The original institutional and evolutionary economics

As far as we are aware, Veblen (Reference Veblen1898a: 393–394) was the first to use the term ‘evolutionary economics’. He also stressed the role of institutions, a behavioural focus, and applied Darwinian notions of variation and selection to institutional development and change (Camic and Hodgson, Reference Camic and Hodgson2011; Hodgson, Reference Hodgson2004; Veblen, Reference Veblen1898a, Reference Veblen1899). Walter Hamilton (Reference Hamilton1916: 863, n. 5) reported that Robert Hoxie, who was one of Veblen's students, had described himself as an ‘institutional economist’ (McNulty, Reference McNulty1973; Rutherford, Reference Rutherford1997). This is the first known use of this term. Hamilton (Reference Hamilton1919) then made it famous by his launch of a movement that he described as ‘institutional economics’ at the December 1918 meeting of the American Economic Association. From its beginning, Veblen was seen as a foremost inspiration for this school.

Hamilton's founding statement is still well worth reading today. One section is headed as: ‘The proper subject-matter of economic theory is institutions’. This is immediately followed by another section titled: ‘Economic theory is concerned with matters of process’ (Hamilton Reference Hamilton1919: 313–314). Thus, institutions and evolution were joined together at the beginning. Hamilton (Reference Hamilton1919: 315) wrote: ‘We need constantly to remember that in studying the organisation of economic activity . . . we are dealing with a unified whole which is in process of development’. In a challenge to orthodox assumptions, Hamilton (Reference Hamilton1919: 316) criticised the ‘extreme individualism, rationality, and utilitarianism which . . . still finds expression in neo-classical economics’.Footnote ³ Instead, Hamilton (Reference Hamilton1919: 316–317) emphasised that ‘a theory of motives must be used which is in harmony with . . . modern social psychology’. By contrast, the notion of ‘self-interest [offers] nothing more than a blanket formula’. One has ‘to go beyond it to find the concrete influences which animate the behaviour of individuals’, including ‘conflicting values’ as well as ‘instinct and impulse’. The ‘most important’ omission of previous orthodoxy was its neglect of ‘the influence exercised over conduct by the scheme of institutions under which one lives’.

As well as embracing psychology, the original institutionalism welcomed contributions from other disciplines outside economics. Hamilton had been taught by the American sociologist Charles Horton Cooley, who he described as one of the ‘leaders’ of the emerging institutional economics. In turn, Cooley's (see especially 1902: 33–34) work influenced J. M. Clark (Reference Clark1918) and Simon (Reference Simon1957, Reference Simon1996), and it is one of the likely origins of the concept of bounded rationality.

Institutional economics after Veblen

Despite making important contributions, the original institutional economics failed to produce a systematic theoretical alternative that could rival or supplement the rising neoclassical economics.Footnote ⁴ This failure was in part due to seismic shifts in philosophy, psychology, anthropology, and other disciplines after 1914 (Cravens, Reference Cravens1978; Degler, Reference Degler1991). For at least two generations, biological ideas and rhetoric were driven out of the social sciences, partly as a reaction against the misuse of ‘survival of the fittest’ phraseology to justify belligerence in the First World War.Footnote ⁵ The instinct–habit psychology, of William James (Reference James1890) – which had been strongly influenced by Darwinism – became rapidly unfashionable and was replaced by behaviourism. Evolutionary terminology and ideas were widely abandoned. Even Schumpeter (Reference Schumpeter1934: 57) wrote: ‘The evolutionary idea is now discredited in our field’.

Consequently, Veblen's research programme to develop an ‘evolutionary economics’ based on Darwinian principles was deserted by his followers (Hodgson, Reference Hodgson2004; Homan, Reference Homan1932; Rutherford, Reference Rutherford1998). Instead, the term ‘evolution’ was retained by institutional economists to refer loosely to change. Reference to Darwinism was dropped. Much less use was made of psychology. The research programme of Veblen and a major part of Hamilton's (Reference Hamilton1919) manifesto were abandoned. This also helps to explain why Nelson and Winter (Reference Nelson and Winter1982) made no reference to Veblen, although they have since emphasised his importance.Footnote ⁶

