Sociological dimensions in classical/neoclassical economics: Conceptions of social economics and economic sociology

The proposal for social economy as economic science

First and foremost, classical and neoclassical economic theory contains certain explicit conceptions, i.e. projects and proposals, of social economics and to that extent of economic sociology as it is understood. In addition, it entails implicit conceptions or latent elements of social economics in that the idea of the latter is substantively implied, although the term is not formally used. In that sense and degree, classical political economy and neoclassical economics also at least in part represent what Knight (1958) and other economists (Lewin, 1996; Solow, 1990) call ‘sociological economics’, and not just pure market economics. This is a non-trivial and consequential statement and finding so long as ‘purist’ economists define and reconstruct economic science in exactly opposite terms, as an application of mathematical physics or mechanics to economy (Mirowski, 1989), and therefore anti- or non-sociological economics (Lewin, 1996).

Probably the first explicit conception or proposal of social economics is found in J.B. Say’s classical economic theory, the earliest and, as Ricardo remarks, most important European follower of Smith. Thus, some analysts identify and emphasize Say’s ‘social economics’ (Forget, 1999; see also Swedberg, 1987) in the sense of a science entailed and incorporated within his economic theory epitomized by what economists call ‘Say’s law of [equilibrium] markets’ (Baumol, 1999; Samuelson, 1997). First and foremost, Say (1852) proposes that ‘social economy’ is the ‘better’ conception or term for economic science than is the ‘name of political economy’. He does so on the grounds that the subject of ‘political economy’ is what he significantly calls the ‘economy of the society’ or societies analyzed in relation to ‘different parts of the social body’ (Say, 1852). ⁴ Also, following Smith and anticipating J.S. Mill, Say considers ‘political economy’, and thus economic science, to be part of social or political science. Say thus implies a clear and consistent differentiation between various levels of scientific analysis, specifically between the general level and comprehensive analysis of social science as a whole and their particular and delimited forms in economics as its integral element. Such a distinction between various levels of analysis also characterizes Walras’ division of economic science into pure, applied, and social economics, as considered later. ⁵

In this connection, Say (1972) defines social/political science as the general ‘science of the organization of societies’ (la science de l’organisation des sociétés), thus essentially equivalent to Comte’s sociology. In turn, he defines political economy as the specific study of the formation, distribution, and consumption of wealth which satisfies the ‘needs of societies’, briefly the ‘simple exposition of the laws governing the economy of societies’ (‘l’économie des sociétés). Say implies that the ‘organization of societies’ logically and empirically incorporates the ‘economy of societies’ as its integral element and hence ‘sociology’, while not using the term, involves economics, known as political economy, until Marshall in 1890. Evidently, the ‘economy of societies’, like its equivalent ‘material forces of society’, forms the substantive ‘proof’ or ground of Say’s social economics by suggesting that economy only emerges, exists, and operates within society, while being specified by such expressions as ‘capital of society’, ‘social wealth’, and the like. In short, economy, with various economic actors, actions, and interests, belongs to and exists in what he calls the ‘social body’ or ‘social order’ (see also Forget, 1999).

As a corollary, the ‘economy of societies’, with the various specifications mentioned, posits and emphasizes what Say’s inspiration Smith implies and Schumpeter suggests to be the societal setting of economic actors and actions. Alternatively, the notion of the ‘economy of societies’ implies rejecting as a logical non-sequitur or historical fiction the model of an asocial, isolated ‘Robinson Crusoe economy’, which remains prominent in contemporary pure economics (Conlisk, 1996) through the expression ‘simplest “Robinson Crusoe” economies’ and other terms. Also, the ‘economy of societies’, including its specifications such as ‘social wealth’, implies that society, including politics, shapes economic processes and outcomes, as Say acknowledges in observing that ‘social circumstances’ and ‘governmental action’ are ‘favorable’ or ‘harmful’ to the production of wealth. Specifically, Say (1972) recognizes that ‘governmental action’ or the state often provides a ‘powerful stimulus to individual productive energy’ through activities such as ‘well-planned public works (roads, canals)’ – ironically, anticipating his major critic Keynes and perhaps FDR and the New Deal – while rarely being a ‘successful producer’, which in turn anticipates Walras’ (1926) assertion that the ‘state is not an entrepreneur’. Also, seemingly sharing or expressing, like Smith, the liberal-democratic project of the Enlightenment, Say suggests that political and, by implication, civil liberty has a positive impact on economy and society alike, including education. Thus, he observes that political liberty is ‘more favorable’ (although ‘indirectly’) to the production of wealth, just as it is ‘more favorable’ to education (Say, 1972).

Next, the ‘economy of societies’ also operates or appears as a sort of stage of class stratification and inter-class emulation and imitation, which is another element of Say’s social economics. Say (1972) registers the existence of class stratification by identifying the ‘different classes of society’ (les diverses classes de la société), including the middle class, working or low classes in some industries (plus courtisanes), and, by implication, the upper class. He notes inter-class emulation, envy, and imitation, observing that society is divided into a small upper class providing ‘pleasures sought’ in society and large lower classes showing envy for and attempting to emulate or imitate the first and to move from ‘one class to another’ using virtually any means to ‘enrich’ themselves (Say, 1972). In retrospect, Say thereby anticipates Veblenian economic institutionalism and Durkheimian ‘positive’ economists (e.g. Halbwachs) in the analysis of material consumption, especially of its ostentatious or conspicuous form, as ‘social process’ (Gislain & Steiner, 1999). ⁶

Generally, Say implies that the economy, in virtue of being l’économie des sociétés, is, besides being evidently Comte’s ‘social economy’, what Durkheim would call an ‘economic society’ (like ‘political society’), and what Pareto and Parsons would term a particular or partial social system. Conversely, Say, like Comte and Durkheim, does not conceive or depict society, including politics and culture, as being the mere extension or reflection of the economy, notably an overarching marketplace, as does ‘rational choice theory’ and also orthodox Marxism with its ‘economic interpretation of history’, though, in contrast to the latter’s microscopic, individualistic thrust, at a macroscopic structural level. En passant, this is indicated by some proposals for reconciling the two theories into ‘rational choice Marxism’ (Elster, 1990; see also Coleman, 1990).

Hence, Say’s ‘political economy’ has a stronger convergence and affinity with social economics or economic sociology than, if ever, with the ‘rational choice model’ as, in his term, the generalization to the entire ‘social body’ of the ‘law of markets’ that aggregate supply ‘creates its own demand’ (Gide & Riste, 1947; Schumpeter, 1954; yet Baumol, 1999) resulting in necessary, stable equilibrium (Keynes, 1936; ⁷ Samuelson, 1997). Regardless of its (in)validity, however, Say does not attempt or propose to generalize the ‘law of markets’, based on the supply, demand, and prices of material goods and services, beyond the economy into the rest of society construed as a system of ‘implicit markets’ involving the ‘supply’ and ‘demand’ of non-material ‘invaluable goods’ (Arrow, 1997) à la children and spouses as ‘commodities’ with ‘imputed prices’. To that extent, he does not construct or propose the ‘rational choice model’. By such an inaction or omission, Say implies that the ‘law of markets’ simply holds only where actual markets exist and operate, as in the (market) economy, and alternatively it does not hold where they do not exist and operate, as in the non-economic fields of society. So do, following directly or indirectly Say, virtually all the major classical and neoclassical economists considered, from Mill to Walras and Pareto (with the secondary exception of Edgeworth). In this sense, Say’s seemingly surprising, for rational choice theorists, inaction or abstention from the extra-economic generalization of his ‘law of markets’ is what Weber would call the ‘best course of action’ within his economic theory, since none of these classical and neoclassical economists (minus Edgeworth) before or after generalizes such laws beyond their proper market domain. In sum, Say’s economic theory entails the explicit conception and substantive elements of ‘social economy’ combined or coexisting with market economics based on the ‘law of markets’, but no ‘rational choice model’ thus understood.

In another related pertinent instance, J.S. Mill presents and develops an explicit conception and substantive elements of social economics, which he terms the ‘science of social economy’. First, in so doing, he follows and elaborates on Say’s idea of ‘social economy’ (Swedberg, 1998), and is also, as his neoclassical follower Marshall (1961) remarks, greatly influenced by Comtean sociological theory and method (Robert, 2002). For instance, a recent analysis shows that Mill’s ‘science of social economy’, and generally the evolution of his epistemology and methodology, can be ‘explained by Comte’s [sociological] influence’ (Robert, 2002). ⁸

Crucially, Mill adopts or evokes Say’s ‘social economy’ as economic science and Comte’s notion of the ‘social economy’ as a social system. Also, Mill embraces Comtean general sociological notions of ‘social statics’ and ‘social dynamics’ and, as Schumpeter (1954) comments, introduces them into political economy as ‘economic statics’ and ‘economic dynamics’ to the displeasure of ‘pure’ anti-sociological economists like Hayek (1950). ⁹ Consequently, Mill’s ‘science of social economy’ is classical political economy’s version of and another name for economic sociology – not yet so called until, ironically, his sharp critic, neoclassical economist Jevons, during the 1870s (Swedberg, 1998) – specifically its Comtean embryonic version or proxy, and even, judging by its domain, of Comtean general sociology. Schumpeter (1950) estimates that ‘one-third’ of Mill’s major work Principles of Political Economy ¹⁰ is devoted to ‘economic sociology’ in the substantive sense of the theory and analysis of social institutions in terms of their impact on the economy (see also Gislain & Steiner, 1995, 1999; Smelser & Swedberg, 2005; Swedberg, 1987; Trigilia, 2002). Similarly, a recent analysis (Robert, 2002) suggests that Mill’s Principles shows that the ‘explicit recognition of the role of a General Science of Society’ à la Comte’s sociology is accompanied by an opening on ‘cultural, historical, and sociological considerations’ in economics itself. In this connection, Mill is described as ‘a herald of social economics’ (Jensen, 1996) and, to that extent, of economic sociology as understood (Swedberg, 1998) in classical political economy and generally traditional economics, and in that sense his economic theory has elements of sociological economics.

