- 1 For an in-depth presentation of this particular point, see Bridel 1997, 130-9, and more particularl (...)
1For anyone having read The Élements d’Économie Politique Pure and thought he or she had understood the nature of Walras’s failure at integrating money into the general equilibrium framework, Pascal Bridel’s Money and General Equilibrium Theory: From Walras to Pareto is a lesson of humility. As Steindl (1998) notes, this is a tightly written and carefully documented book. Bridel traces the origin of the plain fact that there is no room for money in general equilibrium models to Walras’s own inability to find such a place, starting from the very first edition of the Éléments, going through his Théorie de la Monnaie and all the way to the fourth and final edition, where Walras attempts to include money in the general equilibrium framework, by incorporating it into the utility and production functions. Bridel demonstrates that Walras’s failure has deep roots: it cannot be reduced to the sudden lack of rigour that characterizes Walras’s slipping into approximation, when he discusses the tâtonnement of the equations of circulation.1
2Bridel’s book is a step by step voyage throughout the writings of the University of Lausanne’s first professor to hold a chair of economics. It ends with a chapter devoted to the contributions of three of Walras’s immediate followers, those who took on the challenge to pursue his journey, namely Pareto, Schumpeter and Schlesinger. This chapter will be the starting point of our discussion of Schumpeter’s contribution on this subject.
- 2 “The work was read avidly in Vienna even long after the First World War, and its youthful freshness (...)
3Schumpeter does not emerge unscarred from Bridel’s scrutiny, although he finds pride of place among Walras’s continuators, and even if Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie (The Essence and Main Contents of Economic Theory, hereafter WH) is, in Bridel’s words, the key, be it a “modest” one, to the “spread of Walrasian theory into the German-speaking world” (Bridel, 1997, 160), to which Morgenstern’s obituary of Schumpeter bears witness (ibid., fn. 36, 173).2 This said, Bridel’s less than two pages devoted to WH ends with the statement that “Schumpeter’s strictly theoretical discussion displays no originality”. His argument is supported by the facts that not only does Schumpeter not even summarize the principle of encaisse désirée but also that he makes no attempt at building upon Walras’s monetary foundations. All that Schumpeter’s discussion would really contain is a “comparative static quantity theory”.
- 3 See Bridel (1997, 5) for Walras’s more than probable inspiration from Isnard (1981) in the espousal (...)
4 However, Bridel does recognize the presence of two important “methodological distinctions” in WH. One is the differentiation between the numéraire and money, the other is the derivation of monetary theory from price theory. On the question of the numéraire, Bridel quotes Schumpeter’s covering letter that accompanied the copy of WH he sent to Walras, where he expressed his veneration and intention to continue the master’s work in the direction he had opened. Interestingly, Walras’s copy of WH contains an annotation in the margin where Walras translates Wertmaßstab by “numéraire” and Tauschmittel by “money”, a “crucial (and at the time still novel) distinction” (ibid., fn. 38, 174).3 On the second issue, Bridel underscores that this is the first time this is accomplished outside Lausanne. Taken literally, this statement might downplay the contribution of Carl Menger, whose influence on Schumpeter’s professors in Vienna (notably Böhm-Bawerk and Wieser) is certainly one of the motives for writing WH, whereas there is definitely a link between price theory and monetary theory in Menger’s works.
5 These comments will guide our own assessment of Schumpeter’s contribution to monetary theory in WH. For anyone wishing to understand some aspects of Schumpeter’s thought, WH is an important book, if only because it is the one where he puts in writing his vision of methodological individualism. Schumpeter was 25 when it was published and it is a book infused with enthusiasm but also sometimes with awkwardness. On the other hand, it reveals an already productive scholar, able to handle huge amounts of literature and to set in words an ever prolific and meandering thought.
6 The paper is structured as follows. Section 2 briefly reviews Schumpeter’s purpose in writing WH, as well as the methodological aspects it contains, which result in an idiosyncratic interchangeability between the phrases “value function” and “marginal utility function”. Section 3 examines the relation that Schumpeter establishes between price theory and money. We show that indirect exchange is the pathway that Schumpeter takes to introduce money in the general equilibrium theory, in a way that owes nothing little if anything to either Menger or Walras. Section 4 highlights the surprising fact that Schumpeter almost had a go at the money-in-the-utility-function theory. The final section contains a few concluding remarks, whose purpose is to situate Schumpeter’s chapter on monetary theory in WH within his life’s work on money, credit and banking.
