Introduction
During the post-war era, most developed capitalist entities witnessed unprecedented changes in their industrial structure. Most of these changes centred around the shift of labour and capital from goods production to service provision. A general consensus exists regarding the major post-war social achievements, such as rapid and new technological advances; the tertiarisation of industries; increased internationalisation; and ongoing demographic changes, such as rapid ageing and the advancement of women in the workplace, which has, in turn, led to the destabilisation of the family.
Among these, the shift towards the service economy has been commonly accepted as a profound phenomenon that has captured the focus of scholars by addressing the forces that have propelled these post-war changes and the resulting new perspectives.
In fact, social scientists have coined various concepts to explain the dynamics of such development and change. The most important and influential among them is Bell’s (1999) “post-industrial society.” Although disagreements exist over what social scientists call “the demise of class conflict” at this historical stage, post-industrialists acknowledge the existence of only two categories of employment—goods-producing and service-rendering employment. Thus, the notion underlying the post-industrialists’ arguments is a dichotomy. This view was supported by Clark (1940) and Fisher (1952), who asserted that the service economy has replaced the manufacturing industry, although no proper examination has been conducted to assess its validity.
However, as Walker (1985) stressed, in this type of argument, many disparate phenomena have been haphazardly brought under the umbrella (and overburdened) concept of “services.” Attempts have not been made to circumvent such misconceived notions of homogeneity, by thoroughly exploring the conceptual meaning of “services” and criticising its conventional use. Accordingly, Marxists have taken the responsibility of examining this concept in detail.
The topic of service labour is embedded within the concept of productive/unproductive labour (PUPL). The socio-economic implications of these categories have led to heated debates amongst economists not only in the bygone eras of Smith 1 and Marx, but also amongst those who subscribe to modern Japanese Marxist ideologies since the end of the Second World War. While most people on both sides of this debate ascribe great importance to Smith’s distinction between these two types of labour, there has been much disagreement over him categorising service labour as “barren and unproductive” (Smith 1977, 885).
The majority—not only including the direct successors of Smith, such as Cannan (1894), Germain Garnier (as cited in Cannan 1894), and Say (1861), but also some Marxist economists and modern economists, especially the advocates of “tertiary industry”—argue that service labour is productive, since it creates surplus value, and state that it would be unfair to consider it unproductive simply because it does not result in a material commodity. Challenging the notion that productive labour must result in a material commodity, they have tried to prove that any type of labour can create value if exchanged against capital; nearly all of them believe that Smith treated service labour as non-material. Even some Marxist economists, extensively quoting Marx to buttress their claim, believe the materialist interpretation to be consonant with that of Marx.
Curiously, none of them admits that the representation of service labour in its original context must be significantly revaluated. In other words, while they draw on the concepts of Marx and Smith to fortify their claims, they do not seem to care as much about what Smith or Marx themselves meant. Consequently, their arguments on service and productive labour have inevitably paralleled those of the other side, leading to the arguments of both sides being limited by this misinterpretation of the classical meaning of service labour. Some Marxists, in particular, being unaware of the social background to the classical meaning of service labour, have succumbed to a methodological trap, by accepting the post-industrialists’ “vulgar” definition of “service” and failing to see that the post-industrial viewpoint is only a superficial reading of fundamental changes in contemporary capitalism.
Therefore, acknowledging this misunderstanding and arriving at a more accurate representation of service could be the key to resolving this dispute. To this end, our study provides some considerations. To avoid controversy, we refer to Adam Smith’s original definition as the “classical” definition of service labour. We find that this connotation was also referred to as “service” in pre-capitalist societies.
Accordingly, this study is divided into two parts. The first part attempts to understand what Smith and Marx precisely meant by the term “service.” We explore Smith’s dual criteria for productive labour, within which the classical definition of service labour is embedded. When Smith talked about labour in which the service relationship [Dienstverhältnis] prevailed, it appears he was thinking of his first criterion [Formbestimmung] of productive labour, which has nothing to do with the materiality of labour. This misinterpretation of Smith’s work is especially apparent in what Smith referred to as specific activities which was later mentioned as “service in kind [Naturaldienst], or service objectified in a thing” 2 by Marx (Marx and Engels 1981, S.374). Thereafter, we reveal why this misinterpretation occurred (and even prevailed) in modern times by exploring the socio-economic implications of service, which are deeply embedded in Marx’s theory of capital.
In the second part, we attempt to clarify how theoretical consistency could be sustained between service labour and the theory of PUPL. Note that service labour has different connotations in Marx’s context; in the conventional view, the distinction between PUPL and its relation to the definition of service labour is crucial in a theoretical sense.
We will address this issue as follows. First, we distinguish service labour in Marx’s context, “labour as service,” from non-material labour and “tertiary activities.” Second, we derive two distinct connotations: labour productive of capital and labour productive of value. Third, we define the materiality of production, to clarify the misinterpretation and misunderstanding regarding the materiality of service labour. Finally, we discuss the question of whether tertiary activities are productive of value.
The Classical Definition of Service and Its Misinterpretation in Political Economy 3
The term “service” is conventionally defined in contrast to goods. Naturally, this definition assumes that services are not tangible, are neither discrete nor mobile, and disappear after consumption. This view of service has resulted from the misinterpretation of Adam Smith by vulgar economists.
Duality of Smith’s Criteria for Productive Labour
Classical Definition of Service Labour Embedded Within That of Productive Labour
Scholars before Smith—such as William Petty and Francois Quesnay 4 —also commented on service labour, but their definitions have little impact on the aforementioned debate; thus, we shall begin our study from Smith himself.
Smith’s primary concern was obviously not the function of labour, but its form (or “productivity”):
Adam Smith (1723–1790) had no interest in services as such. He was concerned with labour, and tried to describe and understand how the society in which he lived, functioned and developed, and in particular, how capitalists accumulated wealth. But the theoretical analysis [of Smith] for that [service in classical term] development remains embedded in the concept of productive and non-productive labour. (Delaunay and Gadrey 1992, 16)
Therefore, Smith’s perspective of productive labour is a vital clue for us to understand his views on service labour in the classical sense.