Furthermore, Mill (1968) defines the subject matter of the ‘science of social economy’ almost à la Comte’s definition of that of general sociology, the ‘conduct or condition of man in society’, ¹¹ especially, by implication, the impact of the latter on economic action, thus including economic sociology analyzing societal influences and elements in the economy. As a corollary, Mill considers the ‘science of social economy’ to be virtually identical to Comte’s science of society, incorporating ‘political economy’ – defined à la Say and Ricardo as the study of the ‘production and distribution of wealth’ – including market catallactics – defined as the ‘science of exchanges’ – as ‘a branch’. Alternatively, Mill suggests that the ‘science of social economy’ thus understood is not to be equated to and named – as Say in his view did ¹² – ‘political economy’ as the ‘large extension of the signification’ of this term, because the latter is ‘not the science of [society or politics], but a branch of that science’, alongside the ‘other branches’ of sociology or ‘social philosophy’ à la Comte (Robert, 2002).

Mill’s ‘science of social economy’ or ‘economic sociology’, in Schumpeter’s sense of a sociological-institutional analysis of the economy, involves certain substantive elements. One of these elements is the impact of social institutions such as traditions and legal rules on the economy, including the market, especially during early periods. Mill (1884) remarks that during these periods ‘all’ market processes are under the ‘influence of fixed customs’ so that only during modern times has competition functioned as the ‘governing principle of contracts’ and even its operation can be limited by ‘law and custom’, ¹³ as some contemporary economists also recognize (Young & Burke, 2001). Another related substantive element of Mill’s ‘science of social economy’ or ‘economic sociology’ is the influence of social institutions on wealth distribution. Mill (1884) observes that the distribution of wealth is the problem of ‘human institution’ to the effect of being dependent upon the ‘laws and customs of society’, even acknowledging, almost like Marx or Weber, that the rules determining this seemingly objective market process express ‘the opinions and feelings of the ruling portion of the community’ (see also Acemoglu, 2005; Acemoglu & Robinson, 2008). An additional element of Mill’s ‘science of social economy’ is what sociologically minded economists and economic sociologists consider the process of social-cultural determination or formation of individual tastes or preferences and motivations (Akerlof, 2007; DiMaggio, 2005; Frank, 1990; Prendergast, 2008). Mill suggests this process by observing that social actions are in reality the ‘result of a plurality of motives’ – and not only the ‘desire of wealth’ – notably of those motivations as ‘modified by the social state’.

Next, and perhaps most surprising and ‘disappointing’ to rational choice theorists, is that Mill holds neither an explicit nor implicit variant of their model, the all-embracing economic approach to social phenomena. Moreover, he effectively rejects or rules out such an approach as a sort of non-sequitur or pretension. Thus, he admonishes that ‘political economy’ as economic science ‘does not pretend’ (sic) that its premises and inferences are ‘applicable’ to ‘those parts of human conduct of which wealth is not the principal object’, thus to the realms of society beyond that of the market and economy (Mill, 1968; see also Cournot, ¹⁴ 1960).

Last, recall he suggests that ‘political economy’ is ‘not the science of [all society], but just a branch of that science’. Comtean sociology thus does not apply its theoretical and methodological approach to all social action and structure, nor expand, as the economics of society, ‘rational choice theory’. Mill thereby follows Say and Comte in differentiating various levels of social scientific analysis, the most general and comprehensive level of sociology or ‘social economy’, from the specific and limited level of ‘political economy’ or economics, and anticipates (though likely does not influence) Walras (doubtless more influenced by his compatriots), who also makes such a distinction.

Walras’ neoclassical economic theory cum marginalism involves yet another explicit conception of social economics, perhaps as surprising or unexpected, at least to those economists and rational choice theorists knowing or attentive only to Walrasian ‘pure’ market economics, as that of Jevons, considered below. Leon Walras, another marginalist pioneer and even according to Schumpeter (1954) the ‘greatest of all economists’, worked not only in pure market economics (his main and best-known book being Elements of Pure Political Economy) but also in applied economics (Applied Political Economy). ¹⁵ Moreover, he devotes a special book to social economics, entitled Studies of Social Economy (Walras, 1936a; see also Sandmo, 2007), which is at first glance as incredible and yet true as Jevons inventing the term ‘economic sociology’ (Swedberg, 1998). Hence, ‘social economy’ in the sense, like ‘pure political economy’, of science à la Say casts doubt on what analysts see as the widespread ‘misperception of Walras’ as simply a ‘pure’ narrow neoclassical – i.e. marginalist, perfect-competition and general-equilibrium – theorist (Blaug, 2001; Schumpeter, 1956; Stigler, 1957) in contemporary economics (Burgenmeier, ¹⁶ 1994) as well as a sociological rational choice theorist (Coleman, ¹⁷ 1990). Instead, this makes Walras too, even if secondarily, a social economist.

Specifically, Walras (1936a) defines ‘social economy’ as the ‘theory of the division of social wealth’, notably economic distribution as influenced by considerations of ‘justice’ in society. And he proposes ‘social economy’ as a complementary discipline to ‘pure political economy’, defined as the ‘theory of social wealth’, notably the ‘theory of price determination under the hypothetical regime of absolutely free competition’ and laissez-faire (Walras, 1926), as well as to ‘applied political economy’, understood as the ‘theory of the production of social wealth’ (Walras, 1936b). A relevant substantive element of Walrasian ‘social economy’ that relaxes the hypothesis of perfect competition or laissez-faire of his ‘pure political economy’ is that he acknowledges that the economy cannot effectively function without what he calls ‘intervention by some authority’ in society, notably the state (also, Perroux, ¹⁸ 1960), while, like Say, denying that the latter is an entrepreneur (l’Etat n’est pas un entrepreneur) (Walras, 1926). Specifically, while extolling free competition or laissez-faire as the ‘superior and general’ rule or principle of the production of wealth, he recognizes that it is the state that institutes and maintains ‘free economic competition in society’ through ‘very complex legislation’ (Walras, 1936a), influenced by or evoking Comte’s insight into the government’s function in the ‘social economy’.

Walras substantively distinguishes market-economic from other social laws. In his view, the ‘law of economic relationships of people in society’ is utility or material interest, while that of non-economic social phenomena is moral good or justice (Walras, 1936a). This implies what contemporary sociologists critical of neoclassical economics’ purist separation of economic and social fields call ‘social logic subject to the rule of equity’ (Bourdieu, ¹⁹ 1998; see also Burgenmeier, 1994). Hence, Walras’ sharp and consistent distinction suggests that the laws of the economy, including the market, do not actually operate beyond their specific and relatively narrow realm, and therefore cannot be theoretically assumed to; and, alternatively, those of non-economic domains are autonomous forces, not to be reduced to the law of utility or material interest.

As a corollary of this distinction, like Say and Mill and his contemporary and theoretical ally Jevons, Walras constructs or implies no ‘rational choice model’, in the sense of theoretically extending the laws of the economy to all society, thus negating autonomous social non-economic forces like equity/morality or dissolving them into utility/material interest, as such a procedure is logically a non sequitur because it contradicts his initial premise and is thus inadmissible. No wonder, while doing so for social economics as well as pure (and applied) economics, Walras does not write a special book on ‘rational choice’ economics thus understood. The latter would violate or self-contradict his substantive distinction between the market-economic and the non-economic laws and realms of society, specifically utility and material interest as distinguished from equity and morality. This is a theme his successor (at Lausanne’s chair of political economy) Pareto further argues and develops in his antinomy of ‘interests’ and sentiments (‘residues’) determining rational economic and non-rational social actions, respectively. Furthermore, Walras explicitly incorporates ‘social economy’, alongside ‘pure political economy’ and ‘applied political economy’, in the disciplines of economic science, but does not even imply the ‘economics of social relations’. For social relations are viewed as being governed by ‘laws’ other than utility or material interest that instead govern only the economy. This effectively rules out the ‘rational choice model’ as an ‘impossibility theorem’ or logical contradiction from Walras’ rendition of economic science. In sum, like Say, Comte, and Mill, Walras differentiates various levels of analysis of economy in his differentiation of economic science into pure, applied, and social economics (but not the ‘rational choice’ theory of society).