After Veblen, the other most important original institutional economists were John R. Commons and Wesley C. Mitchell. Veblen died in 1929, Commons in 1945, and Mitchell in 1948. Although the original institutional economics remained influential in America at least until the 1940s, by then it had lost much of its former vision of what ‘evolutionary’ and ‘institutional’ approaches might mean. The movement partly fragmented and gradually declined, with Clarence Ayres becoming the most important influence on North America. Ayres (Reference Ayres1944, Reference Ayres and Dowd1958) explicitly rejected Veblen's instinct–habit psychology and saw institutions negatively as constraints on technological progress. Also, contrary to Hamilton (Reference Hamilton1919), Ayres generally abandoned the analysis of institutions as such, and the use of psychology in the analysis of individual motivation. He preferred the term ‘evolutionary economics’ to ‘institutional economics’, and saw ‘evolution’ as principally the interplay between the positive forces of technology and restraining institutional inertia.

John K. Galbraith was also strongly influenced by the original institutionalism (Dunn, Reference Dunn2011; Galbraith Reference Galbraith1952, Reference Galbraith1958, Reference Galbraith1969). But while his works had a huge popular appeal, they have made little mark on theoretical economics. The fractured post-War original institutionalism made little headway outside the United States, apart from important singular cases such as Gunnar and K. William Kapp. Myrdal (Reference Myrdal1957, Reference Myrdal1968), of course, made several pioneering contributions and won the Nobel Prize. Kapp (Reference Kapp1950) was one of the founders of ecological economics. But neither left a lasting school of followers in their European homeland or elsewhere.

By the time that Williamson (Reference Williamson1975) coined the term ‘new institutional economics’, its ‘original’ predecessor had little influence upon what had become a mainstream economics dominated by mathematical technique. To their great credit, Williamson and others brought the study of institutions back on the agenda, but they showed little interest in the achievements or errors of the original institutionalism. Except for a nod in the direction of Commons, it was largely ignored or dismissed. For example, Coase (Reference Coase1984: 230) wrote that the original ‘American Institutionalists were not theoretical but anti-theoretical . . . Without a theory, they had nothing to pass on except a mass of descriptive material waiting for a theory or a fire’. Williamson (Reference Williamson1996: 1792) likewise alleged that the ‘older style institutional economics was content with description, whereas newer style institutional economics holds that institutions are susceptible to analysis’.

Allegations that the original institutional economists were ‘atheoretical’ or ‘against theory’ have been amply refuted elsewhere (Hodgson, Reference Hodgson2004; Rutherford, Reference Rutherford2001, Reference Rutherford2011). One has only to read Hamilton (Reference Hamilton1919) to see that they were far from ‘content with description’ and wished to prioritise analysis. The unfortunate rift between the original and the NIE is the central theme in Hodgson's (Reference Hodgson2014) reflection in this issue.

Evolutionary economics after Veblen

After the original institutionalism had abandoned Veblen's project to develop an evolutionary economics on Darwinian foundations, the term ‘evolutionary’ was used less frequently by economists. Revival in its usage was a long and gradual process. Two important post-War attempts to use versions of Darwinian selection were by Armen Alchian (Reference Alchian1950) and Milton Friedman (Reference Friedman and Friedman1953). Nevertheless, the general equilibrium theory was increasingly seen as the core of economics and dynamic considerations were pushed aside. When eventually the tide began to turn, it came from outside neoclassical economics, with important contributions to dynamic analysis by the Austrian school economist Friedrich Hayek (Reference Hayek and Hayek1967) plus other creative dissidents such as Nicholas Georgescu-Roegen (Reference Georgescu-Roegen1971), Janos Kornai (Reference Kornai1971), and Kenneth Boulding (Reference Boulding1981). And there was the work by Dick Nelson and Sidney Winter that would set the tone for the various streams of research that would in our times come to be associated with the label ‘evolutionary economics’.