This perspective is two-fold. It is well-known that Smith referred to labour’s productiveness (and, therefore, unproductive labour as well as service labour) in Ch. 3, Book 2 of The Wealth of Nations. However, it is lesser known that other books and chapters of this work (such as Ch. 4 and 5 of Book 2, along with Books 4 and 5) also contain references to service labour, which could be an indispensable supplement to his determination of productive labour.
Productive Labour: Primary Criterion
We begin with the famous narrative in Ch. 3 of Book 2, where Smith discusses his first criterion for productive labour. On “whether it produces a return on the money invested in it [work]” (Delaunay and Gadrey 1992, 12), Smith stresses that “whatever part of his stock a man employs as a capital, he always expects it to be replaced to him with a profit. He employs it, therefore, in maintaining productive hands only” (Smith 1977, 441).
Thus, for labour to be productive, it must be maintained by capital and not by revenue. Different perspectives of a similar narrative are found in Ch. 4 of Book 2: “If he uses it as a capital, he employs it in the maintenance of productive labourers, who reproduce the value with a profit” (Smith 1977, 465), and “all capitals are destined for the maintenance of productive labour only” (477). Furthermore, Smith states, “the persons whose capitals are employed in any of those four ways are themselves productive labourers” (480), referring to “improvement or cultivation of lands, mines, or fisheries,” “all master manufacturers,” “all wholesale merchants,” and “all retailers” (477).
These references should not be wrongly interpreted, as does J. B. Say, and Smith’s criterion for productive labour is based on industry classification, since all productive labourers in the aforementioned four industries are employed by capital. Others employed by revenues are, as we will see later, regarded as menial servants. Contrarily, this specific criterion of Smith could be interpreted through Karl Marx’s phrase, “labour which is not exchanged with capital, but directly with revenue” (Marx and Engels 1989, 12). We shall call this the “form and historical” interpretation.
By this first criterion, whether labour is productive is judged not by its materiality but by its social character—especially, by how it is maintained. This criterion is certainly available for defining unproductive labour.
In Ch. 3 of Book 2, Smith argues that “unproductive labourers, and those who do not labour at all, are all maintained by revenue,” and the workmen (those of productive labourers) who “employ any part of them [unproductive labourers] in this manner” by “his wages” are considered “generally but a small one” (Smith 1977, 442). This is because in Smith’s era, other than those labourers, several classes (such as the yeomanry, independent handicraftsmen, both small masters and journeymen) were not maintained by wages. Those who were excluded from the subsumption of capital would not employ their stocks as one; therefore, the labour maintained by the revenue of the following classes were rendered unproductive, irrespective of whether they belonged to “any of those four ways” previously discussed.
For instance, “[t]he unproductive class, that of merchants, artificers, and manufacturers, is maintained and employed altogether at the expense of the two other classes of that of proprietors, and of that of cultivators” (Smith 1977, 888). The same class which was referred to as productive in Ch. 4 of Book 2, was regarded here as unproductive, when we change the perspectives to those of the “proprietors” and “cultivators.” And thus, Garnier’s claim that “Smith’s assertions contradict themselves” (Cannan 1894, 25) was incorrect. When maintained by proprietors or cultivators, they are not sellers of the power of their labour, but of its products. It is obvious that in this specific case, the stock of proprietors or cultivators is not to “restore the capital and pay the interest,” but “reserved for immediate consumption” (Smith 1977, 465). Therefore, those “generally productive” are regarded as unproductive labourers.
Generally speaking, Smith’s first criterion is concerned with the social form of labour, with factors affecting the materiality or function of labour getting abstracted. Therefore, the only criterion for being productive is the capitalist relationship between the buyer and seller of a specific labour power.
Productive Labour: Supplementary Criterion
However, when discussing the result of paid work, Smith referred to a different criterion (Delaunay and Gadrey 1992, 12): productive labour “fixes and realizes itself in some particular subject or vendible commodity, which lasts for some time at least after that labour is past” (Smith 1977, 438). By contrast:
The labour of the menial servant does not fix or realise itself in any particular subject or vendible commodity. His services generally perish in the very instant of their performance, and seldom leave any trace or value behind them for which an equal quantity of service could afterwards be procured. (Smith 1977, 439)
In these narratives, in addition to the first criterion (concerning the social character of labour), another one (regarding the durability of its results) was given. Obviously, whether labour “fix[es] or realize[s] itself in any particular subject or vendible commodity” has no bearing on the buyer–seller relation; therefore, “[h]ere the definition of form, the determination of productive and unproductive labourers by their relation to capitalist production, is abandoned” (Marx and Engels 1989, 18).
Curiously, it is Karl Marx (rather than Smith’s successors) who correctly noticed the duality in Smith’s criteria on productive labour. Most of Smith’s successors (who Marx called the purveyors of “vulgar economics”) ignored his first criterion and quoted his second criterion to conclude that he “wrongly” restricted the wealth of a nation to the material domain. The editor of Smith’s work, Cannan, for instance, criticised him, saying he “did not see that the manufacturer and merchant are maintained by the menial services of cooking and washing just as much as the cooks and laundresses are maintained by the manufacturer of bonnets and import of tea” (Cannan 1894, 24).
However, certain forms of labour could be unproductive under Smith’s first criterion while still being productive under his second criterion, even if we consider it to be realised in any “vendible commodity.” As Marx noticed,
Smith knows quite well, a seamstress whom I get to come to my house to sew shirts, or workmen who repair furniture, or the servant who scrubs and cleans the house, etc., or the cook who gives meat and other things their palatable form, fix their labour in a thing and in fact increase the value of these things in exactly the same way as the seamstress who sews in a factory, the engineer who repairs the machine, the workers who clean the machine, or the cook who cooks in a hotel as the wage labourer of a capitalist. These use values are also, potentially, commodities; . . . Thus[,] these persons have potentially also produced commodities and added value to the objects of their labour. (Marx and Engels 1989, 20)
Marx saw this contradiction as Smith’s ambiguity in regards to the ideas held by both physiocrats and mercantilists (Marx and Engels 1989, 17–18). In fact, considering the sociohistorical background of Smith’s era, the two criteria were compatible with each other at the time. 5 However, Smith’s successors did not share the same background, in which the capitalistic mode of production underwent rapid developments and thereby radically changed the form of labour. Under this circumstance, unless scholars take a historical standpoint like Marx, it would be difficult for them to understand Smith’s actual meaning when he refers to productive labour or “service labour.”