A further explicit conception and substantive elements of social economics are present in Wieser’s neoclassical economic theory as part of the Austrian School of marginalist economics founded by Menger, developed also by Böhm-Bawerk, and continuing through Mises and Hayek, as well as in part Schumpeter (Stolper, 1994; Swedberg, 1998), as its ‘younger’ members. Above all, like Walras, Wieser writes a separate book on social economics as part of Weber’s interdisciplinary project, titled Foundations of Social Economics, involving also the young Schumpeter from the Austrian School as well as some members of the rival German Historical School, during the 1900s (Swedberg, 1998). Generally, according to his younger contemporary Schumpeter ²⁰ (1956), Wieser’s ‘first’ and ‘last’ scientific interest was ‘historical sociology or sociological history’, reflected in his ‘great sociological book’ on power (published when he was 74).

Specifically, echoing Walras and Jevons, and crucially influenced by Weber (Swedberg, 1998), Wieser (1967) outlines and defines social economics as an ‘inquiry into the social relations of the economy’, including the individual’s economic behavior such as utility maximizing and cost minimizing in the ‘light of his social environment’. ²¹ Social economics, or more precisely economic sociology, thus involves what he significantly describes as the ‘sociological problems of economic theory’. Wieser proposes social economics cum economic sociology on the ground that a ‘sophisticated theory of society’ is necessary for completing ‘modern economic theory’ if the latter wants to be ‘convincing’; specifically ‘sociological phenomena’ used in a ‘description of the economic process’ are also required for its ‘explanation’. Moreover, Wieser goes further than Jevons and Walras, while following or evoking Mill and Menger, by considering economics to be ‘only one phase of social science’, specifically an ‘advance guard of sociology’. Wieser grounds this vanguard status of economics within sociology, especially in the economic marginal-utility theory of exchange value as a ‘commensurable quantity’ – and notably what Schumpeter (1908–1909) calls a ‘social value’ ²² – allowing ‘more rapid and certain progress’ in this field than in ‘other sociological fields’. ²³ And, like Jevons and Walras, he divides economic science into pure market economics, or catallactics, based on the marginal-utility theory of prices, including the marginal productivity (‘imputation’) principle of wealth distribution (Wieser, 1956), and social economics as defined, but not into the economics of society as a whole, so ‘rational choice theory’.

Wicksell’s neoclassical qua marginalist economic theory also contains an explicit conception of social economics largely influenced by and elaborating on and extending that of Walras. In general, Wicksell (1934) conceives, as an extension of Walras’ definition, ‘social economy’ as a science in terms of an ‘investigation’ of the application of ‘economic laws and practical precepts’ for the purpose of attaining the ‘most possible social gain’ and of the indispensable modifications to that purpose in the ‘existing economic and legal structure of society’. In particular, like Walras, Wicksell specifies ‘social economy’ as focusing on the social conditions of the distribution of wealth or property among factors of production, notably capital and labor. Thus, à la Walras, Wicksell (1934) states that the ‘question of property rights of various factors of production’ is a social problem and hence the subject of ‘social economy’, while the ‘share of the product which goes to any particular [factor]’ is an ‘economic problem’, as determined by the rule of marginal productivity, and thus belongs to the domain of ‘pure political economy’. Wicksell (1934) implies that the ‘social problem’ and thus ‘social economy’ as science is based on what Walras calls the law of equity or morality, while the ‘economic problem’ and so ‘pure political economy’ has its basis in the law of utility or material interest. He thus adopts the Walrasian consistent dichotomy between the laws of non-economic realms of society and those of the economy.

And exactly like Walras, Wicksell classifies economic science into the trinity of ‘pure political economy’ (or market economics), ‘social economy’, and ‘applied economy’. Alternatively, he does not incorporate into this classification or envision the universal economic theory of social action and structure – so the ‘rational choice model’, though seen as one of the ‘precursors’ of public choice theory, alongside Schumpeter, etc., by some economists (Mueller, 1997; Wagner, 1997). Further, Wicksell’s less or even non-marginalist contemporary compatriot Gustav Cassel (1929) writes, like Walras and Wieser, a book entitled Theory of Social Economy (curiously translated in French as Traité d’économie politique). In particular, he assumes and analyzes ‘a rationally regulated social economy’ and ‘a social economy under dynamic conditions’ (see also Cassel, 1927). In passing, both Swedish economists influence or anticipate their unorthodox successor Gunnar Myrdal’s (1953) implied conception of social economics, indicated by the proposed ‘sociological explanation’ of market variables such as supply, demand, and prices.

The idea of economic sociology

An explicit conception of social economics under the term economic sociology can be identified in none other than Jevons’ ‘pure’ neoclassical economics, more precisely marginalism premised on marginal-utility theory, which rejects and supersedes the classical, labor–cost version. First and foremost, incredibly enough, Jevons, commonly viewed as a hard-core marginalist economist, probably coined the very term ‘economic sociology’, preceding classical sociologists (Swedberg, 1987, 1998) like Durkheim and Weber, and succeeding, in part influenced by their earlier colleagues, Comte and Spencer. That such a ‘pure’ neoclassical economist should invent the name, though not the concept, of one of the most central and vibrant branches of sociology is one of those remarkable ironies in the development of social science. In turn, this is an invention that deliberately projects sociological elements (Parsons, 1935a) onto Jevons’ economic theory hence involving social economics conjoined with his pure-market economics of marginalism. Furthermore, Jevons (1965) entertained certain ‘great rational expectations’ about what he terms a ‘branch of economic sociology’, which he proposes, alongside some other branches, as no less than a sort of savior of economics in virtue of its attributed ‘mission’ of rescuing the latter from its admittedly ‘confused state’. ²⁴ This indicates that Jevons considers economic sociology to be a branch of economic science and, by assumption, of sociology alike and thus as an interdisciplinary field or what Schumpeter (1956) calls ‘no man’s or [rather] every man’s land’, analogously perhaps to social psychology and similar disciplines.

In this connection, he defines à la Comte and Spencer by implication economic sociology as the ‘science of the evolution of social relations’ in terms of their effects on the economy, especially the market. This implies that what later sociologically minded economists call the ‘economic sociology of the market’ (Boulding, 1970) is the main specialty of Jevons’ ‘branch of economic sociology’. As a corollary, the major substantive element of Jevons’ economic sociology involves the impact of social factors like political institutions on markets. Thus, he observes that what he describes as the ‘political information of the moment’ exerts an influence on the ‘future supply and demand’ in markets and generally ‘other than strictly economical grounds’ often shapes or settles market transactions (quoted approvingly, as is virtually everything else from Jevons, by Edgeworth, 1967). ²⁵ Last, like Say and Mill – whose Ricardian labor-cost theory of value and prices, however, he vehemently rejects in favor of that of ‘final utility’ – Jevons does not explicitly propose or imply a ‘rational choice model’ in his conception and classification of economics. For instance, Jevons (1965) classifies economic science into ‘mathematical theory of economics’, ‘systematic economics’, ‘descriptive economics’, and ‘economic sociology’, alongside some other branches (‘commercial statistics’, ‘science of finance’). However, no overarching economic approach to all social phenomena is present or even implied in this classification, in contrast to contemporary ‘imperial’ economics.

An explicit conception of social economics, which is even more explicitly sociologically minded than those of Walras and Wicksell, pervades Wicksteed’s neoclassical economic theory, also built on marginalism, by mostly developing and specifying Jevons’ ‘economic sociology’, while also being influenced by Comte’s general sociology. For instance, as the indicator of his sociological influence and mindedness, incredibly for such a supposedly principled marginalist economist, Wicksteed uses a statement from Comte in the original French ²⁶ that effectively posits the conception and necessity of social economics or economic sociology as the prologue and credo of his main work, The Common Sense of Political Economy. Then, literally following Comte, Wicksteed (1933) considers economics as a particular branch of sociology to the point of treating the first as the ‘handmaid’ (sic) of the second. Thus, Wicksteed (1933), like Comte and Pareto, to be considered next, observes the ‘connection between the narrower problems of Economics and the broader problems of Sociology’, hence between ‘sociology generally’ and economics particularly, i.e. the ‘area[s] of economic and sociologic knowledge’, as well as economics and ‘the other and vaguer branches of sociological study’. In this connection, Wicksteed (1933) implies, à la Comte and Say, that the ‘true political economy’ is essentially social economics, in virtue of its being the study of the ‘economy of the polis, or regulation of the resources of the community’ and distinguished from market, ‘commercial’ economics.

In particular, developing Jevons’ early insights that market supply, demand, and thus price are dependent on social-political processes, Wicksteed suggests the need for what some later sociologically minded economists term the ‘economic sociology of the market’ as the study of the impact of society on markets (Boulding, 1970). Thus, Wicksteed (1933) observes that, in view of its ‘true function’, the market has in reality ‘never’ existed and operated in isolation (‘left to itself’) from society but by implication always within a societal context, even suggesting that it ‘never’ will or should, an apparent and perhaps surprising deviation from or relaxation of the doctrine or utopia of laissez-faire. ²⁷ And the implied social setting of the market serves as the compelling empirical ground and rationale for the sociology of markets and generally for his seemingly ‘shocking statement’ ²⁸ – ‘indecent proposal’ for most pure economists as well as rational choice theorists – -that ‘economics must be the handmaid of sociology’ ²⁹ and to that extent represents ‘sociological economics’. Last, like Jevons and Walras, Wicksteed neither applies nor advocates a universalistic economic approach to all human behavior and society, and thus his vision of economic science does not include the ‘rational choice model’ but encompasses instead social economics, notably the economic sociology of the market, alongside pure and applied ‘political economy’.