- 4 Schumpeter (1908, v). The author’s own words are: “Ein guter Sinn liegt in dem geflügelte Worte: Al (...)
To understand is to forgive. A better way of saying this is: He who understands sees there is nothing to forgive. And this also applies to the area of knowledge.4
- 5 There is a translation, by Bruce A. McDaniel, but it can hardly be recommended. See McCaffrey (2011 (...)
7Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie is Schumpeter’s Habilitation thesis, published in 1908. Unfortunately, no utilizable version of it exists in English,5 whereas there is a Japanese translation, as attested by Kesting (2008, 78), as well as an invaluable Italian translation by Giuseppe Calzoni, available since 1982. As already mentioned, WH is the work of a young scholar. This circumstance explains the energy but sometimes also the maladroitness that show through its 626 pages. This drawback contrasts with the erudition, the capacity to elaborate at length on abstract and methodological issues, thus the stunning maturity that such an exercise requires and that also transpires.
- 6 The hostilities were launched by Menger in Untersuchungen über die Methode der Sozialwissenschaften (...)
8 Schumpeter’s purpose in writing the book was twofold: first, to expound “what we should think of pure economics”; second, to “familiarize the German audience with economic theory” (Schumpeter 1908, xvii and xviii). In other words, Schumpeter’s second goal was to “solve” the Methodenstreit (battle of methods) – which the emerging Austrian school, with Carl Menger at its head,6 had been waging with the German Historical School since 1883 – in a way that was “both clever and innovative”, i.e. by suggesting “a peaceful coexistence of the two approaches” (Kesting, 2008) and thus of their distinct methods, as both had a contribution to make and each should acknowledge the other. Hence the opening statement of WH, quoted above as an epigraph to this section.
- 7 See, for instance, Graça Moura (2002 and 2003).
9 This historical circumstance raises the question as to what might be Mengerian about WH. Among other aspects, this book is where we find Schumpeter’s first exposition of his methodology. Although it has been recently challenged,7 we will go by Shionoya’s (1997) analysis of Schumpeter’s instrumentalist methodology, which is patently present in WH’s Chapter 6 on “Methodological Individualism”. This is the only chapter available in English. In the preface of the translation, Hayek describes it as “still entirely Mengerian”, a view that soon appears disputable to the reader of Schumpeter’s text.
10 First, Schumpeter’s individualism is purely methodological and it precludes any commitment to political individualism:
[T]here is not the slightest connection between individualistic science and political individualism. … We are … fully in agreement with any objection that is raised against the misuse of the theory to defend indifference to social misery. However, it would be wrong to reject the theory for this reason alone. (Schumpeter 1980, 3)
11Second, Schumpeter adopts this methodological approach for what he considers practical reasons.
[Methodological individualism] has no specific proposition and no prerequisites, it just means that it bases certain economic processes on the actions of individuals. Therefore the question really is: is it practical to use the individual as a basis. ... The question is purely methodological... (ibid.)
12Third, methodological individualism is what is at work in his brand of “pure economics”, whose goal is not to produce truth, but to generate results. Hence, Schumpeter’s “instrumentalism”:
[W]hat economics really is—whether the individual is the driving force or whether something else is—is unimportant. … What counts is not how things really are, but how we put them into a model or pattern to serve our purpose as best as possible. In other words which concept is the most practical format from the viewpoint of the results of pure economy. (ibid., 5)
13Finally, it is significant that Menger is not explicitly referred to in WH, setting aside the Preface, where his name is stated as one of the three initiators of the revolution in pure economics, alongside Walras and Jevons.
14 The methodological choices made by Schumpeter are even more apparent in the fifth and final part of WH. This is where Schumpeter defines the “nature” or “essence” of “pure economics”, also referred to as “exact” or “static”, the three being often used interchangeably, depending on the context. For Schumpeter, “pure economics” is an abstract construction, a theoretical schema for the description of economic facts. It is based on hypotheses that are chosen by the theorist and its goal is not to produce truth, but to bring out the “essence” of economic facts, that is, to simplify the exposition of what would be otherwise irremediably complex.
- 8 Schumpeter (1934). First German edition: 1911.
- 9 “Damit Wert entstehe, muß sich zur Nützlichkeit auch Seltenheit gesellen” (Schumpeter, 1908, 100).
- 10 For an analysis of the openness of Menger, Wieser and Böhm-Bawerk to the inclusion within the realm (...)