However, we stress here that Smith blended the two criteria only in the chapter on labour productivity. The rest of his The Wealth of Nations, as we will notice in the next part of this article, is based exclusively on his first criterion, especially when he talks about service labour in classical terms.
“Service in Things”: Why the Classical Definition of Service Labour Cannot Be Considered “Non-material”
Since Smith’s arguments on service labour (other than those in Ch. 3 of Book 2, which have been covered by almost every previous study on the topic) are spread across several chapters, it is necessary to comprehensively inspect his narratives to fully understand his thoughts on “service labour.”
Let us begin with the labour of menial servants and similar cases (such as “the servants of the public” or “the most respectable orders” [Smith 1977, 439]). These are, considering Smith’s first criterion, unproductive because they are maintained by society’s annual revenues. Smith’s second criterion, while also considering labour as unproductive if it generally “does not fix or realize itself in any particular subject or vendible commodity” (439), did make some exceptions for cases such as a seamstress (eine Nätherin), a repairman, or a cook (eine Köchin).
If we restrict “service labour” to those servants and consider “privileged standings” from the same perspective as the “vulgar economists,” we are to “open the flood-gates for false pretensions to the title of productive labour” (Marx and Engels 1989, 27; emphasis in the original). It is true that most of these servants perform their labour in a non-material form, and we shall call this interpretation a materialistic one. For convenience, we will treat the service labour in materialism terms in the manner of “service.” It would be misleading since the materialist interpretation is coherent with Smith’s original idea if we restrict our scope to Ch. 3 of Book 2 of The Wealth of Nations. However, does “service” today include exactly the same trades as it did during or before Smith’s time?
This is what we call “failure to see the forest for the trees.” If we investigate contexts apart from those in Ch. 3 of Book 2, the inconsistency between the materialist interpretation of service and Smith’s narrative is revealed.
Smith used the phrase “service” in Book 1 of The Wealth of Nations before sharing the criteria for productive labour (also the criteria for service labour) (Smith 1977, 106). The connotation of this word is similar to that in Books 2 and 3, where Smith referred to the labour of several manufacturing trades as “service.” 6 These services are procured for direct consumption or the fulfilment of personal needs, and thus, they are exchanged for revenue and not for capital.
These narratives could be understood with Smith’s first criterion; however, there are significant inconsistencies between his second criterion and the materialist interpretation. A decisive case appears in Book 3, where the phrase “the inhabitants of the town and those of the country are mutually the servants of one another” (Smith 1977, 503) was used.
It is impossible to explain those narratives under the materialist interpretation of “service,” for these trades (such as artificers, smiths, carpenters, and masons) do “fix or realize [themselves] in any particular subject or vendible commodity” and, therefore, “fix their labour in a thing and in fact increase the value of these thing . . . [t]hus these persons have potentially also produced commodities and added value to the objects of their labour” (Marx and Engels 1989, 20).
These specific cases would be rendered as “service in things” (Naturaldienst), if they are consistent with Smith’s first criterion. For instance, exchanges between citizens and farmers proceed with their incomes; therefore, “the inhabitants of the town and those of the country” are de facto “maintained by a part of the annual produce” (Smith 1977, 439), and the annual produce is considered the revenues of their partners.
We can conclude that Smith’s first criterion is consistent throughout The Wealth of Nations and, therefore, should be considered his primary criterion. Simultaneously, his second criterion has a narrow base and is, therefore, considered a supplementary criterion. Naturally, the “form and historical” interpretation of service provided by Karl Marx is consistent with this prior criterion. Under this preliminary, we define the “form and historical” interpretation of service as “service labour in the classical term.”
While the materialist interpretations of Cannan, Say, Garnier, Senior, and some of the Marxists do not cover the sphere of “service in things,” they should be considered a misinterpretation of service labour in classical terms. However, why Smith gave a supplementary criterion in addition to his primary one, why the primary criterion and its “form and historical” interpretation have gradually lost their impact, and why a misinterpretation of service labour is prevalent in recent times, are questions that remain to be answered.
Sociohistorical Background of the Classical Definition of Service Labour
Dienstverhältnis and the “Form and Historical” Interpretation of Service
A scholar like Smith would, undoubtedly, have based his research and theoretical criteria on socio-economic practice. Karl Marx, probably the only scholar of his time who favoured Smith’s standpoint on how labour can be regarded as productive, genially pointed out that Smith’s primary criterion was meant to distinguish between two kinds of labour: the capitalistic form and the service form (in “service relation”).
Service relation (Dienstverhältnis) refers to a specific legal and economic relationship between the upper and lower standings, where the lower standings bear the obligation to present “service in things” and/or labour (persönliche Dienst) (Watanabe 1985, 90–91) to their masters (Masuda 1986, 16). It is thought that in the era when The Wealth of Nations was written, serfs (villeins) and some of the yeomanry were still ruled by service relations. These facts are discussed in several chapters (Smith 1977, 468, 519, 1112). Moreover, the service relationship could also be confirmed in communes, where one could gain his residence “neither by notice, nor by service, nor by apprenticeship, nor by paying parish rates” (Smith 1977, 196–197). Marx quoted the British census in 1861, stating that most of the population belonged to the so-called “service class” (dienende Klasse) (Marx and Engels 1962, S.470). Therefore, Smith apparently did not ignore service relations when he referred to the phrase “service labour.”
As long as service relations, as a mode of production, paralleled and competed with the capitalistic mode of production, it was obligatory for capitalist spokespersons to ensure that service labour under the service relation was ideologically unproductive. This was a natural obligation for classical economics.
When Smith granted the labour maintained by capital for productive labour, and those maintained by expenses, “who in the common apprehensions of men are regarded as the most productive,” for “barren and unproductive” (Smith 1977, 885). Through his narratives, we can see that he credited capitalistic activities with being productive.
By contrast:
bourgeois narrowness, which considers the capitalist forms of production to be the latter’s absolute forms—and therefore the eternal natural forms of production—is able to confuse the question of what productive labour is from the standpoint of capital with the question of what labour is productive in general, or what productive labour is in general. (Marx and Engels 1994, 128; italics in the original)
Adam Smith, who lived in a society where the two modes of production competed against each other, knew “what productive labour is from the standpoint of capital” and knew what the real issue was (Marx and Engels 1994, 128; italics in the original).
