economy might be a Nash equilibrium as well" (Hahn 1987: 29). As he acknowledges, if barter were "the accepted mode of transaction," it would be risky and therefore disadvantageous to accept a promissory note. Of course, it is simple to show that, in certain historical circumstances, the risk of accepting a promissory note that was embedded in a trustworthy network of banking families would be far less than transporting gold. But this is outside the methodological micro-foundations of economic analysis. Having pushed the spare principles to their limits, Hahn is left to ponder that "it is open to argument how far the given transactions institutions . . . are part of the basic description of the economy" (Hahn 1987: 42). Armed with the set of axioms and assumptions he employs, one could never explain why a rationally advantageous monetary system, includ- ing credit, banks, and so on had not existed for all time.
The explicit aim of neoclassical "real" analysis of money is to establish the logical preconditions for money's existence. However, it displays overwhelming theoretical deficiencies which, as a con- sequence of the methodological neglect of "historical" and "social" structure, are not remediable within the framework of its assumptions. In the first place, as I have already stressed, neoclassicism is mostly concemed with money as a cost-reducing medium of exchange for the individual. There is no attempt to account for the "concept" of money as a measure of value (or unit of account)—or even to recognize that this might constitute an intellectual problem. Second, as Hahn inad- vertently shows, money's general acceptability (or its liquidity) cannot be explained in terms of individual rational maximizing; this form of theorizing becomes stuck and cannot proceed because it must pre- suppose that which it sets out to explain. Finally, the preoccupation with the medium of exchange in the hypostatic model of barter exchange necessarily involves the "neutrality" assertion and the neglect of those
515
REVIEW OF SOCIAL ECONOMY
monetary forms which are not only characteristic but also constitutive of capitalism's production of credit money.
Monetary systems are the result of the long term historical develop- ment of a complex structure of social relations and practices which cannot be grasped by of neoclassicism's methodology. In this respect, Smithin has observed that "the micro-foundations of standard monetary theory have been left extremely weak" (Smithin 1994: 14). In fact, we need to go further: money cannot have "micro-foundations" if these are sought exclusively in the formal deductive model of the individual agent's rational choice of holding a "veil" or "lubricant" as simple medium in a "real" exchange economy.
MONEY AS A STRUCTURE OF SOCIAL RELATIONS
The dominance of neoclassical deductive methodology, in which money is a neutral veil, is the direct outcome of the triumph of the "theorists" during the Methodenstreit. However, the ultimately van- quished "historical school" in this struggle held to a theory of money's "origins" which differed significantly from the one propounded by their opponents.'^
The "historians" (including some sociologists) held to a «on-market theory of the origins of money and, in addition, stressed the essential role of the "state" (or "polity') in the reproduction of monetary systems.'^ In contrast to the theoretical economists' focus on the
12 For a brief references of the work of Roscher and Knies, see Einzig (1966). Despite his methodological critique of their work, Weber's analysis of money is heavily influenced by them and also Knapp (1924). Indeed, Menger's vehement and dogmatic insistence that money had only the one function as medium of exchange was probably as much a result of the intellectual enmity during the Methodenstreit as it was of scholarly conviction.
13 The state has no conceptual place in the "real" analysis of money; the theoretical "primitives" of this tradition do not incorporate social structure beyond the most elemental pure abstract exchange between utility-maximizing agents. More recent neoclassically based analysis has attempted to explain the state's existence as a response to "market failures" such as "extemalities," in the provision of "public" and "merit" goods, and in securing "property rights." But the state is accorded only a secondary or reactive role and older "real" theories of money are readily accomodated in this framework. In a Mengedan account, for example, the market might yeild the most saleable commodity as the medium of exchange; but the state can economize on transactions costs by being a more effective and trusted guarantor of that commodity's quality (Hodgson 1992).
516
MONEY IS A SOCIAL RELATION
medium of exchange function, they sought the origins of money as a measure of value/money of account and (unilateral) means of payment of fines, tithes, taxes, compensations—that is, debts between the political community and its members (See Schumpeter 1994, chapter 4; Einzig 1966; Melitz 1974).