15 Part Five of WH is also where Schumpeter strongly underlines the limits of pure economics, its incapacity to deal with issues that belong undeniably to the realm of economics, such as interest and entrepreneurial profit. Actually, Schumpeter states economic development, or dynamics, as a much more important issue than statics, repeatedly in WH, three years before the publication of the Theory of Economic Development.8 In Part Five, he also specifies the relations between pure economics and other sciences, such as biology, sociology, ethnology, and psychology. About psychology, Schumpeter clearly states that there are no psychological foundations to the theory of value. From this point of view he distances himself from the rising Austrian school. Accordingly, earlier in WH, this point is elaborated at length in “The Concept of Value”, the chapter where Schumpeter affirms that he considers “utility” and “value” as absolutely synonymous. This has two consequences. Firstly, he parts company with those he calls the psychological theoreticians, among them Böhm-Bawerk, whom he quotes for stating: “For the emergence of value there must be scarcity as well as usefulness”.9 This statement could have been taken from Menger, who had however retired by the time Schumpeter attended lectures at the University of Vienna. Menger, we know, distinguished between economic and non-economic goods, the latter being those which “not only do not have exchange value, as has previously been supposed in the literature of our subject, but no value at all, and hence no use value” (Menger,  1976, 118).10 Secondly, the assimilation of value to utility leads Schumpeter to designate the marginal utility curve as the value function (Wertfunktion). This is to be borne in mind as we move forward in the present discussion.
- 11 Actually Part Two of WH is made up of two sections. Three chapters make up the first section, which (...)
- 12 “Fragen wir uns, was dem, was wir soeben ableiteten, in der Wirklichkeit entspreche, so lautet die (...)
16Schumpeter asserts repeatedly that the subject of pure economics is value, prices and money. This contrasts with the small amount of space he assigns to money per se in WH. To be fair, money is evoked in passing several times throughout the book. Nonetheless, only 22 pages are devoted to the “Foundations of the Theory of Money” (Grundlagen der Geldtheorie) and the chapter bearing this title appears belatedly, actually not far from the middle of the book. This suggests some discomfort on the part of Schumpeter with the subject. This is further confirmed by the place the chapter occupies in what is the centerpiece of the book, namely Part Two, dedicated to “The Problem of Static Equilibrium”. As a matter of fact, Part Two includes eight chapters, the chapter on money appearing as the second to last one.11 Yet, more perplexing perhaps is the fact that the chapter on the “Elements of Price Theory” is not better treated, as it merely precedes the one on money and includes only 16 pages. Interesting, however, is the connection between the two chapters, as the chapter on prices ends with a section on “Indirect Exchange”, to which refers the opening statement of the chapter on money, which reads thus: “Let us ask ourselves to what corresponds in reality that which we have just derived; the answer is: The phenomenon of money”.12
- 13 Menger uses the term absatzfähig, which translates as “saleable”, hence the choice of the translato (...)
- 14 See Arena and Gloria-Palumbo, 2008, for an in-depth discussion. Also see Menger ([1892c] 2002).
17 In other words, examining the activity of indirect exchange is how Schumpeter introduces money. So far, we are, it seems, light years away from Walras. Whereas Schumpeter takes the pains in WH to explain the rich cognitive contribution of the system of “exact” economics, hence his repeated reference to Walras, here one might be tempted into seeing the link between “pure” economics and money as borrowed rather from Menger. Indeed, in “The Origins of Money”, Menger (1892a) explains how, over time, money slowly emerged. Indeed, the problem he seeks to tackle is that of “the genesis of a medium of exchange”. In the course of his discussion, he begins by showing the advantages of exchange, the fact that its organization raises difficult issues, so that agents quickly understand their interest in holding goods that are “marketable” or “saleable”.13 Menger, we know, rejects the idea of the uniqueness of price, precisely because commodities are characterized by their greater or smaller degree of marketability. So, over time, agents have learnt that there are benefits to resorting to “mediate exchange”, hence an increase in the demand for more marketable goods, leading to a growing and self-reinforcing differentiation among commodities, which slowly created the conditions for the generalized reliance on the precious metals as money.14 Consequently, although Menger recognizes the positive role played by “state recognition and state regulation”, he asserts that “money has not been generated by law” (Menger, 1892a, 255). Money is indeed not a “pragmatic” but instead an “organic” institution, or a “self-organizing” rather than a “designed” institution, to use Hodgson’s (2006, 13) terminology. This is clearly spelt out in the following passage:
We can only come fully to understand the origin of money by learning to view the establishment of the social procedure, with which we are dealing, as the spontaneous outcome, the unpremeditated resultant, of particular, individual efforts of the members of a society, who have little by little worked their way to a discrimination of the different degrees of saleableness in commodities. (Menger, 1892a, 250)
18In sum, Menger shows how money came to be, as the need for a medium of exchange led “economizing” agents, as he calls them, to understand the gains reaped from entering indirect exchange and thus from holding certain commodities, precious metals more specifically, in order to carry out their trade. As a result, indirect exchange takes centre-stage in Menger’s explanation of the phenomenon of money.