We summarise this particular issue in Table 1.
Smith’s primary criterion is shown in the rows of Table 1, whereas his supplementary criterion is depicted in the columns. Since Marx explains,
where money is exchanged directly for labour . . . it is bought as a service; this is nothing more than an expression for the particular use value provided by labour, just like every other commodity; but it is a specific expression for the particular use value of labour, in so far as labour does not provide services as an object but as an activity. 7 (Marx and Engels 1994, 138; italics in the original)
We exclude labour “served as a thing” from the discussion of service labour.
Meanwhile, when discussing “labour served as an activity,” even the non-material labour provided by actors or clowns—which “perish[es] in the very instant of their [its] performance” and is thus rendered unproductive under Smith’s second criterion—should be categorised as productive under his first criterion, if the labour itself is maintained by capital (Marx and Engels 1977, SS.519–520; Marx and Engels 1989, 164).
Simultaneously, the material labour of a tailor, which is “fixe[d] and realize[d]” in the form of clothes, is rendered unproductive, because it is exchanged against the revenues of farmers or manufacturers (Marx and Engels 1977, SS.443–444). Menial servants, whose labour “fixes itself in a material product and could just as well (in its result) be a vendible commodity, as it in fact is for the hotel proprietor” (Marx and Engels 1989, 21), or “certain labours of MENIAL SERVANTS may therefore equally well take the form of commodities (potentia) and even of the same use values considered as material objects”; thus, “productive” under Smith’s second criterion, should be rendered “unproductive” under his first criterion for “they are not productive labour, because in fact they produce not ‘commodities’ but immediate ‘use values’” (Marx and Engels 1989, 28; emphasis in the original).
Marx’s “form and historical” interpretation is, therefore, completely consistent with Smith’s primary criterion. Marx also pointed out that Smith’s supplementary criterion was generally understandable if we consider that the works of non-material labour that “perish in the very instant of their performance” were seldom maintained by capital in Smith’s era.
Meanwhile, material labour that “fixes and realizes in some particular subject” was generally subsumed by capital’s domination in Smith’s era, resulting in the disappearance of most of the “service in things.” This was the reason why, as Marx noticed, Smith considered the labour of menial servants to “generally perish in the very instant of their performance” (Smith 1977, 439). 8
We summarise these changes in forms of labour in Table 2.
Historical Background for Smith’s Second Criterion: Forms of Labour in His Era
Smith’s era was a time when capitalism conquered almost every perspective of material production, leaving very limited exceptions for “service in things,” such as the labour of cooking women (as menial servants). In that era, money (as revenues) could exchange
Aut 9 against commodities which capital alone produces and sells, aut against labour, which just like those commodities is bought in order to be consumed; that is, only for the sake of its particular material determination, its use value—for the sake of the services which, through its particular material determination, it renders to its buyer and consumer. (Marx and Engels 1989, 14; italics in the original)
This particular historical insight forms the basis for Smith’s supplementary criterion, where the “service in things” field (remember that this category accounts for most of the contradiction between Smith’s two criteria) was rapidly disappearing, as shown in Figure 1.
Consequently,
only an inconsiderable part of them (like cooks, seamstresses, jobbing tailors and so on) will produce material use values . . . consequently only a quite insignificant part of these unproductive labourers can play a direct part in material production once the capitalist mode of production has developed. (Marx and Engels 1989, 15) 10
The subsumption of personal services by capital was a relatively recent event for Smith. “This is one of the aspects which lead Adam Smith to put forward other points of difference, in addition to the first and in principle determining differentia specified,” and therefore
more and more a material difference between productive and unproductive labourers, inasmuch as the former, apart from minor exceptions, will exclusively produce commodities, while the latter, with minor exceptions, will perform only personal services. Hence, the former class will produce immediate material wealth consisting of commodities, all commodities except those which consist of labour capacity itself. (Marx and Engels 1989, 17; italics in the original)
At that time, the use value produced was “never an immediate object of consumption, but a bearer of exchange value” (Marx and Engels 1989, 15). Against this background, from a capitalist perspective, productive labour was distinguished from other types using both the “form and historical” and material criteria, if we add the limiting word “generally”—as Smith had done.
Therefore, the particular background for the materialist interpretation of Smith’s second criterion can be summarised as in Table 3.
When perceived thus, the practical differences and the contradiction between Smith’s first and second criteria disappear. This was the reason Smith mentioned his supplementary criterion in addition to his primary one.
However, due to a failure to accurately consider Smith’s sociohistorical background, most of his successors could not distinguish his first criterion from his second one. This led them to stick to the materialist interpretation, which should be considered a misinterpretation of Smith’s arguments.
The failure of those economists to recognise Smith’s prior concern about distinguishing between capitalist and other competing modes of production is exactly why Marx coined the term “vulgar economics.” However, we shall point out that it is the unprecedented rapid development of the capitalist economy that made it hard for economists to catch up with Smith’s ideas, though they are de facto Smith’s successors, for they adhere to the same ideology of defending the capitalist mode of production.
Classical Definition of Service: Its Materialistic Interpretation and the Expansion of Capitalistic Productive Labour
The unprecedented expansion of the capitalist economy finally led to the subsumption of personal services by capital. This historical movement began during Marx’s lifetime and peaked in recent times. This might be the reason why mainstream economists and some of their Marxist critics commonly assume that the category of “services” must be defined in contrast to that of goods.
Obviously, one has no reason to classify service labour as unproductive while rendering goods-producing labour as productive, since both forms of labour are currently maintained by capital. As a result,
as the dominion of capital extended, and in fact those spheres of production not directly related to the production of material wealth became also more and more dependent on it . . . the sycophantic UNDERLINGS of political economy felt it their duty to glorify and justify every sphere of activity by demonstrating that it was “linked” with the production of material wealth, that it was a means towards it; and they honoured everyone by making him a “productive labourer” in the “primary” sense, namely, a labourer who labours in the service of capital, is useful in one way or another to the enrichment of the capitalist, etc. (Marx and Engels 1989, 31; emphasis in the original)
Just as Smith’s assertion,
in opposition to the Physiocrats, that the NON-AGRICULTURAL, INDUSTRIAL CLASS reproduces its own wages, that is, it does after all produce a value equal to the value it consumes, and thereby “continues, at least, the existence of the stock or capital which employs it.” Hence arises, under the influence of and in contradiction to the Physiocrats, his second definition of what is “productive labour.” (Marx and Engels 1989, 19; emphasis in the original)
Modern economists manage to persuade industry capital that those labours outside the production process are “industrially necessary” or “useful” (Marx and Engels 1989, 22) so long as they are maintained by unproductive capital (capital invested outside the production process).