In this issue, Winter (Reference Winter2014) touches upon the debate about the usefulness of evolutionary arguments in economics triggered by Alchian (Reference Alchian1950) and Friedman (Reference Friedman and Friedman1953).Footnote ⁷ He notes that where ‘Alchian had proposed that evolutionary ideas offered a way for economists to avoid reliance on maximization assumptions, Friedman's similar ideas were advanced as a basis for such reliance’ (Winter, Reference Winter2014: 3). In his PhD thesis, Winter (Reference Winter1964) showed that Friedman (Reference Friedman and Friedman1953) had misused Darwinian ideas, and that the logic of selection did not support Friedman's conclusions. He subsequently developed some of these ideas in a classic article (Winter, Reference Winter1971). Winter's seminal collaboration with Nelson led to a number of pioneering papers and the outline of a new evolutionary approach (Nelson and Winter, Reference Nelson and Winter1973, Reference Nelson and Winter1974, Reference Nelson and Winter1977). The big impact came with their Evolutionary Theory of Economic Change (Nelson and Winter, Reference Nelson and Winter1982).

A major feature of the post-1982 evolutionary economics is its reversal of Friedman's conclusions. Moreover, Nelson and Winter (Reference Nelson and Winter1982) showed how evolutionary thinking could help overcome two lacunae in the optimization-cum-equilibrium paradigm of neoclassical theory: the analysis of administrative behaviour and technical change. Their seminal work can be understood as using evolutionary thinking to link insights from work on administrative behaviour by the Carnegie School (Cyert and March, Reference Cyert and March1963; Simon, Reference Simon1947) to Schumpeterian innovation themes. In doing so, they explicitly adopted the metaphor of biological evolution as the starting point for a theory that explained economic change as involving variation, selection, and retention in populations of firms, resulting in a changing ‘gene’ pool of organisational routines.

In spite of Nelson and Winter's use of the Darwinian variation, selection, and retention logic, the post-1982 wave of evolutionary economics emerged without any explicit connection to Darwin or Veblen. In their book, Nelson and Winter (Reference Nelson and Winter1982) only mentioned Darwin once in passing. Instead, they saw their approach as Lamarckian.Footnote ⁸ The affinity of their work with Veblen's went unnoticed, and instead a foremost accolade went to Schumpeter: their approach was described as ‘neo-Schumpeterian’ and the label stuck.Footnote ⁹

Of course, there were good reasons to celebrate Schumpeter. He had highlighted the vitality of entrepreneurship and innovation in economic development and forcefully articulated the limitations of the optimization-cum-equilibrium approach of neoclassical theory in addressing these phenomena. Schumpeter (Reference Schumpeter1934) saw the central dynamism of capitalism as entrepreneurship changing the economic system from within through relentless processes of innovation: ‘This process of Creative Destruction is the essential fact about capitalism . . . a theoretical construction which neglects this essential element of the case neglects all that is most typically capitalist about it; even if correct in logic as well as in fact, it is like Hamlet without the Danish prince’ (Schumpeter, Reference Schumpeter1942: 83–86). His point, of course, was that in its worship of equilibrium analysis, neoclassical theory had edited Hamlet out of the drama of restless capitalism.

There are clear parallels between Schumpeter and Veblen.Footnote ¹⁰ Together they emphasised economic dynamics, the driving power of technology, and positive feedbacks leading to cumulative change. But there are also important differences. In contrast to Veblen's approach, Schumpeter's evolutionary economics drew little inspiration from Darwin (Hodgson, Reference Hodgson1993, Reference Hodgson1997a; Witt, Reference Witt2002). Schumpeter explicitly wanted to avoid biological language, and he developed his arguments without giving the population logic (which underlies Darwin's theory) a central role in explaining economic change. In this sense, Nelson and Winter's approach, with its emphasis on variation, selection, and retention, is more Veblenian than Schumpeterian. On the other hand, institutional analysis – which was central to Veblen's approach – remained in the background. It received greater attention only later, in Nelson (Reference Nelson1993), Nelson and Sampat (Reference Nelson and Sampat2001), Nelson and Nelson (Reference Nelson and Nelson2002), Nelson (Reference Nelson2005), and other works.