Probably the most sociologically minded, advanced, and consistent conception of social economics is found in Pareto’s neoclassical, mostly marginalist, economic theory, along with that of Schumpeter, in late neoclassical or early contemporary economics. This is not surprising because Pareto is a sort of dual, complex economist-sociologist. On the one hand and earlier, he is a neoclassical pure economist, notably marginalist, in his theory of exchange value and prices (Pareto, 1927), en passant unlike Weber as an historical-sociological and thus unorthodox economist (Knight, 1958). ³⁰ On the other hand, Pareto is a classical sociologist, especially in his later career, perhaps (as implied in Parsons, 1967) the ‘best sociologist among neoclassical economists, and the best economist among classical sociologists’, like Schumpeter among early contemporary economists and sociologists.

Generally, Pareto (1932) treats à la Comte economic science, like other ‘specialized disciplines’, as no more than a part of sociology, which he characterizes as the Comtean ‘synthesis of them all’ in virtue of being the science of ‘human society in general’. ³¹ Notably, Pareto suggests that the theoretical ‘results’ of ‘pure economics’ ³² represent an ‘integral and not unimportant part of sociology’, however, they are ‘only a part’ to be considered ‘in conjunction with other parts’ in order to produce a full ‘picture’ of social ‘reality’. ³³ In particular, Pareto (1932) suggests the need for social economics or economic sociology, proposing that a ‘great many’ economic phenomena cannot be studied ‘without the aid of sociology’ to the effect of considering ‘not just the economic phenomenon taken by itself, but also the whole social situation’ because the ‘economic situation’ is ‘only a phase’ of the latter. Moreover, he observes that there are ‘situations’ in which the ‘economic problem’, far from being more important than ‘the sociological problem’, is actually ‘subordinate to [it]’, as is consequently ‘pure economics’ to social economics and sociology overall.

As a corollary, Pareto (1932) regards the economy as an integral element of the total social system as more comprehensive and complex. In his words, ‘economic states’ emerge and exist ‘only as particular cases of the general states of the sociological system’, a view that is incidentally adopted or echoed by some initially pure Paretian–Walrasian, general-equilibrium and subsequently sociologically minded theorists, in a theoretical trajectory almost à la Pareto (Arrow, 1994). Specifically, adopting Walras’ law of utility or interest governing the economy, Pareto characterizes the economic system as constituted of rational-logical actions induced by tastes and appetites in relation to the ‘connections of obstacles’ to the acquisition of ‘economic goods’ (i.e. market demand and supply, respectively), and thus as driven by material interests. However, Pareto considers the ‘sociological system’ to be ‘much more complicated’. This is because it involves also and mostly non-rational or irrational forces that he calls ‘residues’, as ‘manifestations of sentiments’, and ‘derivations’, as false justifications of conduct induced by sentiment, and hence both logical-rational and non-logical/non-rational actions, with the latter predominating. ³⁴ Pareto therefore effectively restates and broadens Walras’ ‘law’ of equity governing society beyond the economy. Consequently, while in ‘pure economics’ equilibrium and thus what has come to be known as the ‘Pareto optimum’ is established from what Pareto (1927) calls the ‘opposition between tastes and obstacles’ (i.e. demand and supply), he suggests ‘much greater complication in sociology’. For in sociology or society non-rational or non-logical actions determined by ‘residues’ combine with and dominate rational-logical actions ‘alone considered by economics’, as do illogical ‘derivations’ in the case of ‘logical reasoning’.

In general, Pareto apparently adopts, develops, and expands Walras’ coherent dichotomy between the laws of the economy and those of society and thus various levels of analysis. Specifically, he does this by incorporating the Walrasian economic ‘law’ of utility or material interest as ‘interests’ and by expressing and expanding the non-economic ‘law’ of equity into ‘residues’, namely, ‘Class IV residues connected with sociality’, including ‘self-sacrifice for the good of others’, and residue ‘Class V Integrity of the individual and his appurtenances’, especially moral ‘sentiments of equality in inferiors’. ³⁵ To that extent, this makes Paretian sociological theory less original than claimed by the author and usually supposed, as by Parsons et al., enamored with Pareto’s sociology, while being avoided by hard-core ‘pure’ economists like Samuelson, at Harvard during the 1930s. In passing, Pareto’s conception of ‘residues’ cum ‘manifestations’ of sentiments or emotions prima facie converges with or resembles Weber’s ideal-type of affective or emotional and, though less manifestly, if Parsons (1967) is correct in his convergence thesis, value-rational social action and even Durkheim’s normative-institutional sociological theory, and thus Durkheimian economics (Gislain & Steiner, 1999). ³⁶

Last, and perhaps most surprising or disappointing to most ‘pure’ economists and rational choice theorists as one of their main ‘heroes’, is how Pareto goes a step further than virtually all neoclassical as well as classical economists, except for in part Mill. He does so in explicitly and categorically rejecting the idea and possibility of the extension of the assumptions of ‘pure economics’ to all social action and society, so ‘rational choice sociology’ as a sort of non sequitur in empirical terms, simply is invalidated by social reality. On the one hand, he maintains that the ‘hypothesis’ of logical-rational actors and actions motivated by material interests or tastes and their satisfaction are ‘not too far removed from realities’ in economy, though not entirely devoid of non-rational forces. Yet, Pareto argues that the generalized hypothesis that humans in society deduce ‘logical inferences from residues’ and act logically-rationally is ‘far removed from realities’ because human actions are induced more by sentiments than by rational factors, and hence derivations, as spurious rationalizations of non-rational actions, are used ‘more frequently than strictly logical reasonings’ (Pareto, 1932).

Pareto infers that sociological ‘rational choice theory’ is empirically an impossibility or a contradiction. This is because, first, such irrational forces and non-rational actions prevail in society beyond the economy. Second, and as a corollary, theoretically deducing rationality from such pervasive irrationality or non-rationality, or assuming away the latter by dissolving it into the former, is an illogical and empirically invalid procedure. In his words, a science premised on the ‘hypothesis’ of invariably logical inferences and rational actions presumably derived from ‘certain given residues’ as irrational forces, simply of universal rationality deduced from non-rationality in society, would represent a ‘sociology like a non-Euclidean geometry’, because such a generalized model of social phenomena would have ‘little or no contact with reality’ (Pareto, 1932). Alternatively, because irrational forces such as sentiments prevail in social action and structure and even sometimes influence markets ³⁷ and economy (Parsons, ³⁸ 1967; Schumpeter 1991), and because deducing universal rationality from such irrationality is a logical fallacy, Pareto suggests a sort of irrational or non-rational choice theory. The latter is in the sense that actors have and make choices in society but these are mostly non-rational or irrational choices, which implies that rational choice theory conflates ‘choice’ and ‘rational’ by equating the first with the second. He considers such irrational or non-rational choice theory, premised on the concepts of residues and derivations, to be the defining theoretical feature of his project of general sociology. Consequently, Pareto’s rendition or vision of economics and sociology does not incorporate ‘rational choice theory’ cum the generalized economic model of society (Morishima, 1998) but does encompass social economics or economic sociology as an indispensable complement and corrective to ‘pure political economy’. In sum, while his rendition of economic science comprises social economics, his project of theoretical sociology is built on the basis of irrational or non-rational rather than rational choice theory.

Schumpeter’s conception of social economics is probably the most sociologically advanced, developed, and coherent within late neoclassical or early contemporary economic theory if he is considered primarily as an economist; yet, like Pareto, he is the most sociologically minded of late neoclassical or early contemporary, specifically Austrian, economists. In short, both seek to fuse, and to some succeed in doing so, economics and sociology (Morishima, 1998). Formally, while Schumpeter prefers the term ‘economic sociology’ to ‘social economics’, he typically uses these two terms interchangeably or, following Weber (Swedberg, 1991), as what economist Robbins (1998) calls the ‘most powerful sociological mind of his period’, uses the first as part of the second, alongside economic history, for example. In general, Schumpeter (1954) defines, à la Durkheim’s sociological institutionalism, economic sociology and thus by implication social economics as the ‘analysis of social institutions’ in respect of their effects on economy, while pure economics, after the model of Walras, is defined as the ‘study of economic mechanisms’. Specifically, using Weber’s terminology with Durkheim’s institutional content, for Schumpeter (1949b) economic sociology involves the ‘description and interpretation – or interpretative description – of economically relevant institutions’, involving ‘habits and all forms of behavior in general’ like ‘government, property, private enterprise, customary or “rational” behavior’. The Durkheimian moment is his renaming Weber’s ‘economically relevant phenomena’ as ‘economically relevant institutions’. In sum, Schumpeter (1950) characterized economic sociology as the ‘analysis of social institutions or of ‘prevalent social habits’ and their shaping of economic actions, evoking again Durkheimian and related sociological institutionalism (Boulding, 1957).