The sale of these to the public provides him with wages and profit. And these SERVICES which he has thus bought enable him to buy them again; that is to say, they themselves renew the fund from which they are paid for. (Marx and Engels 1989, 22; emphasis in the original)
However, when personal services were subsumed by capital, the necessity to distinguish them from material labour also disappeared. From the capitalist perspective, it is the form of labour (whether it is maintained by capital) that matters.
The materiality of human labour never becomes the main concern of capitalists unless it forms a necessary part of the valorising process of that specific capital. Naturally, the materialist interpretation of service ceased to be the main concern of modern economists.
Karl Marx—Successor of Adam Smith in the Classical Definition of Service 11
Service in General
Unlike those bourgeois or vulgar scholars who claimed themselves to be the successors of Smith, 12 Marx, who agrees with Smith on this specific topic, has a completely different understanding of service, as evident from volume I of Capital: “A service is nothing more than the useful effect [Wirkung] of a use-value, be it of a commodity, or be it of labour” (Marx and Engels 1996, 203).
Thus, Marx implies two kinds of services: the service of a commodity and the service of labour. He also spells out that service is beneficial for the use value of either commodity or labour.
According to Marx, the concept of service has little to do with the capitalist form of society because it is mainly a question of utility or use value, and not of surplus value. In contrast to the many historically specific concepts in Marx’s theory of political economy, service can be generally viewed from the perspective of the labour process and represented as an ahistorical concept. To distinguish service under this notion from the service conceived from the historical viewpoint, we call the former “service in general.” 13
To understand the conceptual implication, we look more closely at the conceptual relationship between service and useful work in the case of both commodity and labour. In other words, we must ask, “What constitutes a use value?”
Regarding the service of a commodity, Marx makes the following remark in his Contribution:
A commodity as a use-value has an eminently material function. Wheat, for example, is used as food. A machine replaces a certain amount of labour. This function, by virtue of which a commodity is a use-value, an article of consumption, may be called its service, the service it renders as a use-value. (Marx [1859] 1977, 37)
It seems clear that the service of a commodity is defined as the useful work generated either in the process of productive consumption of a machine or in the process of unproductive consumption of wheat. In other words, a machine, if it is a useful component of a production line, renders a service in its operation. Wheat, as long as it is a useful ingredient of consumption, renders a service in the digestive process. It is important to note that the concept of service is mentioned only in reference to consumption (irrespective of whether it is productive). This type of service and its conceptual implications have been overlooked by many Marxist economists. Accepting the conventional dichotomy of goods and services, they dismiss Marx’s remark that even goods could render their own services.
However, referring to another kind of service, namely, service of the labour, Marx states in Result: 14 “In general, we may say that service is merely an expression for the particular use value of labour where the latter is not useful as an article, but as an activity” (Marx [1867] 1990, 1047).
More specifically, in Theories of Surplus Value, he states:
Where the direct exchange of money for labour takes place without the latter producing capital, where it is therefore not productive labour, it is bought as [a] service, which in general is nothing but a term for the particular use-value which the labour provides, like any other commodity; it is however a specific term for the particular use-value of labour in so far as it does not render service in the form of a thing, but in the form of an activity, which however in no way distinguishes it for example from a machine, for instance, a clock. (Marx 1963, 403–404)
Marx seems to insist that labour renders its service “not as an article, but as an activity” and not “in the form of a thing, but in the form of an activity.” If we stick to these passages as the only clue as to what service may mean, we may be fully but wrongly convinced that Marx shares no difference with the misinterpretation of Smith’s understanding of service. However, we must be more cautious in interpreting these passages, as Marx presents only a part of his arguments here.
Another aspect of his idea can be seen in Grundrisse, where Marx distinguishes the case of “service in kind [Naturaldienst], or service objectified in a thing” (Marx 1973, 467), from “the case of personal services, [where] this use-value is consumed as such without making the transition from the form of movement [Bewegung] into the form of the object [Sache]” (466). We can understand the former case historically; according to Marx, the former case arises “for example, when the peasant takes a wandering tailor, of the kind that existed in times past, into his house, and gives him the material to make clothes with,” while the latter case arises when “I give money to a doctor to patch up my health” (465).
This characterisation is briefly recapitulated in Theories of Surplus Value:
Certain services, or the use-values, resulting from certain forms of activity or labour are embodied in commodities; others, on the contrary, leave no tangible result existing apart from the persons themselves who perform them; in other words, their result is not a vendable commodity. (Marx 1963, 405)
Thus, it is clear that any labour (as far as it holds a use value) could render two different types of services: service in kind and personal service. The former is a service embodied in a commodity, while the latter is consumed as a form of movement.
Here, it is important to note that any labour can render its own service. The service of living labour can manifest in the process of its consumption as a productive activity and not as a result of the activity.
In summary, we can define service in general as any form of useful work [Wirkung] of a use value, which represents itself in its consumption and not in the result but in the process itself.
Service in Particular
We now examine the same issue from another angle, that is, the connotation of service in its social form. Therefore, the question is: what kind of social relations are needed for labour and commodities to be consumed only as service or use value? This question is partly answered in the following passage: “If A exchanges a value or money [i.e. objectified labour] in order to obtain a service from B [i.e. living labour] then this can belong within the relation of simple circulation” (Marx 1973, 465).
In such a simple circulation, Marx states:
The money which A here exchanges for living labour—service in kind, or service objectified in a thing—is not capital but revenue, money as a medium of circulation in order to obtain use-value, money in which the form of value is posited as merely vanishing, not money which preserves and realizes itself as such through the acquisition of labour. Exchange of money as revenue, as a mere medium of circulation, for labour, can never posit money as capital, nor, therefore, labour as wage labour in the economic sense. (Marx 1973, 467)
Marx stresses here upon the marked contrast between simple circulation, in which service labour has its own place, and the capitalist situation, in which wage labour is employed. The simple purchase of labour services with money is not the main concern of those who aim to exploit the surplus from labour power by placing it under control.