The expression of this approach in Knapp's State Theory of Money influenced Keynes's study of ancient near eastern currency and the methodology of A Treatise on Money (Knapp 1924; Keynes 1930). Here Keynes opens with unequivocal assertion that "Money of Account, namely that in which Debts and Prices and General Purchas- ing Power are expressed is the primary concept of Theory of Money." In what is probably an oblique reference to Menger's insistence that money has only the one function—as a medium of exchange, Keynes dismissively observes that: "Something which is merely used as convenient medium of exchange on the spot may approach to being Money But if this is all, we have scarcely emerged from the stage of Barter" (Keynes 1930: 3). In this possibly unwitting endorsement of the German historical school approach, Keynes argued that the "state or community" was the source of money of account (measure of value) and means of payment (Keynes 1930: 4-5, 11-15).
A related divergence from the conception of money simply as a neutral medium of exchange is to be found in the work of those who, like Keynes, understood bank money as "book money"—that is, an abstract money of account for recording debt. As I have indicated, post- Keynesian monetary theory focuses on the supposed distinctiveness of credit money's non-neutrality; but from the present standpoint I wish to emphasize the fact that the symbolic representation, or signifiers, of these debt-relations eventually became accepted means of payment and media of exchange. In other words, a particular social relationship— promise to pay—^became money.
Thus, I intend to follow the "historical school," Keynes (and the later Hicks) and see money primarily a measure of value and a credit relation, and, furthermore, to argue that these attributes are to be seen as the structural properties of society. '"*
14 Given Hicks's initial reaction to Keynes, it perhaps a little ironic that he should reach exactly the same conclusion in the posthumously published A Market Theory of Money. After arguing that debt contract rather than spot payment was the most typical
517
REVIEW OF SOCIAL ECONOMY
1. The Social Bases of Money as a Measure of Value: State and Society
Although the German historical school paid more explicit attention to money as a measure of value than the "theorists," its origins were not fully explained beyond pointing to the need for the precise measure- ment of "debts" of fines, tithes, taxes, etc. In this respect, although we might agree with Keynes's metaphor that the state or community claims not only to enforce but also to write the monetary "dictionary" (Keynes 1930: 5), it should be pointed out that both activities presuppose the existence of a "language" (Frankel 1977).'^ Independently of any economic analysis of money, the numismatist Grierson similarly decided "on the test of money being a measure of value. Unless the commodities used for exchange bear some fixed relation to a standard, we are still dealing with barter" (Grierson 1977: 16). "The parties in barter-exchange are comparing their individual and immediate needs, not values in the abstract" (Grierson 1977: 19 emphasis added).
But Grierson goes further and attempts to explain the origins of the concept of money as a measure of value:
Behind the specific phenomenon of coin there is the phenomenon of money, the origins of which are not to be sought in the market but in a
economic transaction and discussing the standard functions of money, he concluded: "We seem thus to be left with two distinguishing functions of money: standard of value and medium of payment. . . . [and] money as a means of payment implies money as a standard." (Hicks 1989: 43). However, even though it appears that Hicks, in his very last work, was moving to a "credit theory of money" (as opposed to a commodity- money theory of credit), he felt it necessary to take Keynes to task for defining money by its perfect "liquidity." Hicks's partial retention of the older style of theorizing is evident: "liquidity cannot be defined . . . except in terms of exchangeability for money. So to define money as an asset with perfect liquidity is to argue in a circle. It is the other functions of money which are intrinsic; the liquidity property follows from them." (Hicks 1989: 42). Obviously, money performs "functions," but as we have seen their existence cannot be derived simply from their individual rationality or utility and that this applies especially to measure of value/money of account as an intersubjective, that is, social institution.
15 There is a long tradition of using a money/language analogy in economics and sociology The former is traceble to Turgot and more recently is used by Tobin and Hahn who seem to imply that to indicate that they are aware that money is "social" is all that is necessary to escape some of the restrictions of economics "methodological individualism." This analogy obscures as much as it illuminates, but the argument cannot be pursued here.
518
MONEY IS A SOCIAL RELATION
much earlier stage in cotnmunal development, when worth and wergeld were interchangeable terms.