It is interesting to note how the three main themes in Nelson and Winter's (Reference Nelson and Winter1982) work (the biological metaphor, administrative behaviour, and innovation) have fared more recently in modern scientific thought. First, Schumpeterian innovation themes are very much central to modern evolutionary economics, and many would probably first and foremost associate the label evolutionary economics with work on technological innovation and industry evolution. Second, the micro-orientation on administrative behaviour has been taken up much more enthusiastically in management science and organisation studies than in evolutionary economics itself. In evolutionary economics, the focus is primarily on the meso-level (technological regimes, firm networks, industries, clusters, and regions) with increasing links to macro-analysis (innovation systems, varieties of capitalism), which brings institutional analysis back.Footnote ¹¹ Third, the Darwinian logic has continued to play a metaphorical role, but it has not become the common analytical core of evolutionary economics. In fact, the ongoing lack of a common core has triggered a recent stream of research on the ontological foundations of evolutionary economics (Hodgson, Reference Hodgson2002, Reference Hodgson, Davis and Wade Hands2011; Hodgson and Knudsen, Reference Hodgson and Knudsen2010, Reference Hodgson and Knudsen2012; Stoelhorst, Reference Stoelhorst2008b; Vromen, Reference Vromen and Mäki2001; Witt, Reference Witt2008). The contributions of Witt (Reference Witt2014) and Stoelhorst (Reference Stoelhorst2014) to this issue fall within this latter stream of research.

Why institutional economics needs to be evolutionary

While some new institutional economists such as Douglass North (Reference North1990, Reference North2005) explicitly focus on economic development and growth, transaction-cost approaches have so far been largely confined to equilibria or comparative statics. This is true of the classic works by Coase (Reference Coase1937) and Williamson (Reference Williamson1975, Reference Williamson1985). The dynamic theory is less well developed, and its importance has been a rallying cry for complementary developments and rival approaches (Foss, Reference Foss1993; Langlois, Reference Langlois1992, Reference Langlois2007; Nooteboom, Reference Nooteboom1992; Teece and Pisano, Reference Teece and Pisano1994; Teece et al., Reference Teece, Pisano and Shuen1997, Tunzelmann and Wang, Reference Tunzelmann and Wang2003). An adequate theoretical approach to the study of institutions must address dynamics as well as statics. Although static analysis can be useful, the dynamic theory is vital. This is the first reason why institutional economics could benefit from closer links to evolutionary economics, in which the dynamic theory takes centre stage by definition.

In fact, even within the realm of static analysis, the dynamic theory is often necessary to justify some basic claims. For example, Williamson (Reference Williamson1975, Reference Williamson1985) argued that evolutionary competition leads to greater efficiency. He claimed that evolution leads to the selection of relatively superior or near-optimal organisational forms. He concluded that because hierarchical firms exist, they must be both more efficient and more suited to survival than cooperative or less hierarchical firms.Footnote ¹² But the evolutionary analysis behind Williamson's claim is threadbare. The detailed dynamics of evolutionary selection reveals that special conditions are required for competition to lead to the survival of the globally more efficient, more productive, or more profitable firms (Boyd and Richerson, Reference Boyd and Richerson1980; Hodgson, Reference Hodgson1993, Reference Hodgson1994; Schaffer, Reference Schaffer1989; Winter, Reference Winter1964, Reference Winter1971). For example, there may be scale effects, where potentially more efficient firms are prevented from entering an industry because they are out-competed by other firms before they can reach a size to benefit from economies of scale. There can also be frequency effects, where incumbent firms benefit (through synergies or externalities) from the existence of other firms of the same type, and potentially more efficient firms cannot reach the required critical mass to benefit from others in a similar way. Detailed analysis and empirical evidence are necessary to establish particular conclusions from evolutionary theory.

Rigorous evolutionary analysis must distinguish between developmental processes governing singular entities on the one hand, and selection, diffusion, and other processes relating to populations on the other. When addressing populations – such as competing business firms or political states – both developmental and selection processes must be considered. Developmental processes, such as self-organisation, are themselves honed and modified by selection and diffusion. The over-arching theoretical structure developed to deal with these multiple, complex processes is known as generalised Darwinism (Aldrich et al., Reference Aldrich, Hodgson, Hull, Knudsen, Mokyr and Vanberg2008; Hodgson and Knudsen, Reference Hodgson and Knudsen2010; Pelikan, Reference Pelikan2011; Stoelhorst, Reference Stoelhorst, Finch and Orillard2005, Reference Stoelhorst2008b). Combined with a careful consideration of the interacting units at different levels of analysis (i.e. individuals, groups, firms, and broader institutions), it holds out the promise of making institutional dynamics susceptible to rigorous analysis.