19If we return now to Schumpeter’s chapter on the theory of prices, the question to be raised is whether Schumpeter also points to the advantages of relying on a single medium of exchange. The answer to this question is surprisingly “no”. The problem raised by Schumpeter is different:
- 15 “Wir haben gesagt, dass für jedes Wirtschaftssubjekt und jedes Gut der Preis gleich dem reziproken (...)
20We have said that for each agent and each good the price must be equal to the reciprocal value of the ratio of the marginal utilities of the goods that are exchanged… Only, for the level of our marginal utility to emerge, something else must be added. It is not sufficient that this formula be verified for each individual good; rather it is necessary that the prices of all the individual goods behave in the same way as the marginal utilities of the goods to which they correspond. (Schumpeter, 1908, 273)15
- 16 “Eine solche tritt erst dann hervor, wenn wir auf beide Fälle ein und dasselbe Wertmaß anwenden.” ( (...)
21The aspect Schumpeter is investigating here is price determination, but not à la Menger. Schumpeter is seeking to explain how equilibrium prices are determined, but is his line of approach the same as Walras’s? If we move one step forward, it soon becomes apparent that the problem that Schumpeter wishes to discuss concerns the attainment of the equality between exchange ratios and marginal utilities. To illustrate his line of reasoning, he considers an individual who trades a good A for a good B and a good C for a good D. Can it be affirmed, he asks, that the exchange ratios between these goods are equal to the inverse of their marginal utilities? To be able to be affirmative, one needs to rely on one single measurement of value. 16 To make his point, Schumpeter goes on and examines the case where an exchange ratio differs from the marginal utility ratio and describes the arbitrage activity that would need to unfold for the discrepancy to disappear. There is of course nothing unusual about this reasoning, except that Schumpeter adds that, for the process to take place, individuals must accept to acquire goods only for the purpose of undertaking a future exchange:
- 17 “Sie müssen und werden vielmehr Gütermengen erwerben, die sie nicht brauchen, lediglich, um sie geg (...)
They must and will rather acquire quantities of goods that they do not need, exclusively to exchange them again for goods they really need. Only in this way could the price ratios reach the level allowing us to achieve our maximum utility… There is and therefore there must be, in by far most cases, a demand for goods—one or several—which is not explained by “wants” in the strict sense, but by technical necessities of the workings of the market. (Schumpeter 1908, 274-5)17
22Now, Schumpeter is not discussing here the best way for individuals to carry out the exchanges that correspond to equilibrium prices and quantities, which have previously been reached through the solving of the system of equations of the static economy. Rather, what is examined here is the process that would ensue from a situation where prices are not at their equilibrium level. So, Schumpeter is not discussing the reason why individuals would need to acquire a more “saleable” good, so as to be able to exchange it later for the goods they really need. In other words, Schumpeter is not reflecting on the emergence of money as a medium of exchange, but rather he is arguing that a particular good is needed to “measure the value” of other goods, so that arbitrage could take place in a situation where prices are not at their equilibrium level.
23 The fact that there is the need for an instrument that measures value shows that Schumpeter’s view here is quite opposite to Menger’s. Not only does Menger focus on the genesis of the medium of exchange, he also expresses deep reservations about considering money as a measure of prices and even more as a measure of value. For Menger, money is not a measure of value. On the other hand, Menger recognizes that, in practice, the medium of exchange is used for measuring prices, as he argues in the following passage:
Although the theory of “exchange value” in general, and as a necessary consequence, the theory of money as a “measure of exchange value” in particular, must be designated as untenable after what has been said, observation of the nature and function of money teaches us nevertheless that the various estimates just discussed (as distinguished from measurement of the “exchange value” of goods) are usually most suitably made in terms of money. (Menger  1976, 276)
24This idea is further developed by Menger (1892b) in the article he wrote for the Revue d’Économie Politique, where he insists on the subjective valuation of goods, the latter offering the foundation of his explanation of price formation. As a consequence, goods do not present themselves as certain quanta of value that can be compared to the quantum of value contained in the monetary unit:
The function of serving as a measure of price is not, therefore, contained in the concept of money. Several economists have fused the concept of money and the concept of a “measure of value” together, and have involved themselves, as a result, in a misconception of the true nature of money. (Menger,  1976, 278)
25This does not prevent Menger from advocating that the state should limit the quantity of instruments of exchange in order to achieve the invariability of what he calls their “inner value of exchange” (and that is established by the general theory of the determination of prices).