(Grierson 1977: 33)
Wergeld ("worthpayment") was widespread in premarket societies and comprised the scales and tariffs of compensation for injuries used as an alternatives to socially and economically debilitating blood feuds and lex talionsis (Grierson 1977: 28).
It cannot be shown that wergeld actually predates the use of media of exchange in more narrowly market transactions and Grierson accepts that the "generalized application of monetary values to commodities could scarcely have come about before the appearance of market economies." But Grierson is concerned with what Schumpeter refers to as "logical origins" and argues that the idea of "moneysworth" could not have been produced by the "market."
The conditions under which these laws were put together would appear to satisfy much better than the market mechanism, the prerequisites for the establishment of a monetary system. The tariffs for damages were established in public assemblies, and Since what is laid down consists of evaluations of injuries, not evaluation of commodities, the conceptual difficulty of devising a common measure for appraising unrelated objects is avoided.
(Grierson 1977: 20-21)'^
Grierson makes no use of sociological theory, but his argument could be expressed as a Durkheimian hypothesis that money as a measure of value is a "collective representation" for which the analogue is the structure of society. In taking this lead, we can develop our argument by noting that wergeld expressed two meanings of "worth" which derive from two elements of basic societal structure: the utilitarian and the moral.
It is possible, but by no means easy, to construct reasonably objective indemnity schemes for functional impairment caused by loss or injury to individuals, social groups, or society at large.''' However, society's normative or moral order is inextricably bound up with notions of
16 There is considerable etymological support for this theory. In addition to Grierson, see Einzig [1966]: "the word Geld is said to have originated from Vergeltung which . . . implies the settling of scores or revenge." [379]; also Simmel (1978 (1907), 357) notes that shilling is derived from "skillan," meaning killing or wounding.
17 In their very important critique of neoclassical analysis of money, Heinsohn and Stieger (1989) follow Keynes in linking money of account to contracts and, thereby, argue that money has its origins in the institution of private property without which
519
REVIEW OF SOCIAL ECONOMY
functional "worth" and, in this respect, wergeld was a matter of both injury and insult. Payment for killing a king, for example, was set at an absurdly high level that would have involved the selling into slavery of the murderer's whole extended family. On a more trivial level, it "cost four times as much to deprive a Russian of his moustache or beard as to cut off one of his fingers" (Grierson 1977: 20). Here, loss of face was literally more important than at least some small loss of function. Moreover, it is quite clear tbat tbe expiation of tbe culprit was not intended to compensate precisely for destroyed functional "value," but also involved punishment or vengeance for tbe transgression of tbe symbolic values of tbe "sacred" social realm. Tbe defense of sucb values migbt bave some utilitarian or functionally instrumental con- sequences; but it is difficult to explain tbem entirely in tbese terms (Elster 1990: 238). Ratber, tbe meaning of, say, insults and revenge must be sougbt in society's system of values wbicb, following Durkbeim, is a moral order sui generis. Of course, tbese arguments are simply a restatement tbe familiar sociological critique of the idea of tbe primordial "market"; tbat is, tbe tbeory of society based upon tbe advantaged contractual interdependence of rationally calculating maxi- mizing individuals.'^
Tbis is not tbe place to plow tbis already well-worked ground, but perbaps it migbt be useful to illustrate tbe general argument, as I bave applied it to tbe "origins" of money as a measure of value, by looking at an analysis tbat proceeds along identical lines in its explanation of tbe "logical origins" of inequality. Over tbirty years ago, Dabrendorf
there cannot be contracts (see also Wray 1990). However, as Keynes, Hicks, and Grierson argue, the measure of value is anterior to contract.
18 Sociology's development owed much to its rejection of the assumption to be found in "real" economic analysis that there exists "natural" substratum of social and economic order that is in "in the nature of things" (Parsons 1937). Much of Durkheim' s writing, for example, is devoted to an elaboration of this critique. Most notably he argued that economic institutions, which of course include "money," could not be founded entirely on "contracts" which supposedly expressed these natural interests. Durkheim insisted that the very idea of a contract presupposed an anterior social or moral order: "all in the contract is not contractual...wherever a contract exists, it is dependent on regulation which is the work of society not that of individuals" (Durkheim 1964: 189, 194). Durkheim was not merely suggesting that contracts would be unworkable without norms of custom and practice and a degree of trust, but also literally unthinkable without the social construction of the idea of contract and procedure that, furthermore, could not have been generated by the immediate parties themselves.