In addition to offering a general analytical framework for dynamic analysis, evolutionary theory provides additional theoretical cues for the analysis of institutions. A basic explanatory requirement is that suppositions about individuals or organisations must be consistent with our knowledge of evolutionary processes, in both cultural and biological spheres. At least in principle, there must be some evolutionary explanation of key assumptions. This has been termed ‘the principle of evolutionary explanation’ (Hodgson, Reference Hodgson2004). Concerning individuals, assumptions made about their rationality (or otherwise), capacities, and dispositions must be consistent with our theoretical and empirical knowledge of human evolution. Veblen (Reference Veblen1898b: 188) implied that our presumptions should be consistent with human evolution when he asked mockingly how self-interested, utility-maximising economic man ‘has achieved his emancipation from the law of natural selection’. Concerning organisations and institutions, we must be able to at least sketch a viable evolutionary explanation of their emergence and persistence, which can be subjected to scrutiny in terms of its logic and assumptions.Footnote ¹³

This opens a door to naturalistic approaches. By these we refer to attempts to take biological as well as cultural evolution into account in our understanding of human nature and institutions.Footnote ¹⁴ The general claim is that both biological and cultural processes matter in human society (Boyd and Richerson, Reference Boyd and Richerson1985), although their relative causal importance may be open to dispute (Hodgson, Reference Hodgson2007). It does not amount to biological reductionism, where attempts are made to explain everything in genetic or other biological terms. The relative importance of biology and culture is largely a matter of empirical investigation, rather than of prejudicial dogma. Work in this area is bringing important insights concerning the explanation of human cooperation, moral motivation, and the theory of the firm (Bowles and Gintis, Reference Bowles and Gintis2011; Hodgson, Reference Hodgson2013; Stoelhorst, Reference Stoelhorst, Finch and Orillard2005; Stoelhorst and Richerson, Reference Stoelhorst and Richerson2013; Witt, Reference Witt2003). Notably, both evolutionary naturalism and generalised Darwinism were promoted in the original institutional economics of Veblen (Camic and Hodgson, Reference Camic and Hodgson2011; Hodgson, Reference Hodgson2004).

A further reason why institutional economics needs to be evolutionary concerns the need to bring technology into the analysis of institutional change and economic development more generally. The post-1982 evolutionary economists have been at the forefront of modern thinking about technological change (Dosi, Reference Dosi1982, Reference Dosi2000; Dosi et al., Reference Dosi, Freeman, Nelson, Silverberg and Soete1988; Lipsey et al., Reference Lipsey, Carlaw and Bekar2005; Metcalfe, Reference Metcalfe1994, Reference Metcalfe1995, Reference Metcalfe1998; Nelson, Reference Nelson1995, Reference Nelson2005; Nelson and Winter, Reference Nelson and Winter1977, Reference Nelson and Winter1982; Saviotti, Reference Saviotti1996). The technological lacuna in transaction cost economics has been noted by Williamson (Reference Williamson2000: 60) among others. Ménard (Reference Ménard2014) devotes significant attention to this in his paper in this issue. All forms of institutional economics need to take on board the rich insights of modern evolutionary economists regarding technology.

In standard transaction cost analysis, technology is taken as given, and different governance modes are then compared. This implies the possibility of a hermetic separation of governance structures on the one hand from technology on the other, and between transaction costs and production costs. But Paul Milgrom and John Roberts (Reference Milgrom and Roberts1992: 33–34) have highlighted some of the problems involved in trying to separate production and institutional governance, and their corresponding costs. The transaction costs argument assumes that production costs are given and do not differ across governance or transaction modes. But technologies are often linked with transaction modes and structures of governance. When technology is endogenously determined, its choice may be for reasons other than cost minimisation. This becomes clearer when we dispense with a purely ‘engineering’ view of production and see production costs as also affected by social relations between agents. Clearly, the ability and motivation of workers to learn will often depend on the organisation of production. This reinforces the argument that production costs cannot be independent of social relations.

Several important works underline the interdependence and interaction of organisations and technology (Button, Reference Button1993; Orlikowski, Reference Orlikowski2010; Pagano and Rowthorn, Reference Pagano and Rowthorn1994; Zuboff, Reference Zuboff1988). Ugo Pagano (Reference Pagano1991, Reference Pagano1992, Reference Pagano2011) and Vernon Ruttan (Reference Ruttan1997, Reference Ruttan2006) have explored techno-institutional complementarities and shown that the causality between the two can run in both directions. Consequently, institutional economics has to take the role of technology on board. Holding technology constant when analysing organisational or governance structures can be a useful first step, but it is highly preliminary until techno-institutional interactions and complementarities are brought into the analysis.