- 18 “Über das Problem der Messung des Wertes”.
- 19 “Die Massen der Körper sind nur Verhältniszahlen und können beliebig ausgedrückt werden; aber wenn (...)
- 20 “In der theoretischen Konstruktion unserer Einheit liegt der eine Grundstein der Geldtheorie. Sowei (...)
26 This contrasts altogether with Schumpeter’s position, which is articulated, relatively early in WH, actually in his (seventh) chapter on the “Concept of Value”. The latter ends with a section titled “On the Problem of the Measurement of Value”.18 When it comes to confronting the values of different goods, one can take one as a unit and express all the others in terms of it. Here Schumpeter uses the analogy with mechanics, where: “the masses of bodies are only relative numbers and can be expressed arbitrarily ; but if one of them is determined, so are all the others”. 19 He pursues by affirming that: “In the theoretical construction of our unit lies a founding principle of the theory of money. In so far as money functions as a measure of value, we understand its essence on the basis of such considerations.”20 Consequently, we can safely assert that Menger was not Schumpeter’s source of influence with regard to his conception of money. Quite naturally, this leads to questioning whether Walras may have provided greater inspiration.
27When perusing the first section of “Grundlagen der Geldtheorie”, the chapter of WH that is dedicated entirely to the theory of money, the reader cannot help feeling somewhat puzzled, if not amused, by some of Schumpeter’s comments. The young economist is utterly fascinated by what he (thinks he) has discovered, that is, the fact that the theory of prices contains the mainspring for the emergence of the theory of money. This is achieved, he argues, by applying the methods of mathematical inference, an accomplishment that provides evidence that exact theory is not a mere tautology. No preliminary hypotheses, he notes, specific to the phenomenon of money, are needed for making this (quasi) new discovery.
- 21 The first problem is whether money can consist of something having no “use value”, and the second i (...)
28 The chapter continues with a section that is devoted to the critique of existing monetary theories, including those of Walras and Jevons, and ends with some rather ambivalent observations on the quantity theory. The core of his discussion is contained, however, in the third section, where Schumpeter proposes to look in two directions. Accordingly, he introduces two “founding principles” and a discussion of two “important problems”.21 The two founding principles are: 1° the need to resort to indirect exchange, already taken up in the chapter on the “Elemente der Preistheorie”; 2° the separation between the two functions of money, i.e. measure of value and means of exchange. This distinction leads the author to assert that: “The theory of money as a measure of value and the theory of money as a medium of exchange are two completely different things” (WH, 289), a conception we know Schumpeter has approvingly borrowed directly from Walras.
29We now focus on the first founding principle. In order to assess the proximity between Walras and Schumpeter, we will go over a few aspects of Walras’s monetary theory, by drawing sparingly on Bridel (1997). In the Éléments, money appears first as the standard of measure (to use Jaffé’s terminology): the numéraire is introduced quite late, i.e. in Lesson 25 of the first edition, where Walras writes:
- 22 This quote is borrowed from Bridel (1997, 3), who cites the 1954 translation by W. Jaffé, as well a (...)
The situation of a market in a state of general equilibrium can be completely defined by relating the values of all commodities to the value of any one particular of them. That particular commodity is called numéraire [or standard commodity]; and a unit quantity of this commodity is called a standard [étalon]. (Walras, 1874, 144) 22
30This quote remained unchanged throughout the different editions of the Éléments. Accordingly, it is found in identical form in Lesson 14 of the fourth edition, which bears the title “Theorem of Equivalent Redistributions of Commodity Holdings. Concerning a Standard of Measure and a Medium of Exchange” (Jaffé’s translation), indicating a joint introduction of these two functions of money, although these two functions are clearly differentiated in Walras’s construct. An aspect that is to be forcefully underscored is that the numéraire is always considered by Walras as a concrete commodity. Indeed, Walras writes:
In general…the commodity which serves as the numéraire serves also as money and acts as a medium of exchange. The standard of measure of value thus becomes the monetary standard. The two functions are, nevertheless, distinct, even when they are found in the same commodity. (Walras, 1954, 189)
31So, as regards the relationship of money as a unit of account with money as a means of exchange, Walras and Menger’s positions are diametrically opposed. Furthermore, in Lesson 14, Walras already announces that:
In the real world the matter presents itself quite differently. Every trader keeps available a stock of money for eventual exchange; and this being the case the use of a commodity as money does affect its value in ways that we shall study later on. (ibid., 190)
32This is further discussed in Lesson 30, where Walras solves the “equations of circulation and money”.