Schumpeter (1949b) distinguishes economic sociology thus understood from ‘economics proper’ described à la Weber as the ‘interpretative description’ of ‘economic mechanisms’, notably ‘market mechanisms’, operating ‘within any state’ of social institutions as the prime subject of the former discipline or social economics. ³⁹ However, Schumpeter (1956) suggests that economics and sociology overlap in respect of analyzing institutions and other social forces shaping ‘economic behavior’, and thus converge on developing economic sociology as ‘a no man’s land or everyman’s land’. ⁴⁰ The overlap in social institutions implies that economic science is just as inherently or by necessity social, more precisely institutional, economics as it is pure economics dealing with market mechanisms. A specialty of economic sociology is what Schumpeter (1951) calls the ‘sociology of enterprise’, including by implication the ‘sociology of the corporation’ (Akerlof, 2007). He characterizes this specialty as analyzing the social conditions generating, sustaining, or restricting ‘entrepreneurial activity’, classifying entrepreneurs ‘according to origins and sociological types’, and even extending into the ‘structure and the very foundations of capitalist society’ (Schumpeter, 1951). ⁴¹ Hence, another related specialty is the economic sociology of modern capitalism, including the study of ‘sociological reasons’ for modern economic crises such as the 1929 Great Depression (Schumpeter, 1939) and perhaps in extension the ‘Great Recession’ of the 2000s (Stiglitz, 2010). ⁴² And Schumpeter’s (1949a) famous ‘theory of economic development’, in particular, the ‘creative destruction’ typifying capitalism, is premised on the sociological ‘theory of cultural evolution’ on the apparently Durkheimian basis of considering the ‘social process’ to be ‘really one indivisible whole’ from which ‘economic facts’ are ‘artificially’ separated.

Finally, despite some implications, such as defining political democracy in terms of competition for power, or interpretations, especially within the ‘economics of politics’ (Mueller, 1993), Schumpeter does not propose or incorporate the universal economic model of society, including polity, and therefore not ‘rational choice’, in particular ‘public choice’, theory in his classification of the disciplines of economic science. ⁴³ Recall that Schumpeter’s disciplines of ‘economic analysis’ clearly and explicitly include economic sociology, and in that sense social economics, alongside economic theory, economic history, economic statistics, and economic policy or political economy, but ‘rational choice theory’ is missing not just in name but in substance as the generalized economics of social action and structure. Unlike Pareto’s explicit rejection of and Mill’s admonition about such a generalization, Schumpeter neither proposes nor rejects ‘rational choice theory’. This likely implies that in his view constructing or using the generalized economic model of society is simply a non-issue meriting no serious consideration or elaboration, and thus the model itself a sort of non-entity within his rendition of economics.

In this respect, Schumpeter essentially embraces and develops Walras–Jevons’ conception and classification of economic science as involving pure economics and social economics or economic sociology, plus applied economics. Yet, he does not anticipate contemporary ‘rational choice’ economists pursuing and extolling orthodox, especially neoclassical, economics as a ‘unifying’ paradigm for all the social sciences, simply an ‘imperial science’. Schumpeter is a highly sociologically minded, complex, broad, and multifaceted, or eclectic economic theorist and so unlikely to conceive, pursue, and propose some sort of generalized, simplistic or one-sided, and rigid ‘economics of society’, including the ‘economics of politics’ and the ‘economics of religion’, defining or typifying ‘rational choice theory’. Unlike the latter and like virtually all great or complex economists he appreciates, including Walras, Schumpeter refuses or is disinclined to proceed simply by the economic version of (Edgeworth–Fisher) ‘mechanical analogies’, notably what his more heterodox contemporary Keynes calls ‘by false analogy’ from the perfect market and the fictitious ‘Robinson-Crusoe economy’ to analyze and construe society as the generalized marketplace. ⁴⁴ For like Walras, let alone Pareto, Schumpeter considers society, including polity, to be a complex social system rather than the mere extension or analogy of what his hero calls the ‘hypothetical regime of absolutely free competition’, i.e. the hypothesis rather than reality of a perfect market and laissez-faire, and the ‘law’ of utility or material interest. It is thus more than simply the set of explicit and implicit ‘markets’ with money and imputed ‘prices’ for material and non-material ‘invaluable’ goods (e.g. shoes and children alike), and the like.

From Smith’s political economy to Marshall’s economics

In addition to their explicit versions, various implicit conceptions and substantive elements of social economics obtain in classical and neoclassical economic theory. For instance, Smith’s classical economic theory contains an implicit conception as well as substantive elements of social economics or economic sociology in the sense of substantively harboring the idea, though not formally using the term. Thus, Schumpeter (1949a) identifies what he significantly denotes as the ‘economic sociology of Adam Smith’, in particular in the latter’s main work in economic theory, the Wealth of Nations. In his view, the ‘most important elements’ of Smith’s ‘economic sociology’ are the ‘division of labor, the origin of private property, increasing control over nature, economic freedom, and legal security’ as phenomena concerning the ‘social framework of the economic course of events’ (Schumpeter, 1949a). Also, some contemporary economists analyze Smithian ‘sociological economics’ (Reisman, 1987), including institutional economics (Razeen, 2002), and describe Smith as a ‘sociological economist’ (Reisman, 1998), and not just a ‘pure’ economic theorist, and as such often ‘written down’ (Samuelson, 1977). Others credit Smith for being the ‘vanguard’ in linking ‘politics and economy’ and thus implicitly in social economics, and even ‘politics and society’ as the subject of political sociology, and generally treat his as a ‘critical’ and modernist theory (Shapiro, 1993; see also Tribe, 1999). ⁴⁵ Generally, Smith considers economics, at the time called ‘political economy’, to be a particular ‘branch’ of social or political science (‘the science of the statesman’) and in turn divided into several ‘branches’, including what he refers to as the ‘system of natural liberty’, ⁴⁶ a term that he probably adopted from the French Enlightenment with its laissez-faire doctrine (Keynes, ⁴⁷ 1931) or philosophy (Arrow & Scitovsky, 1969) via its economic ramification in the Physiocratic School (represented by Turgot, Quesnay, etc.).

Notably, Smith’s ‘economic sociology’ contains certain substantive elements in addition and connected to those identified by Schumpeter. One of these elements is the relation of justice to economy and society. Smith considers justice to be the prime condition of society, to the effect of its being more important than economic ‘beneficence’ to the ‘existence of society’, thus the economy itself, including implicitly modern capitalism (as also suggested by Keynes, ⁴⁸ 1936). Conversely, he admonishes and predicts that the ‘prevalence of injustice’ is likely to be ultimately destructive to society, including the economy or capitalism, which can persist ‘without beneficence’ ⁴⁹ but not without justice as understood by Smith. Another important and related element of Smith’s ‘economic sociology’ concerns what Schumpeter and Durkheim would call the institutional framework of the economy. This involves the economic functions of political institutions like government, which Smith calls the ‘duties of the sovereign’, ⁵⁰ thus recognizing, like other classical economists, the ‘necessity of the state’ rather than advocating ‘anarchism’ (Parsons, 1935a). Hence, this element can be considered as exemplifying institutional social economics. However, this holds more in the sense of the old Veblenian heterodox institutional economics (Hodgson, 1998; see also Coase, 1998; Greif, 1998; North, 1994) and Durkheimian and other sociological institutionalism (Boulding, 1957; Stinchcombe, 1997) that are found to have mutual connections and commonalities (Gislain & Steiner, 1999) than the ‘new institutional economics’ (Williamson, 2010), especially its variant premised on the rational choice or cost–benefit efficiency, transaction cost model of social institutions (Fligstein, 2001; Granovetter, ⁵¹ 1985).

Smith’s ‘economic sociology’ also involves an element of class stratification in the specific sense of capital–labor relations and the tendency and action of the capitalist class in economy and society. This may be surprising, perhaps ‘shocking’ for contemporary ‘libertarian’ economists à la Mises (1966), Hayek (1991), Friedman (1982) and others (Buchanan, 1991), who extol capital and its associations, including openly or implicitly monopolies and oligopolies, as a sort of ‘unmitigated good’, and condemn labor and its collective organizations or unions as equally ‘evil’. Thus, Smith describes the typical tendency or action of the capitalist class in economy and society as no less than a concerted ‘conspiracy against the public’, including ‘some contrivance to raise prices’. In short, what Smith registers and implicitly predicts is essentially a pattern of monopolizing closure of markets by extending the market but restricting competition as directed ‘always’ against the public and serving ‘only’ capital. Cynics may comment that Smith’s diagnosis of the capitalist class’s ‘conspiracy against the public’ is the only true case of ‘conspiracy theory’ in economy and society. Generally and also surprisingly, Smith posits that the economic interests of the capitalist class are ‘never exactly the same’ as and ‘even opposite’ to the interest of society as a whole, to the point of the first having an ‘interest’ in deceiving and ‘even’ oppressing the public and ‘upon many occasions’ acting ‘accordingly’. ⁵²