Clearly, from this stage, the context of the argument has changed. It is no longer a question of the simple expenditure of labour in the labour process, nor, for that matter, do Marx or Adam Smith mean to imply that service labour is only that which results in a socially useful effect. Rather, since the production of surplus value is the basis of capitalist production, only labour that produces or brings surplus value can be regarded as wage labour from the viewpoint of that mode of production. That labour which remains outside the grasp of the capitalist mode of production must be otherwise called service labour. Then, service labour can be defined as labour that is exchanged for revenue rather than capital. By contrast, wage labour is employed only in the expectation of surplus value rather than use value but can only be exchanged for capital and not for revenue. Thus, the same labour could be either service or wage labour, depending upon the manner and way in which it is employed. In sum, “service” under Smith’s first criterion, 15 or “labour as service” in Marx’s terminology is the unproductive social form in which labour is exchanged for revenue.
This is another point where Smith and Marx’s view of service has been dismissed. Therefore, it is quite important to clearly distinguish between service labour and wage labour, even though the former can produce a use value as well, while the latter, so far as it is useful, might render a service in its realisation of potential abilities.
Marx criticises this misinterpretation of Smith’s first criterion on PUPL:
Instead of speaking of wage-labour, the term “service” is used. This word again omits the specific characteristic[s] of wage-labour and of its use—namely, that it increases the value of the commodities against which it is exchanged, that it creates surplus-value and in doing so, it disregards the specific relationship through which money and commodities are transformed into capital. “Service” is labour seen only as use-value (which is a side issue in capitalist production) just as the term “productions” fails to express the essence of commodity and its inherent contradiction. (Marx 1968, 501)
Here, Marx is clearly criticising J. B. Say’s vulgarised view and his misinterpretation of Adam Smith. Say states nothing about capitalist relations but much about simple circulation, nothing about the secret of the production of surplus value but much about the consumption of use value. Marx stresses the relevance of this distinction in the following passage: “The distinction made between labourers who live on capital and those who live on revenue is concerned with the form of labour. It expresses the whole difference between a capitalist and a non-capitalist mode of production” (Marx 1971, 432).
This distinction between service labour and wage labour is identical to that made between PUPL from the point of view of capitalist production.
Productive labour, in its meaning for capitalist production, is wage labour which, exchanged for a variable part of the capital, reproduces not only this part of the capital (or the value of its own labour power), but in addition produces surplus-value for the capitalist. (Marx 1963, 152)
Further:
This also establishes absolutely what unproductive labour is. It is labour which is not exchanged for capital, but directly for revenue, that is, for wages or profit (including, of course, the various categories of those which share as co-partners in the capitalist’s profit, such as interest and rent). (Marx 1963, 157)
To summarise, while labour being exchanged for capital, that is, wage labour, brings forth surplus value and is productive of capital, labour being exchanged for revenue (i.e. service labour) is consumed merely as use value rendering a service and is therefore unproductive of capital.
Theoretical Consistency between Tertiary Activities and the Theory of PUPL
Lessons from Classical Texts
Before we verify the theoretical consistency between tertiary activities and the theory of PUPL, which is crucial for understanding “service economy,” several lessons can be drawn from the classical texts of Smith and Marx in the first part of this article.
First, the misinterpretation of Smith’s ideas based on Ch. 3 of Book 2 can be resolved if we broaden our horizons to other chapters—we will find that the classical term of service labour is consistent with his first or prior criterion on productive labour.
Second, the particular form of labour rendered by Marx as “service in things” can also be confirmed in Smith’s narrative when the latter refers to facts on economic history. The particular category of “service in things” is coherent with Smith’s first or prior criterion, while contradicting the misinterpretation of materialism with his second or supplementary criterion. Meanwhile, rediscovering Marx’s historical interpretation of Smith’s first or prior criterion allows us to understand Smith’s considerations better.
Third, the criterion for service labour in the classical term has nothing to do with the materiality of the labour process or the durability of its results. This concept was originally used to describe a unique social relationship between labour and its buyers and, therefore, shall never be related to the discussion of industry classification.
Fourth, service labour in classical terms had theoretical necessity from the specific sociohistorical background, where pre-capitalistic social relations (“service relations”) paralleled and competed with capitalistic ones. However, with the “service in things” subsumed into capital’s domination, distinguishing material labour from non-material labour became ideologically unnecessary from the standpoint of capitalists, leading to criticisms of Smith’s second or supplementary criterion, while his first criterion remained rediscovered.
Fifth, service labour in classical terms was regarded as opposing capitalistic productive labour by both Smith and Marx. This “form and historical” interpretation of service labour has nothing to do with the modern understanding of “service as an industry,” based on the misinterpretation of Smith’s second or supplementary criterion. Thus, we must condemn some Marxists for misusing Marx’s (and in fact Smith’s) theories on productive labour and unproductive labour in the context of the tertiary industry, irrespective of whether they grant the activities of the tertiary industry as productive or value-producing.
Concerns of Smith or Marx regarding the modern category of tertiary activities, that is, labour maintained by capital (and therefore productive under Smith’s first criterion) while being placed outside the production process, shall be examined in another article.
Labour Productive of Value and Labour Productive of Capital
It follows that wage labour is the only possible form of labour that is productive of capital, since it brings surplus value to capitalists. However, not all wage labourers that can be mobilised by the capitalist accumulation strategy are productive of value. Some kinds of wage labourers, although they are necessary for the capitalist economy as a whole, cannot create value or surplus value, since they perform unproductive functions such as commerce, finance, insurance, marketing, advertising, and speculation. 16 Their functions contribute only to the realisation, transformation, and appropriation of the value that has already been created by productive labourers.
Of course, they are, as Marx refers to commercial wage labour, “directly productive” for the capitalist who employs them, because “their unpaid labour, even though it does not create surplus value, does facilitate his ability to appropriate surplus value, which as far as this capital[ist] is concerned, gives exactly the same result, that is, it is its source of profit” (Marx [1867] 1990, 407).