33 How does this square with Schumpeter’s exposition of the “Foundations of the Theory of Money”, or rather its preamble in the “Elements of Price Theory”? If we pick up from where we left off previously, the latter chapter ends with an intriguing statement:
There is and therefore there must be, in by far most cases, a demand for goods — one or several — which is not explained by “wants” in the strict sense, but by technical necessities of the workings of the market.
- 23 “Es wird und muß daher in weitaus den meisten Fällen eine Nachfrage nach Gütern — einem oder mehrer (...)
This result is extremely important, not only because it sheds light on an essential aspect of the exchange process, but also because it provides for a very important application. We shall take up this point in the next chapter. The fact that, to my knowledge, what was said about the theory of prices was expounded before us by only one theorist, namely L. Walras, who in fact did not develop this decisive application, explains so many weaknesses in today’s economic theoretical system.23
34In other words, Walras was, according to Schumpeter, the only theorist to have devised a correctly conceived theory of prices. However, Schumpeter claims, Walras did not derive from this the need for a unit of account so that prices may be brought to reach their equilibrium level. In sum, Schumpeter espouses Walras’s theory of prices, hence his defence of “pure” economics in the form of Walrasian “exact” economics. However, he claims to have detected an unexplored offshoot of Walras’s theory of prices, and pointed to the need for a unit of account so that indirect exchange may take place. Let us now see what Schumpeter makes out of this alleged “finding”.
35 Having discarded existing theories, Schumpeter sets out in the final section of “Grundlagen der Geltheorie” to state the small number of basic principles that are necessary for building a theory of money. Indirect exchange is its founding principle. When more than two goods are exchanged among more than two individuals, there are acts of exchange that solely fulfil a technical function allowing the market to operate. This is what Schumpeter reasserts in the following passage. Moreover, he tediously argues that the possibility of exchange would involve an additional utilization for goods, which would therefore affect their value (utility) functions:
- 24 “Wir sprachen also den ersten fundamentalen Grundsatz der Geldtheorie aus, in dem die Erklärung des (...)
We thus stated the first fundamental principle of the theory of money in which lies the explanation of the essence of the phenomenon of money. The laws determining the value of those goods that serve as money derive directly from it. The possibility of exchange alters altogether the value function of every single good [eines jeden Gutes] … It represents indeed a special use of that good from the standpoint of its owner. It is also easy to see how the value functions are altered. The values of those goods [jener Güter], which can be exchanged for the goods in question [betreffenden Güter] and, among these, those that are in any case actually exchanged, must be introduced into the value functions of the former.24 (Schumpeter 1908, 287)
36Now what Schumpeter means in this passage is not crystal clear. Does he mean that the value of the good(s) that serve(s) as money must enter the value (utility) functions of all goods? This might be how we should interpret the passage. Indeed, Schumpeter seems to have in mind that it is more specifically the value of “monetary” goods that should enter the value functions, as seems to be indicated by his addition of the parenthetic clause: “those that are in any case actually exchanged” [welche tatsächlich eingetauscht werden]. If this is what Schumpeter is driving at, we have here an instance of money-in-the-utility-function.
37 This interpretation is supported by what Schumpeter writes in the next passage, where he focuses on the good that actually serves as means of exchange. The value (utility) function of money is altered by the inclusion of the value of goods for which it is exchanged. However, this alteration is described as a “further” change, which is unfortunately not discussed, as the author leaves that to responsibility of the reader:
- 25 “Für ein Gut, das nicht nur zum Tausche bestimmt, sondern schon zu diesem Zwecke erworben worden is (...)
For a good that is not intended only for exchange, but has actually been acquired for this purpose, the same applies to an even greater extent. Indirect exchange will therefore bring about a further alteration of the value functions of the goods, which are used to carry it out, in a way that the reader himself can easily explain.25 (ibid.)