Smith’s economic theory implies social economics or economic sociology in conjunction or coexistence with market economics as the explicit ingredient. However, it does not entail or imply the economics approach to human action and society cum the ‘rational choice model’, including the ‘economics of politics’ qua ‘public choice theory’. If anything, he treats political economy as a ‘branch’ of a broader social/political science rather than reducing the latter to the ‘rational choice model’ extended from the market and economy to society, including politics and religion, despite some interpretations linking Smith, especially the Wealth of Nations, with the economics of politics (Buchanan, 1991; Mueller, 1993) and the ‘economics of religion’ (Anderson, 1988; Iannaccone, 1998). For instance, Smith’s Wealth of Nations conjoins what Schumpeter calls ‘market economics’ with ‘economic sociology’ or ‘sociological economics’ (Reisman, 1998). Yet, it hardly applies or proposes a ‘rational choice model’ in the sense of extending the first analysis beyond the realm of markets and the economy, as admitted by some moderate rational choice sociologists (Boudon, ⁵³ 2003). Moreover, his first major work, the Theory of Moral Sentiments, can be considered a sort of inverse of rational choice theory, i.e. non-rational choice theory ⁵⁴ (Boudon, 2003; Parsons, 1935a) almost à la Pareto. It is so in its emphasis on irrational variables like ‘moral sentiments’, including, besides justice, ‘sympathy’ and various other human ‘emotions’ and ‘passions’ ⁵⁵ (Tullock, 1972) as operating in economy and society rather than ‘self-interested considerations’ ⁵⁶ regarded as secondary or residual. In passing, Smith’s theory of ‘moral sentiments’ resembles and perhaps anticipates or adumbrates Walras’ ‘laws’ of social non-economic relations and Pareto’s of ‘residues’ cum ‘manifestations of sentiments’, as well as Weber’s ideal-type of affective or emotional social action. In particular, Smith’s idea of justice as the necessary and primary condition of society adumbrates Walras’ ‘law’ of equity governing social relations and Pareto’s ‘sentiments of equality in inferiors’, and Weber’s notion of ‘substantive values’ including these and related concepts. In sum, neither Smith’s Theory of Moral Sentiments nor even the Wealth of Nations builds or projects a ‘rational choice model’.

Marshall’s neoclassical economic theory proper – as distinguished from pure anti-classical marginalism à la Walras, Jevons, and Menger – provides another instance of an implicit conception of social economics or, in Schumpeter’s view, economic sociology. Schumpeter (1941) suggests that Marshall’s major work (Principles of Economics first published in 1891) contains ‘an economic sociology of 19th-century English capitalism’ supported by ‘historical bases of impressive extent and solidity’, conjoined with (‘behind, beyond and all around’) the ‘core of the analytic apparatus’ or the ‘kernel’ of market economics. In addition, some contemporary economists describe Marshall’s as ‘sociological economics’ (Reisman, ⁵⁷ 1990) and generally as having ‘manifold sources of economics’ (Groenewegen, 1995).

As known to sociologists, Parsons (1967), another former economist though lacking the stature and impact of Schumpeter, Pareto, Wieser, and even Weber within economics, famously posits Marshall’s convergence with Weber, Pareto, and Durkheim on the normative-institutional, value-based (‘voluntaristic’) theory of social action, including economic sociology. In a way, Parsons’ inclusion of what he calls the ‘most eminent economist of his generation’ among ‘major’ European social theorists is, as Schumpeter would suggest, only justified in terms of Marshall’s ‘convergence’ with these sociologists, including Pareto in his later career, on economic sociology or sociological economics. Otherwise, it is arbitrary and accidental, for Marshallian and any ‘pure’ economics would instead indicate divergence or dis-affinity in this respect. ⁵⁸ And this is what Parsons (1967) explicitly or implicitly attempts to demonstrate, though he describes such demonstration in terms of ‘convergence’ on a ‘voluntaristic’ theory of social action, rather than on economic sociology or sociological economics, between Marshall and these sociologists, especially Durkheim and Weber. Moreover, he suggests that Marshall shares with Durkheim and Weber ‘essentially the same view’ that modern capitalist society’s ‘basic element’ comprises ‘certain common values’, including economic and political liberty, as ‘an end in itself and as a condition of the expression of ethical qualities’, rather than being defined ‘exclusively or even predominantly in terms of utilitarian want satisfaction’ (Parsons, 1967). ⁵⁹ He provides another instance of convergence on economic sociology in stating that, in Marshall’s analysis, ‘wants adjusted to activities’ and ‘modes of activity’ in the capitalist economy, which express ‘a single, relatively well-integrated system of value attitudes’, resemble the values typifying ‘Weber’s spirit of capitalism’, both value-systems having essentially societal origins as a sociologically more relevant commonality (Parsons, 1932). Also, he implies that Marshall and Weber converge on considering the ‘development of this value system’ of modern capitalism, ‘along with increasing rationality’ in the latter, to be the ‘primary moving force of social evolution’ (Parsons, 1967).

Marshall himself follows Mill and even Comte in considering economics a branch of sociology regarded as a unifying, though not yet unified, social science. Thus, Marshall (1961) appreciates ‘the present movement towards Sociology in America, England and other countries’ for its recognizing the ‘need for the intensive study of economics and other branches of social science’, and registers ‘some excellent intensive studies’ done ‘under the name of Sociology’. ⁶⁰ As an illustration, he credits Comte for doing ‘good service’ by emphasizing the ‘solidarity of social phenomena’, a fact which makes the ‘work of exclusive specialists even more futile in social than in physical science’. He also notes that, ‘under the combined influence of Comte’ and others, Mill later in his career made prominent the ‘human, as opposed to the mechanical, element in economics’, at the same time citing approvingly the latter’s concession that ‘a person is not likely to be a good economist who is nothing else’ (Marshall, 1961).

An important substantive element of Marshall’s economic sociology or social economics is, as Parsons and Schumpeter suggest, the impact of social values, rules, and institutions on the economy. In particular, Marshall (1961) registers that the indirect ‘effects of custom in preventing the methods of production and the character of producers’ for free development are ‘cumulative’, though ‘not obvious’, thus exercising a ‘deep and control influence’ to the effect that so long as customs or traditions impede the ‘progress of one generation’, the next will be forced to begin from ‘the lower level than otherwise’. Lastly, Marshall’s economics entails or projects no rational choice theory as an implicit ‘impossibility theorem’. This is implied in his defining of ‘Political Economy or Economics’ as the ‘study of mankind in ordinary business of life’, i.e. only of ‘that part of individual and social existence’ that is ‘most closely’ connected with the ‘attainment and use of the material requisites of wellbeing’ (Marshall, 1961), but not of the other parts, and thus not the economic analysis of all social life as in ‘rational choice theory’. Apparently following Mill, Marshall’s economics is ‘on the one side a study of wealth’ and ‘on the other, and more important side, a part of the study of man’ (see also Parsons, ⁶¹ 1935a) and so by implication of sociology à la Comte seen as a unifying social science, thus far from being the universal ‘rational choice’ model of human behavior and society. Therefore, like Mill, Say, and Comte, Marshall makes a distinction between various levels of analysis, notably between that of sociology or unified social science and that of economics as its part.

In addition, Pigou’s neoclassical, specifically Marshallian, economic theory also implies certain ideas and elements of social economics. Closely following on and specifying Marshall, Pigou (1960) considers economic welfare to be an integral element of total social welfare, thus the economy part of society. ⁶² Hence, he identifies, elaborating on Marshall, economic, but not total social, welfare as the ‘subject matter of economic science’ (Pigou, 1960), thus implicitly ruling out expanding the theory and approach of the latter beyond this domain, i.e. the ‘rational choice model’. And as a particular curiosity, Pigou outlines a neoclassical theory of labor exploitation (defined by the discrepancy between the value of marginal productivity and wages at the expense of the latter) ⁶³ and thus inequity in the economy. He hence provides a marginalist analogue to Marx’s (1967) conception that stresses the ‘social [property] relations of production’ between exploiting owners of capitals and exploited laborers (see also Wright, 2002). In so doing, Pigou perhaps has in mind or evokes Smith’s warning about the destructive consequences of injustice for society as well as Walras’ law of equity governing social relations.

Implied elements of social economics in ‘purest’ economic theory

Classical/neoclassical economic theory also implies other, less manifest, salient, and influential conceptions or approximations of social economics. Thus, Ricardo’s economic theory probably involves prototypical ‘pure’ economics (Samuelson, 2004) and less ideas or elements of social economics or economic sociology compared to that of Smith (Schumpeter, 1954) and other major classical economists, especially Say and Mill, not to mention Marx (deemed for the present purpose a ‘sociologist’). Still, Ricardo’s economic theory is not totally expunged from extant elements of social economics in the Smith–Say–Mill sense and generally sociological implications or elements (North, 1915; Parsons, 1935a). This is indicated by the fact that Ricardo links economic distribution, elaborating on Smith and anticipating Marx, with class structure as a determinant, and thus with ‘socio-economic conditions’ as ‘a social or institutional datum’ determining in particular the ‘level of real wages’ (Dobb, ⁶⁴ 1973). For instance, Ricardo (1975) observes that wealth (the ‘product of land’) resulting from ‘the united application of labour, machinery, and capital’ is distributed among ‘three classes of the community’, landowners, capitalists, and workers, as their rent, profit, and wages, respectively. In addition, he recognizes that such non-economic factors as what he calls the ‘habits and customs of the people’ often play a role in the economy, including the market (also, Parsons, ⁶⁵ 1935a).