Why evolutionary economics needs to be institutional

Evolutionary economics has a number of inspirational texts but no generally agreed upon conceptual framework. From whatever chosen perspective, an evolutionary economist has to ask: what evolves? One answer may be technology, as famously emphasised by Schumpeter (Reference Schumpeter1942), Ayres (Reference Ayres1944), and many others. But technology does not evolve of its own account, without the intervention of human agency. Inventors and innovators require motivation, incentives, and resources. The development of science requires networks and institutions of scientists, plus political institutions allowing sufficient freedom of expression to debate ideas that may counter conventional or religious wisdom. Financial institutions have to be sufficiently well developed to fund large-scale scientific or technical projects. In short, technology cannot evolve without suitable legal, political, and financial institutions. The evolution of technology is unavoidably entwined with the evolution of institutions.

Early accounts of social evolution confined their account to the evolution of the attributes of individuals. Social evolution was seen simply as aggregated individual evolution. Both Herbert Spencer and Alfred Marshall took this view (Hodgson, Reference Hodgson2004). A major breakthrough occurred with the works by David Ritchie (Reference Ritchie1896) and Veblen (Reference Veblen1898a, Reference Veblen1899), who first wrote on the evolution of institutions themselves, applying Darwinian principles to their selection and development. Institutions, as well as individuals, were discovered as objects of evolutionary selection.

Whatever theoretical framework is adopted, social and economic evolution must focus principally on the systems of rules and habits that are ingrained in institutions and cultures. As both Veblen (Reference Veblen1914) and Hayek (Reference Hayek1973) asserted in different ways, humans are rule-following animals. Veblen took this down to the level of individual psychology, stressing the importance of shared habits and instincts in sustaining institutions. But because institutions are social structures, and culture involves social relations, they are more than the sum of their individual human parts. Accounts of social evolution must address the evolution of relations or structures, and not simply the attributes of individuals, either aggregated or taken severally. Both Veblen and Hayek understood this. What is needed is an analysis in terms of multi-level interactions between individual instincts, habits, and deliberate choices of individuals on the one hand, and the enabling and constraining role of rule systems on the other. In the words of North (Reference North2005: 49): ‘There is an intimate relationship between belief systems and the institutional framework’.

The post-1982 evolutionary economics has paid much attention to the firm, but much less to other institutions. Nelson and Winter (Reference Nelson and Winter1982) ‘started in the middle’ by assuming the existence of firms (Stoelhorst, Reference Stoelhorst2008a). A fuller account of the vital role of firms requires a combination of considerations at the individual level of analysis as well as multiple institutional levels. In particular, the role of legal institutions in the formation, evolution, and integrity of firms has to be taken into account. The role of financial institutions, especially when it comes to innovation and investment, is also crucial. Furthermore, the productive capacities of the firm cannot be understood fully without recognition of the (partly instinctual and partly habitual) human foundations of cooperation in relatively small groups (Dunbar, Reference Dunbar2011; Hodgson, Reference Hodgson2013; Stoelhorst and Richerson, Reference Stoelhorst and Richerson2013). The evolution of firms and other institutions involves multiple-level evolution and interaction. Perhaps surprisingly, a theoretical framework to deal with institutional evolution – even if it were confined to firms – remains remarkably underdeveloped within the post-1982 evolutionary economics, and relatively little effort has been made to make use of theoretical material from elsewhere to help develop such a framework.Footnote ¹⁵

But it is clear that evolutionary economists and other scholars working on technological change have become increasingly aware of the importance of taking institutions into account. Consider, for example, the literature on national systems of innovation (Amable, Reference Amable2000; Edquist, Reference Edquist1997; Lundvall, Reference Lundvall1992; Nelson, Reference Nelson1993; Nelson and Nelson, Reference Nelson and Nelson2002; Saviotti and Nooteboom, Reference Saviotti and Nooteboom2005; Tylecote, Reference Tylecote2007). What is lacking is a systematic approach, with some consensus over key definitions and other matters, so that the roles and dynamics of institutions can be researched and understood more fully.