38The issue that Schumpeter seeks to raise here is that of how the value of a good that serves exclusively as a means of exchange is affected:
- 26 “Allein noch in einer anderen Art beeinflußt der indirekte Tausch die Werte, diesmal nicht bloß die (...)
One way or another, indirect exchange affects values, that is, not just the value functions of those goods. There is namely the tendency to withdraw those goods from the use they are capable of serving directly. That there be in fact many such goods will be less obvious, however the practice of using only one or only a few goods for carrying out indirect exchange acts, as is currently the case with the precious metals and was certainly typical for all goods serving as "money”, relatively large amounts can continuously be withdrawn from their original use. This results, like any restriction of the quantity of a good, in the increase of its marginal utility.26 (Schumpeter 1908, 287-8)
39In any case, whatever the intention of the author, the reasoning breaks off suddenly and the reader is left to her own devices.
40 Of course, this contrasts with Walras’s own undertaking. As Bridel writes, in the fourth edition of the Éléments, the one that Schumpeter in all likelihood had at hand, “Walras claims to have eventually found room for money in the agent’s utility function; that is, to have integrated money and value theories” (Bridel, 2002a, 271). When writing WH, Schumpeter had no pretence at having brought forward new ideas. So why was he not content with referring to Walras’s Section VI of the Éléments? The reason is that he probably had not entirely grasped the intricacies of Walras’s reasoning, so it is possible that he had not come completely to terms with the “alterations brought by Walras … to the time structure of his model” (ibid.). Indeed:
[Walras’s] whole exercise amounts to a clever adaptation of the relationship between [his] once-and-for-all market and his periodical market in order to make room for money considered as circulating capital.” The once-and-for-all market (including, for the first time, money) is kept as ‘the preliminary phase of timeless tâtonnement…with equilibrium [including money] established in principle’ (1954: 378; original italics): equilibrium prices are instantaneously reached and disequilibrium exchange and production are, of course, ruled out ex definitione. In the fourth edition, the periodical market coincides with “the phase [where] the static equilibrium has actually commenced, so that equilibrium is established in fact” (1954: 378; original italics): equilibrium exchanges and production defined on the once-and-for-all market are actually taking place in a fully “monetized” economy. (ibid.)
41So, some similarity does appear in the treatment of money by Walras and Schumpeter, as it is the tâtonnement process that offers the setting for Walras’s analysis, while Schumpeter discusses the problem of the attainment of equilibrium. However, Schumpeter comes nowhere close to the refinement present in Walras’s text, an unsurprising outcome, given that we are comparing a piece by a very young author with the most accomplished version of the major work of a mature scholar. Indeed the way Walras managed to introduce money into the utility function is quite contrived. To quote Bridel:
To make short a very long and difficult argument, together and simultaneously with the three other sets of markets, the equilibrium price of money is first theoretically determined in principle and defined in terms of numéraire. However, at that stage, Walras has not yet suggested any rationale for the demand of money. Phase 2 of his time structure…offers just that: the equilibrium price of money determined ex ante through a tâtonnement in principle on the money market in phase 1 is a function of the expected supply and demand for money during phase 2. In Walras’ framework, agents know with absolute certainty in phase 1 the exact amounts they must make and receive in payments, and the exact dates on which these payments must take place during phase 2. … Money is conceived as a good yielding no utility in itself, but, by providing “services” of an inventory nature… In this conceptual framework, money, like any other stock of circulating capital, has to be held in quantities that are assumed to be perfectly known along perfectly predictable time patterns of consumption during phase 2. Just as the consumer derives satisfactions from the holding of good inventories, he or she can also be viewed as deriving utility from holding command over future goods in money form. … This is how money is formally introduced into the utility function. (ibid., 272)
- 27 As Bridel writes: “This is the old and well-known Hicksian critique (Hicks 1933: 446–8)”.
42Nonetheless, as Hicks noted many years ago: “Clearly, Walras provides no rationale, in the modern sense of the term, for the holding of money. … Why should agents hold unproductive cash balances during the gap between receipt and outlay when they can instead hold interest-yielding assets” (ibid., 273)?27 Actually, agents are compelled to hold money, not out of choice, but because of the transaction technology that is imposed upon them. Consequently, Walras’s lack of success after nearly 30 years of efforts leads us to judge the shortcomings of Schumpeter’s own attempt with not just only a little indulgence.