While Ricardo’s economic theory represents or implies less social economics than the theories of his predecessor Smith, contemporary Say, and follower Mill, it does not generalize its ‘purest’ market economics, including the ‘law of [equilibrium] markets’ shared with Say (Keynes, 1936), to the non-economic realms of society. To that extent, his economic theory does not construct or envision a ‘rational choice model’. In this sense, even the ‘purest’ and ‘most able’ (according to Schumpeter, 1954) of the classical economists comes closer to Smith’s combination of market and social economics than to the contemporary universal economic approach to social action and society. For instance, Ricardo’s attention to class stratification as impacting wealth distribution is largely lost or irrelevant in the ‘rational choice model’, especially within economics (Lie, 1997). While not as relevant as Smith, Say and Mill, or even Walras and Jevons in terms of social economics, and being primarily a pure economist, Ricardo is a secondary social economist (likely against his intentions), at least in virtue of his concern for the link of class structure and wealth distribution and the economic effects of social customs. Yet, admittedly he is no ‘rational choice’ theorist at all or proponent of academic ‘economic imperialism’ (Tullock, 1972), despite implied or possible interpretations to that effect. ⁶⁶ In sum, Ricardo comes closer to conceiving and analyzing the economy as Say’s ‘economy of societies’ than society as an extended marketplace, a system of economics and non-economic ‘markets’, and in that sense closer to social economics (North, ⁶⁷ 1915; Parsons, 1935a) than to ‘rational choice theory’.

The aforesaid of Ricardo also holds true of William Senior as probably the second ‘purest’ major theorist within classical political economy. Likewise, Senior’s economic theory is primarily pure economics, specifically catallactics – more so than even that of Ricardo, which dismisses market exchange as ‘mere circulation’ in favor of production and distribution – though more openly ‘apologetic’ in that it uses the concept of ‘abstinence’ to justify profit and the capitalist class, as Marx accused and some contemporary economists recognize (Samuelson, 1994). Nevertheless, like that of Ricardo, Senior’s classical political economy is not entirely purified of elements of social economics; if anything, the latter appears more manifest, particularly more institutional. Senior acknowledges the critical impact of social institutions on wealth distribution, elaborating on Smith–Say’s earlier insights into institutional influences in the economy and anticipating Mill’s treatment of distribution as the problem of social institutions. For instance, he proposes that the ‘peculiar institutions of particular Countries’ – invoking slavery, legal monopolies, and poor laws – tend to affect the laws of ‘Distribution of Wealth’, making them less universal, i.e. more relative across societies and in historical time, while those of ‘Nature and Production of Wealth’ are ‘universally true’ (Senior, 1951). Like that of Ricardo, Senior’s economic theory does not really extend pure economics, especially catallactics with its ‘law’ of market exchanges and equilibrium markets à la Say, into society’s non-economic or non-market relations and systems, and to that extent does not represent or construct the ‘rational choice model’. At least, the elements or embryos of institutionalism, manifested in the influence of social institutions on economic processes or outcomes like wealth distribution, surprisingly but effectively make Senior’s classical political economy closer to Smith–Say–Mill’s social-institutional economics than to the contemporary ‘rational choice theory’ as, in his terms, the catallactics of society cum a generalized marketplace.

In addition, an implied idea or element of social economics is found in the economic theory of John Cairnes, deemed the ‘last important’ classical economist (Stigler, 1957) and having the misfortune to witness first-hand and be overshadowed by what Schumpeter (1954) views as the ‘Copernican’ marginalist revolution in economics during the 1870s, specifically by Jevons et al. in England. This is the idea of the social determination or formation of individual tastes or preferences and motivations and, by implication, economic actions (Akerlof, 2007). Thus, apparently following Mill, Cairnes (1965) observes that the ‘desires, passions and propensities’ involved in the individual ‘pursuit of wealth’ are ‘almost infinite’ and, sociologically more relevant, such ‘motives and principles of action’ in the economy tend to develop and so change the ‘progress of society’. The last statement suggests that Cairnes, like Mill, considers individual tastes or preferences and motives to be variables in virtue of their being subject to cultural and historical variation, including ‘progress’, simply social products, just as do most sociologists and many contemporary economists (Akerlof, 2007; Greif & Tabellini, 2010; Prendergast, 2008). Alternatively, he implies that tastes are not constants and given, unlike ‘natural’ propensities – and hence actors are not ‘natural born’ utility-maximizers – as treated in contemporary pure economics with its ‘revealed preferences’ theorem (with qualifications ⁶⁸ in Samuelson, 1983), and in the universal economic approach to human behavior (Stigler & Becker, 1977). Hence, Cairnes suggests, as also some prominent contemporary economists recognize, that individual tastes, preferences, and motivations are mostly ‘socially caused’ (Arrow, 1994), especially by cultural norms (Akerlof, 2007). This is indicated by ‘acquired tastes’ from culture and history (Frank & Cook, 1995) and ‘interdependent preferences’ through social interaction, notably societal imitation or class emulation (Leibenstein, 1950), which are consequently ‘not really fixed’ (Pollak, 1978). At least on this account, like Mill’s version which influences it, Cairnes’ classical political economy implies in part social economics ⁶⁹ (Becker & Murphy, 2000) but not ‘rational choice theory’, or does more the first than the second. Thus, a study shows that, following Mill and thus by extant implication Comte, Cairnes considers political economy a ‘distinct and separate branch’ of ‘sociological theory’, i.e. ‘General sociology’ (Robert, 1995). ⁷⁰

As indicated, in the Austrian neoclassical-cum-marginalist economic school, Wieser’s is the early and Schumpeter’s later most explicit, coherent, and, developed conception and project of social economics. But they are not the only cases if its implicit versions or approximations are considered, starting with the founder, Menger, and extending into the contemporary members like Mises, influenced also by Weber and in part anti-sociological Hayek (outside the present scope). While commonly regarded as a ‘pure’ and staunch marginalist theorist because of his main work (Menger, 1950), Menger still implies the idea of social economics by the notion of a ‘social economy’ composed of ‘individual economies’. Evoking Say and unwittingly Comte, Menger (1963) defines – what incidentally most German-speaking writers from Fichte and List like to call and extol as – the ‘national economy’ as inherently a social economy in virtue of being the ‘social form of [economic] activity’. ⁷¹ Notably, Menger (1892) considers the money medium of exchange a ‘social phenomenon’, namely the ‘social institution of money’, in its origins. In general, Menger’s notion of an inherently ‘social economy’ as an economic system by implication makes economic science necessarily social economics, even if he would prefer its theoretical-formal variant. In particular, he suggests that theoretical, as sharply distinguished from historical and applied, economics has its ‘position in the sphere of the social sciences’. He thus implies that economic science is a branch of sociology in the generic sense, an implication Wieser makes explicit and develops, as does Mises ⁷² (1960), and his and Weber’s student Alfred Schutz ⁷³ (1967) registers (Prendergast, 1986).

Substantive elements of social economics are less salient in the third original member of the Austrian School, Böhm-Bawerk (Schumpeter, 1956), as a sort of this School’s equivalent of Ricardo in the sense of advocating ‘purest’ economic theory, though in a radically and deliberately different version (marginal-utility vs. labor theories of value and prices), but which is not totally purged or denied. Like Ricardo’s, Böhm-Bawerk’s ‘purest’ economics implies or acknowledges such elements, namely his marginal utility theory of value and prices, including his analysis of capital and his mostly psychological explanation of interest. In general, Böhm-Bawerk (1959) notes, with reference to Say, the ‘social formations of value and price’, including the price of capital or interest. In particular, in analyzing the latter, Böhm-Bawerk (1959) acknowledges and distinguishes the ‘social and political problem’ from the ‘theoretical problem of interest’. Specifically, he defines the ‘social and political problem’ in terms of ‘principally’ the ‘effects’ of interest on capital, notably its justice or fairness (whether ‘just, fair, useful, good’ or not, etc.) in contrast to the ‘theoretical problem’ defined ‘exclusively’ by the ‘causes of interest’ (‘why’) (Böhm-Bawerk, 1959). These causes are identified in higher ‘subjective valuations’ of present over future goods (cf. Böhm-Bawerk, 1929). Notably, he registers that the issue of interest on capital has, in economic theory since Smith, attained the ‘rank of a great social problem’ or a ‘social problem of the first rank’ such that economics cannot afford to ‘overlook’ (Böhm-Bawerk, 1959). In methodological terms, Böhm-Bawerk (1929) comes closest to social economics or economic sociology as a multi-paradigm and multi-method discipline when he allows for methodological pluralism in stating that any, including theoretical and historical-empirical, methods are ‘good’ so long as ‘in given case’ they help attain the ‘truth’ as the ‘objective of science’. ⁷⁴ In particular, he does not reject a priori the historical-empirical cum ‘objective’ method but instead recognizes it as an ‘aid to the attainment of real knowledge’ (Böhm-Bawerk, 1959). In so doing, Böhm-Bawerk apparently attempts to reconcile opposite methodological positions of the Austrian (Mengerian) and Historical (Schmoller) Schools, thus making what Menger (1963) calls the ‘battle of methods’ between theoretical and historical economics redundant and useless (Swedberg, 1998).