520
MONEY IS A SOCIAL RELATION
Started from Durkheim's insistence that every society is, fundamentally, a "moral community" in the sense that both social positions and the performances of their individual incumbents are evaluated in relation to systems of norms and values (Dahrendorf 1968). These are supported by both material and symbolic sanctions—that is, punishments (includ- ing "shame") for transgression; and rewards for exemplary conduct— especially "honors." The latter produces an inequality of status. Relatedly, wergeld symbolically represented society's two faces and involved, on the one hand, attempts to quantify the functional contribution of social positions by the imposition of tariffs for the impairment or loss of incumbents. On the other, such schemes, or collective representations, were the codification of the values and norms without which the evaluation of functional "worth" would have remained anomic; that is, an unresolvable contesting of claims and counterclaims backed by coercion. There are then very good reasons for believing that the very idea of money originated outside the "market," as this is conventionally construed in economics.
2. Credit money: Banks and the State
Early intellectual opposition to the commodity-theory of money was largely founded on efforts to understand the theory and practice of new forms of state money and bank money other than as the mere representation of precious-metal money.'^ Money's transformation
19 The existence of state-issued precious metal coinage struck with the image or emblem of the ruler, as quality guarantee, can quite easily be incorporate in orthodox "real" or neoclassical analysis (Hodgson 1992); but such treatments have tended to be at the expense of historical accuracy. While it is acceptable to present an economic analysis of the net benefits of the state's "public goods" role in the provision of commodity money, it is misleading even to imply that this provides an adequate explanation of the emergence of the state's involvement. Even a cursory glance at the history of "seignorage" and "debasement" suggests that the exploitation of political advantage was at the center of the struggle for the monopolization of issue and that the outcome was contingent and not simply determined by any "logic" of cost reduction for market transactors (Spufford 1988). Similarly, the brief references to the history of banking to be found in many economic textbooks merely perpetuates the misleading commodity-theory account of goldsmith's receipts as the origin of banknotes (see for example, Begg, Dombusch, and Fischer 1991: 404). To be sure, precious metals were deposited for safe-keeping and the receipts were often accepted as the direct representation of these commodities and thereby used as money; but this was probably the one of the least important antecedents of modem capitalist banking.
521
REVIEW OF SOCIAL ECONOMY
from, in Simmel's terms, "substance" (commodity) to pure "function" as an acknowledgment of debt (promise to pay) expressed in a money of account and issued by states and banks involved the realization that money was itself a social relation.
The equivocal position of such views in economic theory is apparent in the difficulties that bank credit money has posed for orthodoxy. Insofar as the actual activity of banks has been the subject of "real" or narrowly neoclassical analysis, it has been seen as simple inter- mediation between savers and borrowers and in the reduction of transactions costs through the clearing of checks etc. However, from the fifteenth century, banking practice increasingly involved the "manu- facture" of money as a social relation. Lending involved the creation of a deposit which stood in a relatively autonomous relationship to any approximate incoming "balance" of deposits from other sources. The bank "note" signified neither actual nor virtual goods and commodities, but simply a debt to the bank and promise of repayment. This difference has been expressed in a variety of ways; but perhaps most pithily in the distinction between the "real" analysis conception of banking practice in which "deposits make loans," and the converse "credit" or "monetary theory" view that "loans make deposits". The latter act of bank lending creates a deposit against which checks may be drawn (Schumpeter 1994: 717; also Rogers 1989; Wray 1990).