All these forms of commercial and financial labour, which add nothing to the value of the commodities they handle, are to be classified as wage labour unproductive of value. As Marx states, “What he brings in is the function, not of any direct creation of surplus value, but of his assistance in reducing the cost of realising surplus value, in so far as he performs labour (part of it unpaid)” (Marx [1867] 1990, 414). Therefore, as to the concept of productive labour, it is necessary to distinguish labour productive of value (that is, the value-creating character of labour) from labour productive of capital (i.e. the general form of wage labour).
A further question rises here: “Which kind of labour is value-creating”? In other words, what conditions are necessary for wage labour (labour generally productive of capital) to be productive of value? To mark the boundary of value-creating labour, it is necessary to tackle it from two different perspectives: the division of labour inside and outside the factory.
With the development of the capitalist economy, the division of labour within the factory has become so complex that each worker can now be considered productive only in the sense that different jobs and various functions collectively produce a use value as the final product. The totality or cooperative nature of these workers, which Marx refers to as the collective labourer, makes the concept of value-creating labour even larger and more extensive than before. As individuals, they do not need to put their hands on the object (Marx [1867] 1990, 643–644; see also Marx 1963, 411–412).
Thus, nowadays, large numbers of overseers, engineers, directors, and so on, who have nothing to do with the work-up of raw materials, are included in the category of labour productive of value, so long as they engage in producing the common product collectively. In so far as each of them represents an organ or a constituent part of the collective labourer, it is no longer necessary for the result of their activity to be materialised directly in tangible goods.
The ongoing division of labour within the factory could widen the concept of value-creating labour. The notion of the social division of labour tends to make the boundary between two different types of labour rigidly fixed by applying the concept of value-creative labour to the sphere of material production.
In summary, by following Marx, we can demolish the flimsy analytics of conventional service theory, which focuses its attention merely on the tangible outcome of the labour process. In addition, we can determine the difference between the labour productive of capital and the labour productive of value. The former is generally represented as the form of wage labour, while the latter, being a subset of the former, can be defined as wage labour that is not only exchanged for capital but is also useful as an organ or a member of the collective labourer in such an industrial sphere of material production.
The Materiality of Production and Material Production
However, the category “materiality of production” shall not be rendered as being “some particular subject or vendable commodity, which lasts for some time at least after that labour is past” (Smith 1977, 438). As Marx states:
The materialization, etc., of labour is, however, not to be taken in such a Scottish sense as Adam Smith conceives it. When we speak of the commodity as a materialization of labour—in the sense of its exchange-value—this itself is only an imaginary, that is to say, a purely social mode of existence of the commodity which has nothing to do with its corporal reality; it is conceived as a definite quantity of social labour or of money. It may be that the concrete labour whose result it is leaves no trace in it. (Marx 1963, 171)
In this particular connotation of materiality, Marx and Smith diverged. In material production, all that useful labour can do is merely change the form of materials. As per Smith, it can neither create them nor make them “vanish.” Therefore, it is necessary to note that the materiality of the production process is not the physical nature of the product. As Marx states:
In addition to extractive industry, agriculture and manufacturing, there exists yet a fourth sphere of material production, which also passes through the various stages of handicraft industry, manufacturing, and mechanical industry; this is the transport industry, transporting either people or commodities. (Marx 1963, 412)
Thus, materiality does not lie in the result but in the particular way of change―that is, the chemical, physical, and biological modification, in which the natural scientific law dominates. For example, what the transport industry sells is the actual change of place, and this spatial change is the result of the production process accomplished by the useful labour of transport workers. In the same sense, the biological change of crops constitutes the main part of agriculture, and the manufacturing industry can make use of the physical and chemical change processes to meet its technological and engineering demands. Thus, we can conclude that a spatial change process characterises the materiality of the transport industry. Marx’s reference to transport labour, therefore, makes clear that the connotation of value-creative labour has nothing to do with the natural form of labour or whether it has left a visible trace in its result, and much to do with the nature of the productive process itself.
We can summarise some of the necessary economic characteristics as criteria for the material production process or the materiality of production.
One of the possible criteria has been mentioned by Adam Smith: whether labour “fixes and realizes itself in some particular subject or vendable commodity, which lasts for some time at least after that labour is past” (Smith 1977, 438). However, under Marx’s connotation, some non-material labour such as calligraphy, painting, or writing could leave their traces on “some particular subject or vendable commodity,” while a clear sort of material labour
whose result it is leaves no trace in it . . . In agriculture, etc., although the form given to the commodity, for example wheat or oxen and so on, is also the product of human labour, and indeed of labour transmitted and added to from generation to generation, yet this is not evident in the product. In other forms of industrial labour, the purpose of the labour is not at all to alter the form of the thing [die Form des Dings], but only its position [Ortsbestimmung] . . . Therefore, the materialisation of labour in the commodity must not be understood in that way. (Marx and Engels 1989, 27)
This dilemma discloses the sterility of the “Scottish sense” (on the materiality of production) in economic analysis, for whether labour “fixes and realizes itself in some particular subject or vendable commodity” has nothing to do with the social form of labour, and therefore does not affect the difficulty with which the specific labour is subsumed by capital.
Historically, the capitalistic mode of production has been related to the special characteristic of material labour, which promises a certain amount of labour fruits in a certain period. 17 Namely, if we keep the level of productive forces and the proficiency level of labour skill constant, the production time of a specific product of material labour should be consistent with its reproduction time. Therefore, a correlation should be confirmed between the fruits of labour and labour invested. This correlation is one of the economic characteristics of materiality.
Conversely, even if we hold the level of productive forces constant, the original production time for non-material production is inconsistent with its reproduction time. For instance,
The product of mental labour—science—always stands far below its value, because the labour time needed to reproduce it, has no relation at all to the labour time required for its original production. For example, a schoolboy can learn the binomial theorem in an hour. (Marx and Engels 1994, 87)
Meanwhile, the correlation between capital invested and labour productivity could be rendered as another economic characteristic of materiality. Economies of scale, or more accurately improvement of labour productivity as a result of raising the organic/technical composition of capital, generally works for material production. Therefore, we cannot measure the productivity of non-material labour based on its technical composition.
The “cost disease” of personal services shall be seen as the embodiment of this law. As Baumol (1986) notices, the convergence of national productivity levels in developed countries has become increasingly obvious since the 1970s. Some scholars may conclude that this particular trend is correlated to the so-called “service economy”. However, as the originator of the “theory of cost disease,” Baumol’s review of Fuchs and Leveson’s book The Service Economy points out the “dispersion within the [service] sector” (Baumol 1971, 301) which could result in the dispersion of labour productivity.