Hence, despite their ultimately futile methodological battles, the Austrian theoretical-marginalist and German historical-empirical schools share the substantive concept of a social economy, and to that extent social economics, a commonality overlooked or neglected in the literature favoring their differences. This is what even ‘pure’ theorist Menger acknowledges, and especially Wieser emphasizes in his special book on social economics, not to mention Schumpeter’s sociological-economic fusion. Furthermore, Weber does so in his attempt to synthesize the Mengerian and Schmollerian oppositing methodological positions by the project Foundations of Social Economics, broadly understood as involving, alongside economic sociology, economic theory and economic history, and thus members of both schools (Swedberg, 1998). In turn, as an unorthodox challenge, like Veblenian and Durkheimian institutional economics (Gislain & Steiner, 1999), to orthodox theoretical economics, the Historical School is a historical-empirical version of what some neoclassical economists (Knight, 1958) call ‘sociological economics’ seen as equivalent to economic sociology. ⁷⁵ For example, in addition to Schmoller as the leading representative, etc., it is exemplified by its later nominal members Sombart and Weber (Parsons, 1935a). Overall, the Austrian Economic School contains explicit (as in Wieser and Schumpeter) and implicit (as in Menger and Böhm-Bawerk) ideas of social economics as an economic science and of the social economy as an economic system, but does not contain the universalistic economics of society cum ‘rational choice theory’. This applies to the Austrian School’s early version, thus excluding its later versions like Mises’ ‘praxeology’, largely premised on Weber’s conception of instrumental rationality, as well as Hayek’s anti-sociological ‘true individualism’, as another story outside the scope of this article.

Last, the ‘purest’ neoclassical economic theories imply certain ideas and elements of social economics or economic sociology, as shown by such supposedly purist marginalist economists as John B. Clark, Irving Fisher, and Francis Edgeworth. Implicit elements of social economics are present in Clark’s (1956) neoclassical economic theory, premised on rigorous marginalist principles, such as the marginal utility principle of exchange value and the marginal productivity rule of wealth distribution, and commonly considered, as by Veblen (1899), as its major critic pointing to its and other orthodox economics’ ‘preconceptions’, among the ‘purest’ and ‘hard-core’ versions of marginalism, together with Böhm-Bawerk, Edgeworth, and Fisher. Thus, Clark (1899) proposes ‘a division’ of economic theory on the basis of ‘sociological evolution’ reminiscent of Comte and Spencer, namely into ‘Social Economic Dynamics’, as an apparent variation on Comtean ‘social dynamics’ and an extension of Mill’s ‘economic dynamics’ inspired by the latter. Curiously, in so proposing, Clark seemingly attempts to answer Veblen’s (1898) ‘prayers’ implied in his question ‘[W]hy is economics not [but should be] an evolutionary science?’ Furthermore, Clark acknowledges that society impacts the economy, observing that producers are organized in ‘groups and subgroups’ by the ‘socialization of economic life’ and that labor, capital, and related economic variables are dependent on ‘social organization’. ⁷⁶ Notably, he implies a certain social conception and explanation of exchange values or prices, thus introducing an element of economic sociology of the market by considering ‘market value’ to be (‘also’) a ‘social phenomenon, not a universal [or natural] one’. At this juncture, Schumpeter (1908–1909) describes Clark’s conception of exchange value, like that of Wieser, as positing ‘social value’. In sum, Clark’s ‘hard-core’ marginalism implicates and conjoins elements of social economics, in particular socioeconomic dynamics and the economic sociology of prices, with pure market economics or catallactics, but does not suggest the ‘rational choice model’ as its generalization beyond markets and economy to all society.

By comparison to that of his colleague Clark as well as Walras and Jevons, Fisher’s marginalism is even ‘purer’ than Böhm-Bawerk’s but is still not totally purified of social or socio-psychological elements. Continuing Walrasian mathematical pure economics saturated with ‘mechanical analogies’, such as the market as ‘industrial machinery’ or ‘hydraulic’ (Fisher, 1965; Mirowski, 1989), and becoming influential by his theory of interest and economic crises – yet unable to foresee the 1929 Great Depression, and thus losing his stock investments – he implies social economics without constructing or envisioning ‘rational choice theory’. For instance, he suggests elements of economic sociology of the market, stating that the assertion that supply and demand determine price { XE “prices” } (e.g. the rate of interest) is { XE “supply and demand” } ‘merely’ stating rather than { XE “state” } solving the ‘problem’ (Fisher, 1954). The problem consists in exploring the ‘particular forces’ (Fisher, 1954) determining supply and demand if such factors are observed to operate not only within but also beyond the ‘realm of traditional economic variables’ by considering ‘social and institutional conditions’, including the ‘political sphere’ (Samuelson, 1983). ⁷⁷ To that extent, this is the problem of a ‘sociological [or socio-psychological] explanation of the causes of supply, demand and price’ (Myrdal, 1953). ⁷⁸

Compared to that of Clark and Walras, Fisher’s implicit or proxy social economics and general marginalism is more psychological and less sociological, namely less institutional (a common feature of neoclassical and contemporary economics according to Lewin, 1996). This is indicated by his definition of interest as a ‘psychological depreciation of the future satisfaction of wants’ or an ‘index of impatience’, the notion of ‘psychic income’ or of ‘enjoyment income’ via consumption as a ‘psychological entity’, and the like (Fisher, 1954). If social economics is only intimated or approximated in Fisher’s marginalism, ‘rational choice theory’ is simply missing. Like Walras, Fisher does not attempt or propose to extend the mathematical theory of exchange value and market prices, including the rate of interest as the ‘most pervasive price in the whole price structure’, beyond markets and economy to all society, thus the ‘rational choice model’ with its ‘implicit’ non-economic political, religious, marriage, and other ‘markets’ and ‘imputed prices’ of human ‘commodities’ like spouses and children, etc. (Becker, 1991). In sum, Fisher’s marginalist economics follows and elaborates on Walras’ ‘pure political economy’, confined to the ‘machinery’ of the market and thus not generalized into ‘rational choice theory’, implies or evokes Walrasian ‘social economy’ or Jevons’ ‘economic sociology’ as a branch of economics, and resembles Böhm-Bawerk’s marginalism (e.g. their theories of interest are both psychological).

Lastly, Edgeworth’s marginalism, while appearing as ‘pure’ as or even ‘purer’ than that of Fisher and Böhm-Bawerk, approximates certain elements of social economics, including the economic sociology of the market and prices. As with other classical/neoclassical economists, in Edgeworth social economics coexists with explicit market economics, called ‘pure catallactics’, and also, unlike virtually all of them, with extant or proxy ‘rational choice theory’, what he calls ‘mathematical sociology’ and ‘mathematical psychics’ as ‘social mechanics’. Like that of Wicksteed but even more so, Edgeworth’s economic theory is difficult – not just because of the often obscure prose – to understand and analyze without considering that of Jevons as the main and pervasive influence. On the one hand, his ‘pure catallactics’ builds on what he calls the Jevonian marginal utility ‘Formulas of Exchange’ and is thus less pertinent or original within marginalism. On the other hand, Edgeworth’s implied social economics follows or echoes Jevons’ explicit conception of ‘economic sociology’, notably with respect to the market, and by implication Walras’ ‘social economy’ and even Comte’s general sociology. For instance, this ‘purest’, seemingly least sociologically minded marginalist economist, like Jevons and Wicksteed, approvingly invokes (though imprecisely) Comte in considering marginal utility theory to be the ‘most sublime branch’ of economic and even sociological analysis, or the ‘branch most applicable to Sociology’ (Edgeworth, 1967 [1881]). Edgeworth also approvingly quotes Jevons’ statement that often transactions in the market are concluded ‘upon other than strictly economical grounds’, like bargaining or haggling, i.e. on social considerations, including political influences, as a substantive element of the Jevonian economic sociology of the market. In addition, Edgeworth implies this in observing what he calls the ‘social distance’ between individual agents in markets and its implied effects on market transactions and processes, an impact that some contemporary sociologically minded economists especially emphasize and demonstrate (Akerlof, 2002).

The only major point at which Edgeworth deviates from Jevons and diverges from Wicksteed is his proposal for ‘mathematical sociology’ or ‘social mechanics’ premised on the generalization of marginal-utility theory beyond economy and, to that extent, extant or proxy ‘rational choice theory’. Thus, he suggests that ‘the Calculus of Variations’ in utility is the ‘branch most applicable to Sociology’ ⁷⁹ (Edgeworth, 1967 [1881]), yielding ‘mathematical sociology’ as a sociological generalization of such calculation and in that sense to be interpreted as ‘rational choice theory’. This is still a sort of rationalistic charitable interpretation, for the two are not necessarily identical, as shown by Pareto’s also mathematical sociology yet essentially irrational choice theory (Fararo, 1993; Parsons, 1967; Schumpeter, 1991). On this account, Edgeworth’s ‘rational choice theory’ is a deviant case or an exception rather than a typical element within neoclassical economics, thus effectively confirming rather than contradicting the ‘rule’. In sum, Edgeworth’s is probably the only well-known, though relatively secondary, neoclassical and generally orthodox economic theory implying social economics in conjunction with pure market economics as well as proxy ‘rational choice theory’ cum ‘mathematical marginal-utility sociology’, or ‘social mechanics’. (The preceding consideration is summarized in Table 1 in the Appendix.)