The production of state-credit money is in principle the same process as bank-credit money. Since some Italian cities decreed, in the early seventeenth century, that their promises to pay their debts were legal tender, states have used this means of borrowing and increasing the money supply (Kindleberger 1984; Arrighi 1995). The acceptability of these promises depended, among other things, on the credibility of promises to repay debt. This was achieved by the gradual extension of long-term control over state finances by the creditors through the replacement of arbitrary and relatively untrustworthy monarchical rule by bourgeois democracy and constitutionalism. In effect, the rulers of the Italian bourgeois city-states—Genoa, Florence, Venice— guaranteed each others' debts. For similar reasons, the state's promise to pay became money relatively early and successfully in the Nether- lands and England, but more unevenly and hesitantly elsewhere in Europe (Dickson 1967; Kindleberger 1984).
From the present standpoint two elements in this "leap" in the social development of money are particularly important. First, the growing
522
MONEY IS A SOCIAL RELATION
international bank "giro" gave more prominence to "the idea of a money of account which was defined by quantities of metal and which existed only as a bookkeeping device for facilitating large scale trade and finance in a world of numberless and ever changing currency systems." By the early seventeenth century, distinctions such as the one between moneta immaginaria and moneta reale were being made (Schumpeter 1994: 296-297).
Second, it was the detachment of money, by banks and states, from being merely a signifier of commodities or deposits of commodity money that proved to be difficult to accomodate within "real" analysis (Schumpeter 1994: 317-322).^° This mode of monetary production involved the creation of debt and the acceptance of "promises to pay" as media of exchange and means of payment. As promises, money is not a "commmodity" which stands in a relatively stable relationship to other commodities, nor is it merely a refiection, symbolic representa- tion, or signifier of an underlying existing "reality" of economic relations. Rather, it is a social relation based upon definite and particular social structural conditions of existence involving, among other things, an institutionalized banking practice and constitutional legitimacy of the political authority in which the promises of banks and the states to pay gradually became currency (Dickson 1967; Hicks 1967, 1989; Weber 1992). The fundamental general social structural change that underpinned this critical step was the establishment, in custom and law, of the fungibility (transferability) of debt (Atiyah 1979: 135-138; Weber 1992 (1927): chapter XX). This involved a
20 Credit devices that originally facilitated trade, such as bills of exchange, also became "detached" from any actual commodities that they might have directly represented and were issued by banks to serve as autonomous media of exchange and means of payment, tt should be noted, however, that "commercial-paper" as a medium of exchange that directly signifies actual goods is consistent with the idea in "real" analysis of money as a neutral "veil." The conception of a purely paper credit barter economy in which a numeraire represents "real" values is contained in the Walrasian model; and more recent theoretical developments, such as the "New Monetary Economics," examine the possibilty of large-scale "cashless" barter systems based on advanced computer technology which can match traders and produce the necesssary "double coincidence of wants" (Smithin 1994: 14-16). See also the "real bills doctrine" in relation to the Banking and Currency schools dispute (Smithin 1994: 4). However, in "real" analysis, the equation between signifier (bill) and object (commodity) is either assumed or considered exclusively as a logical relation. "Monetary" analysis points to the inherently problematic relationship that is the result of the relatively autonomous processes of the respective "social" production of money and goods.
523
REVIEW OF SOCIAL ECONOMY
second level of detachment in the transformation of social relations into "money." Not only was credit money separated from any direct relation to "real" commodities, but also from particularistic (person-to-person) debt relations. The transformation of personal into impersonal trust (legitimacy) and, thereby, of IOUs into liquid money, was a long-term historical process which, like the concept of measure of value, cannot be deduced from the axioms of orthodox theory.
Considered in this way, banking practice, including state and "central" banking, is social practice; that is, the social construction of credible rules of thumb that are considered prudent and legitimate by all concerned. It is not quite seventy years since Keynes attempted, in A Treatise on Money, to give a more secure theoretical foundation to the "creation" of bank money:
[I]t is evident that there is no limit to the amount of bank-money which the banks can safely create provided that they move forward in step. The words italicised are the clue to the behaviour of the system— Each Bank Chairman sitting in his parlour may regard himself as the passive instrument of "outside forces" over which he has no control; yet the "outside forces" may be nothing but himself and his fellow-chairmen, and certainly not his depositors.