Among those “stagnant services,” Baumol (1993, 624) pays particular attention to education and medicine. Labour processes in these departments “are inherently resistant to standardization,” and we could not expect an equal amount of quality if we substitute that non-material labour with machines (625). These are economic characteristics of non-materiality.
By contrast, the fourth material production department, that of transportation, “also passes through the different stages of the handicraft system, the system of manufacture, and mechanized industry” (Marx and Engels 1994, 245) by raising its organic or technical composition of capital, even though transportation generally does not fix or “realize itself in some particular subject or vendable commodity, which lasts for some time at least after that labour is past.”
In general, two characteristics of material production are crucial for economic analysis: the correlation between labour invested and its fruits, and the correlation between capital invested and labour productivity. The question is whether service labour is part of material production.
If we refer to service labour in classical terms, that is, labour as service, clearly this conception is only related to the social form of labour and has nothing to do with its materiality. In addition, labour as a service is maintained by revenue and, therefore, is consumed unproductively.
However, if “service” is referred to as an industry, that is, labour in the tertiary industry, the question becomes one of industry classification and has nothing to do with Marx’s category of service or service labour.
The originators and advocates of the term tertiary industry defined the term negatively, that is, industries that do not belong to the primary and secondary industries shall be rendered as tertiary or service industries (Fisher 1952, 824). Indeed, one cannot tell whether an activity is material or whether it belongs to the production process from such a negatively defined term. We could find labour in repairing and transportation to be part of material production; we could equally find labour in wholesale or retail outside of social production, and labour in science or art immaterial. Generally, “service as an industry” is not a proper category for discussing material production.
Is “Service Labour” in the Tertiary Industry Productive of Value?
Most of the arguments on PUPL highlight the problem of whether service in the modern term is productive of value (Haruo 1989, especially Ch. 5 and 6). This could be attributed to the growing importance of the tertiary industry in the national income.
However, Bauer and Yamey state:
The shift of many economic activities from the unpaid to the paid category and the relative growth in importance of the money economy in the course of economic development often exaggerate the growth of national income as conventionally measured. For much the same reason the conventional national income calculations tend to exaggerate the discrepancy in incomes between more developed and less-developed countries . . . particularly in the earlier phases, there tends to be a marked shift of activity (and consumption) from the unpaid to the paid categories. It is highly probable that the inclusion of unpaid activities in the reckoning (a matter involving rough estimates) would reduce the rate of growth of national income, and, even more, the rate of growth of tertiary activity. (Bauer and Yamey 1954, 105; Clark 1949)
We also find it true that some of the tertiary activities in developed countries are usually classified as a part of primary or secondary activities in developing countries (Bauer and Yamey 1951, 742–744, 753–754). This is because such activities are yet to be independently developed into an industry or become part of wage labour and, thus, labour productive of capital. Therefore, the growing importance of tertiary activities or “services” in the national income could be analysed from the perspective of the subsumption of labour by capital.
Accordingly, a growing number of advocates are urging Marxist economists to treat tertiary activities or non-material activities as labour productive of value; they are, in fact, considering all labour productive of capital (i.e. wage labour) also as productive of value.
However, it is clear from its negative connotation that tertiary activities cannot be rendered purely non-material nor can they be classified into the sphere of social production. Furthermore, even if tertiary activities are made up of wage labour or labour productive of capital, note that not every wage labour is productive of value; the tertiary activities outside the material production shall not be viewed as labour productive of value.
Conclusion
The preceding references to Smith and Marx’s text show us that their concepts of service are completely different from the conventional one. Their concepts (namely, “service labour in the classical term”), although they originate from the concept of “service relation,” can be identified in contemporary capitalist societies if we consider interests, rents, wages, and profits as the various forms of “revenues.”
In fact,
the same labour can be productive when I buy it as a capitalist, as a producer, in order to valorise it, and unproductive when I buy it as a consumer, a spender of revenue, in order to consume its use value, no matter whether this use value perishes with the activity of the labour capacity itself or materialises and fixes itself in an object. (Marx and Engels 1989, 20–21)
And thus, all labour could, from the perspective of consumers, be rendered as a service (in the classical sense).
This renewed interpretation of service should not be confused with the materialist misinterpretation discussed previously or the natural successors of that misinterpretation, the advocates of “tertiary industry,” who roughly regard labour in the tertiary industry as non-material (Clark 1940, 182).
Considering the strong evidence shared by Bauer and Yamey (1951, 1954), we have to reject Fisher’s (1952, 822) and Clark’s (1940, 182) theses on the necessity for applying the concept of “tertiary industry” in the case of significant growth of non-material labour in developed countries. By contrast, significant changes should be attributed to the form of labour.
If Adam Smith were alive today, he would certainly grant the activities in the tertiary industry for productive labour under his primary criterion, while being confused with the partial judgements on his second or complementary criterion given by materialists. 18 This dramatic result has several socio-economic implications. The most important one is that it can lead us to a fundamental criticism of the conventional post-industrial argument, which tries to persuade us that we have entered a new era, an era that can be termed as the “service” society.
According to the post-industrial argument, the “service” society is characterised by the fact that most of its employment is in the “service” sector. However, these sectors have different activities. Furthermore, the constituent parts of these sectors are much less homogeneous in terms of their socio-economic functions. Noticeably absent from any discussions along the line of the conventional post-industrial thesis is a clear notion or functional definition of service. Disaggregation of this sector into subdivisions is necessary. Furthermore, functional identification of each activity is needed.
From Marx’s critical viewpoint, the conventional notion and theory of “service” should be dismissed owing to their own misconceived nature. This does not mean a mere replacement of the conventional concept with Marx’s, as is often understood by some Japanese Marxist economists who believe in the post-industrial hypothesis. As stated, the purported new “service” society is not found in Marx’s concepts of service and service labour. In other words, the dramatic change within and outside the advanced capitalist societies, which seemingly indicates the advent of “service” society, can neither be grasped by conventional “service” theory nor by Marx’s.
Rather, as Walker (1985) stresses, the problem posed by post-industrialism should be tackled within the general framework of the industrial capitalist system. Unless we view the concept of service from this perspective, we will fail to determine the structural consistency that causes such a change.