43The reader seeking to appraise Schumpeter’s role in the area of monetary theory as a go-between between Walras and the German-speaking world is likely to end up dissatisfied. More particularly, in Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie his attempt at incorporating money within the general equilibrium framework fizzles out quickly. Of course, no blame can be laid on the young scholar he was then for the difficulties he encounters. It is rather his very approximate knowledge of Walras’s contribution that is somewhat disenchanting. Mention is made in WH neither of Walras’s theory of capitalization and credit, nor of his theory of circulation and money. One might even wonder whether Schumpeter had actually read those sections of the Éléments d’Économie Politique Pure at the time he was writing his Habilitationsschrift. Indeed, this contrasts with his presentation of the Éléments in his 1910 obituary article on Walras. Published only two years after WH, “Marie Esprit Léon Walras – 1834-1910” reveals a more explicit acquaintance with the book’s contents, in particular with the chapters dealing with the “problem of capitalization” and the “equations of circulation” (Schumpeter 1952, 78). For the first time also, Schumpeter refers to Walras’s notion of encaisse désirée, a notion he will later strongly reject in Business Cycles:
This Walrasian idea of an encaisse désirée, which reappears in Marshallian analysis and has recently, it seems, entered upon a new lease of life, is one of the least valuable elements in the great Frenchman's mighty structure. It is harmless only in the analysis of stationary states, although even there it implies a misrepresentation of facts (Schumpeter, 1939, Vol.II, 547).
- 28 See also Bridel (2002a, 268) and Bridel (2002b, 228).
44That Schumpeter’s undertaking to integrate money in the general equilibrium framework runs dry is hardly surprising, as the enterprise was doomed from the outset: “there is no place for money in modern Walrasian general equilibrium models”, Bridel (1997, viii) affirms, [d]espite decades of excruciating theoretical effort”.28 What is interesting, nonetheless, in Schumpeter’s endeavour, is that it is perhaps the first – albeit most likely an inadvertent one – attempt, after Walras’s, at integrating money in the utility function. That Schumpeter will never take up this matter seriously again is certainly explained by his quasi exclusive focus in the following years on dynamics, which would lead him to abandon statics, i.e. the only domain where “exact” or “pure” economics applies. One might also venture that relinquishing this line of research might have been the sign that he intuitively perceived that nothing could come out of it.
45 Dynamics and economic evolution are crucially of much greater importance than statics, Schumpeter argues already in the final chapters of WH. And the three ingredients that provide the foundations of Schumpeter’s theory of dynamics, namely innovation, the entrepreneur and bank credit, receive full attention in The Theory of Economic Development. Interestingly, Schumpeter already had in mind the role played by credit in a context of economic development, when writing the last chapter of WH. By the time he had finished writing The Theory of Economic Development, i.e. only three years after the publication of WH, credit had become his chief focus.
- 29 The first twelve chapters of Das Wesen des Geldes were first published posthumously in 1970 and lat (...)
- 30 This point is argued at length in Dangel-Hagnauer (2002). The interested reader can also refer to L (...)
46 However, Schumpeter never abandoned the project of building an economic theory of money of his own devising. This led to the publication of “Money and the Social Product” (Schumpeter, 1917/1918), written at a time where monetary theory in Germany was in full disarray. It thus contains a critique of both Knapp’s state theory of money and of the quantity theory. It also brings forward ideas that will give rise to full development in Das Wesen des Geldes, 29 his manuscript on money that was published posthumously in 1970, and whose main contribution is a conceptualization of money as an institution.30
47 Money is thus a domain to which Schumpeter returned recurrently throughout his lifelong career. At each stage, he evolved, beginning in his Habilationsschrift with an attempt at integrating money in the general equilibrium framework, and journeying all the way to his unpublished manuscript, in which money is fitted with the attires of an economic institution. Without ever going back on what he had written previously, he constantly moved forward experimenting new solutions. That he never finished his manuscript is perhaps emblematic of the difficulties that are inherent in this area of economic analysis.
48 Over the same time span, Schumpeter’s admiration for Walras never faltered, although in his own writings he gradually moved away from him. In the end, no trace of Walras’s efforts at integrating money are visible in his manuscript, where he recognizes instead that he has become aware of the “doubts and difficulties”, raised by later writers, about Walras’s theory of prices (Schumpeter 1970, 215).
49 Whereas the young Schumpeter writing his Habilationsschrift was excited by the idea that he may have unearthed something even the great Walras had not noticed, the mature and more poised writer had long forgotten his “finding” and come a long way, in his manuscript, towards a conception of money that owed as much to economic sociology as to economic theory. But that is a topic in its own right.
I am indebted to three anonymous referees for their valuable comments and suggestions. The usual caveat applies.