(Keynes 1930: 26-27)^'
The theoretical resistance to this view of bank money continues in a great deal of economics and is almost as mysterious as the production of money itself.̂ ^ This subject would make a very interesting subject for the sociology of economic knowledge. It cannot be pursued here beyond the observation that the proximate reasons for the persistence of "real" monetary analysis must lie in the dominant paradigm's conceptualization of economic relations as those between individual agents and commodities or between economic "variables." But bank money is debt, which is a social relation. The theoretical explication, in economics, of the social process of the production of money remains
21 See the debate on the proposition that the money supply curve is horizontal at a given rate of interest (Moore 1988).
22 As late as 1921, no less an economist than Cannan, writing in the Economic Journal, used the "real" analysis analogy that if a cloakroom attendant loaned out bags left with him this would not involve the "creation" of bags and, moreover, before the owners of the bags could use them, they would have to be recovered from the borrowers (Schumpeter 1994: 1113-1114). But as Schumpeter explains, depositors and borrowers have simultaneous use of the "same" money.
524
MONEY IS A SOCIAL RELATION
underdeveloped for the simple reason that it would have drastic consquences for economic methodology as a whole (deCecco 1987: 1-9). In this respect, it is important to note that after its rehabilitation into neoclassicism, most Keynesian monetary analysis devolved into the study of the "portfolio" selection and "liquidity preference" of rational maximizers. That is to say, mainstream macroeconomics became almost exclusively focused on money's least specific attribute—as a store ^̂
3. Is All Money "Credit"?
Despite the fact that commodity money has long since ceased to have any significance, the distinction between "money" and "credit" persists in learned journals, textbooks, and everyday usage. The obvious empirical point that most money is now some form of credit money is not the one I wish to stress; rather, I would like briefly to argue that the reasoning behind the view that credit money is a social relation be extended to all money, including its archaic "commodity" forms. This general argument has been expressed most forcefully by the sociologist Simmel in his argument that the distinction between barter and money is sociological, not simply logical as in, for example, Clower's well- known formulation that money buys goods and goods buy money, but goods do not buy goods.̂ "* Consequently, Alfred Marshall's school- yard barter or swaps of nuts for apples, for example, is to be seen as an
23 This most "economic" of problems—that is, the value of money—is also, in part, a socially "enacted" or "accomplished" process involving, inter alia, economic theories of the "causes" of inflation; financial journalism; accountancy theory and practice; central banks; ministries of finance, etc. (see Mirowski 1990). In other words, the statistical relationships between the relevant macroeconomic "variables" involved in the determination of the "value" of money are the expression of underlying social and economic processes.
24 However, Smithin reports that shortly before his death. Hicks had abandoned any distinction between "money" and "credit" and maintained that "the evolution of money is better undertsood if one starts with credit" (Smithin 1994: 25 , original emphasis). The evolution of Hicks's thought on money from his early orthodoxy of the 1930s. ("Mr Keynes and the Classics," Econometrica, 5, 1937: 147-159) to the posthumous A Market Theory of Money (1989) and then the above statement displays, to this reader at least, a fascinating gradual movement to an inherently historical and sociological conception. This interpretation relies on some of his substantive arguments in this book rather than its explicit conclusions, with which they are frequently at odds in their adherence to the both the logical and historical primacy of "the market." See also Schumpeter (1994: 321).
525
REVIEW OF SOCIAL ECONOMY
essentially private transaction based entirely on the "subjective" preferences and wants of the two individuals involved. For Simmel, however, money introduces a third factor:
The pivotal point in the interaction of the two parties recedes from the direct line of contact between them, and moves to the relationship which each of them ... has with the economic community that accepts money .. [which]... is only a claim upon society.
(Simmel 1978: 177)
This mode of reasoning achieves two results. First, it draws attention to the fact that the "real" model of pure barter can only be i»jlateral exchange and cannot accurately refer to any historically real economic system of any complexity. Any move to multilaterality with heteroge- neous goods would involve money at least as a unit of account, with or without acceptable tokens. Second, the distinction between "money" and "credit" is dissolved—both are promises to ^̂
[M]etallic money is also a promise to pay and . . . it differs from the cheque only 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: 178)
Admittedly this does not take us very far and a large number of theoretical and empirical issues remain. And, it is not being suggested that distinctions between different forms of money should not be made; for example, commodity money, fiat money, promissory notes, checks, credit cards, local exchange trading scheme (L.E.T.S.) tokens etc. Each has its own particular conditions of existence; but all such conditions are essentially social, and the conventional economic distinction
25 Simmel continues: "This is the core of the truth that money is only a claim upon society. Money appears so to speak, as a bill of exchange from which the name of the drawee is lacking.... It has been argued against this theory that metallic money involves credit, that credit creates a liability, whereas metallic money payment liquidates any liability; but this argument overlooks the fact that liquidation of the individual's liability may still involve an obligation for the community. The liquidation of every private obligation by money means that the community now assumes this obligation to the creditor." (Simmel 1978: 177. See especially, 174-179). Compare Schumpeter's formulation that "money in turn is but a credit instrument" (1994: 320-321).
526
MONEY IS A SOCIAL RELATION
between "money" and "credit" can only obscure this simple but essential observation. Monetary relations are social relations and, consequently all forms of money have a fiduciary character (Dodd 1994).
CONCLUSION
The position taken here is similar in all important respects to the "critical realist" perpective in economics. Here it is argued that the essential methodological core, and fundamental error, of contemporary mainstream economics is its deductivist and/or empiricist methodology (Lawson 1994). This gives primary attention to the "surface" phenomena of, on the one hand, individual calculations of utility as expressed in "revealed preferences" and, on the other, to the statistical description of constant "actual" event conjunctions. In consequence, mainstream economics rarely concems itself with, and indeed cannot easily accommodate any role for, underlying social structures includ- ing, on one level, social relations. In this approach, "social relations" or "social structure" can be said to be conceptualized, at best, as the recognition that economic agents take each others' maximizing strate- gies into account, as in game theory. At the same time, mainstream economists widely observe, and puzzle over, the difficulty of establish- ing an indispensable role for money in their theorizing. My objective has been to demonstrate that money is itself a social relation in the sense that it cannot he adequately conceptualized other than as the emergent property of a configuration (or "structure") of social rela- tions.^^ Thus the inability of contemporary mainstream economics to accommodate money is a specific, and rather fundamental, manifesta- tion of the sorts of failings that "critical realist" and some sociological arguments have established more generally.
REFERENCES
Arrighi, G. (1994) The Long Twentieth Century, London: Verso. Atiyah, P. S. (1979) The Rise and Fall of Freedom of Contract, Oxford:
Clarendon Press.
26 For a discussion of the distinction between social "relations" and social "structure" in relation to economics, see Ingham 1996(a), 1996(b).
527
REVIEW OF SOCIAL ECONOMY
Begg, D., Fischer, S., and Dornbusch, R. (1991) Economics, 3rd edition. Maidenhead: McGraw Hill.
Ciower, R. W. (1984) "A reconsideration of the microfoundations of money," in D.A. Walker (ed.). Money and Markets: Essays by Robert W. Ciower, Cambridge: Cambridge University Press: 81-89.
Cottrell, A. (1994) "Post-Keynesian monetary economics," Cambridge Jour- nal of Economics, 18: 587-605.
deCecco, M. (1987) Changing Money: Financial Innovation in Developed Countries, Oxford: Blackwell.
Dickson, P. G. M. (1967) The Financial Revolution in England: A Study in the development of Public Credit, 1688-1756, New York: St. Martin's Press.
Dodd, N. (1994) The Sociology of Money, Cambridge: Polity Press. Durkheim, E. (1960 [1893]) The Division of Labor in Society, Glencoe, IL.:
Free Press. Einzig, P. (1966) Primitive Money, London: Pergamon Press. Frankel, H. S. (1977) Money: Two Philosophies, Oxford: Blackwell. Ganssman, H. (1988) "Money—a symbolically generalized medium of
communication? On the concept of money in recent sociology," Economy andSociety 17 (4): 285-315.
Grierson, P. (1977) The Origins of Money, London: University of London, The Athlone Press.
Hahn, F. (1982) Money and Infiation, Oxford: Blackwell. (1987) "The Foundations of Monetary Theory," in M. deCecco and J.
Fitoussi (eds.). Monetary Theory and Economic Institutions, London: Macmillan: 2 1 ^ 3 .
Heinsohn, G. and Steiger, O. (1989) "The veil of barter: the solution to the 'Task of Obtaining Representations of an Economy in which Money is Essential'," in J. A. Kregei (ed.) Inflation, Income Distribution and Capitalist Crisis, London: Macmillan: 175-204.
Hicks, J. (\9%9) A Market Theory of Money, Oxford: Clarendon Press. Hodgson, G. M. (1992) "Carl Menger's theory of the evolution of money:
some problems," Review of Political Economy, 4 (4): 396—412. Ingham, G. (1996 a) "Some recent changes in the relationship between
economics and sociology," Cambridge Journal of Economics, 20: 243-275.
(1996 b) "The New Economic Sociology," Work Employment and Society, 10 (3): 549-564.
Jones, R. A. (1976) "The Origin and Development of Media of Exchange," Journal of Political Economy, 84 (4): 757-775.
Keynes, J. M. (1930) i4 Treatise on Money, London: Macmillan. Kindleberger, C. (1984) A Financial History of Western Europe, London:
Allen and Unwin. Knapp, G. (1924 [1905]) The State Theory of Money, London: Macmillan. Lawson, A. (1994) "A Realist Theory For Economics," in R. E. Backhouse,
(ed.) New Directions in Economic Methodology, London: Routiedge: 257-285.
528
MONEY IS A SOCIAL RELATION
(1995) "The 'Lucas Critique': a Generalization," Cambridge Journal of Economics 19 (2): 257-276.
Machlup, F. (1978) The Methodology of Economics and Other Social Sciences, New York: Academic Press.
Mann, M. (1986) The Sources of Social Power, Cambridge: Cambridge University Press.
Melitz, J. (1974) Primitive and Modem Money, Reading, MA: Addison- Wesley.
Menger, K. (1892) "On the Origins of Money," Economic Journal 2: 239-255.
Mirowski, P. (1990) "Learning the meaning of the dollar: conservation principles and the social theory of value in economic theory," Social Research, 57:689-711.
Moore, B. J. (1988) Horizontalists and Verticalists: The Macroeconomics of Credit Money, Cambridge: Cambridge University Press.
Ostroy, J. M. and Starr, R. M. (1974) "Money and the Decentralization of Exchange," Econometrica, 42 (6): 1093-1113.
Parsons, T. (1937) The Structure of Social Action, New York: Free Press. (1951) The Social System, New York: Free Press.
Robertson, D. H. (1928) Money, London: Nisbet. Rogers, C. (1989) Money, Interest and Capital, Cambridge: Cambridge
University Press. Samueison, P. A. (1966 [1958]) "An Exact Consumption-Loan Model of
Interest With or Without The Social Contrivance of Money," in J. Stiglitz (ed.). The Collected Scientific Papers of Paul A. Samueison, Vol. I: 219-233. Cambridge, MA: MIT Press.
Schumpeter, J. (1994 [1954]) A History of Economic Analysis, London: Routiedge.
Simmel, G. (1978 [1907]) The Philosophy of Money, London: Routiedge. Smithin, J. (1994) Controversies in Monetary Economics: Ideas, Issues and
Policy, Aldershot: Edward Elgar. Spufford, P. (1988) Money and its use in Medieval Europe, Cambridge:
Cambridge University Press. Swedberg, R. (1987) "Economic sociology: past and present," Current
Sociology 35,1: 1-216. Weber, M (1981 [1927]) General Economic History, New Brunswick, NJ:
Transaction Publishers. Williamson, O. (1994) "Transaction Cost Economics and Organization
Theory," in N. J. Smelser and R. Swedberg (eds.) The Handbook of Economic Sociology, Princeton: Princeton University Press: 77-107.
Wray, R. L. (1990) Money and Credit in Capitalist Economies: The Endogenous Money Approach, Aldershot: Edward Elgar.
529
