Navigation – Plan du site
Opinions - débats

Fostering Variety in Economics. Interview with Geoffrey Hodgson

Geoffrey Hodgson, Agnès Labrousse et Julien Vercueil

Texte intégral

1After being a university lecturer in economics at the University of Cambridge, Geoffrey Hodgson is currently a research professor in business studies at the University of Hertfordshire in England. He is director of the Centre for Research in Institutional Economics and editor-in-chief of The Journal of Institutional Economics. He is interested in « the big questions of how socio-economic systems work, and how individuals relate to economic institutions ». His books include Economics and Institutions (1988), a « critique of the core assumptions of neoclassical economics and an attempt to outline an alternative approach », Economics and Evolution (1993), studying « the role of metaphor in economics and the sources of creativity, combined with a concern about the fashionable but often ill-informed use of the ‘evolutionary’ label », Economics and Utopia (1999), discussing « possible scenarios for a post-capitalist future and examining the implications of the learning economy. », and How Economics Forgot History (2001), an exploration of « the problem of historical specificity in economics, and the past contributions to this methodological issue by members of the German historical school and others ».

2Distinguished by numerous honours and awards, having published in various academic journals, including the Journal of Economic Literature, Economic Journal, Journal of Economic Behavior and Organization, and Cambridge Journal of Economics, G. Hodgson was listed as one of ‘The Top 40 Living Economists on the Net’ by the Post-Autistic Economics Network (November 2000).

3RR. : Although they are not new in economics, evolutionary and institutional approaches have attracted a growing interest since the 1980s. How do you explain that?

4GH. : When I started working on institutional and evolutionary ideas in the early 1980s, I felt almost a lone voice among economists. Now the situation is very different. There are several reasons for the rise in interest in institutional and evolutionary ideas. Until the 1970s, much of the challenge to mainstream ideas was in macroeconomics, reflecting the influence of Keynesian and Marxian approaches. With some exceptions, such as the work of Herbert Simon, orthodox micro-theory faced less opposition. To a significant degree, the rise of institutional, evolutionary and behavioural approaches reflects the growing development of alternative theories in the microeconomic arena. These became significant in the 1980s, notably with the work of Richard Nelson, Sidney Winter and Oliver Williamson. By the 1990s, some of these critical movements had had a major impact on mainstream approaches, and others became more prominent among heterodox economists. The new institutionalists Ronald Coase and Douglass North were awarded Nobel Prizes in 1991 and 1993. The various forms of institutional and evolutionary thought tackle important questions concerning institutions and economic development that were relatively neglected from the 1950s to the 1970s.

5RR. : Could you present the core propositions of your institutional and evolutionary economics?

6GH. : Both ‘institutional economics’ and ‘evolutionary economics’ are very broad churches. There are as many important differences within both the original and new institutional economics as there are between them. Furthermore, the Nelson-Winter evolutionary approach is very different of evolutionary game theory, and so on. No single approach, including the one I favour, can be representative of institutional or evolutionary economics as a whole.

7The first proposition I would underline concerns the nature of social and economic reality. Institutions, defined as systems of shared rules, are the stuff of all social and economic life. Systems of rules – or institutions – make up the social fabric and provide the means by which we communicate, interact and cooperate.

8The second set of ideas concerns the nature of the individual psyche. Humans have limited cognitive capacities. Our brains have evolved to deal with interactions with others in (relatively small) social groups, and with problems of survival in complex and changing environments with limited information and radical uncertainty. We also have to appreciate the evolutionary context to understand our mental nature and its limits. We are not optimizers – we have survived over millennia by copying others, and testing rough-and-ready solutions. The complexity of the problems involved means that full, rational, deliberation upon every dilemma would be dysfunctional. We acquire skills and internalize them as habits. Habits are acquired dispositions to think or act in various ways. Human intelligence is not only bounded, it also relies essentially on tacit ideas and habits upon which we do not fully deliberate and sometimes we find difficult to articulate. Our individual repertoire of habits and skills itself evolves through time, and this means that our individual preferences also evolve; they are affected by the institutional context and the cognitive rules that we use to interpret information.

9For the above reasons I believe that game theory, while often of heuristic value, is of limited importance as a theoretical tool to explain the real world. Full-blown models of individual interaction in game theory, where every possible human interaction and defined response is considered, and every agent assumes that every other is fully rational, have fallen into widely acknowledged problems of tractability and relevance. In response, Alan Kirman and others have hinted at an altered direction of research, involving a world where rational capacities are bounded, and specific institutions structure agent interactions. The work of Philip Mirowski on markets is also useful in this respect. A principal focus is on the constraining and shaping roles of the institutional structure, rather than the infinite regress of rational decisions that take into account the rational decisions of others, who in turn take into account the rational decisions of others, and so on.

10Key ideas concerning habits, bounded rationality and institutions were articulated in the early part of the twentieth century by original institutional economists such as Thorstein Veblen and John Maurice Clark. They connect with the evolutionary aspect of the story in a number of ways. First there is the emphasis on the context and constraints of human evolution. Second, as Veblen saw clearly, habits and institutions can be treated as objects of selection in an evolutionary process. They are evolutionary entities at the social level, but this does not mean that explanations of social phenomena can be reduced to biological terms. I am not a biological reductionist. Nevertheless, Darwin’s core evolutionary principles of variation, selection and inheritance apply to all populations of complex entities, including institutions and firms.

11There are many more ideas that I consider to be important, but I would start with the above.

12RR. : There has been large discussions among economists about what should – and what should not – be considered as an institution. To what extent the definition of what is an institution is determinant to the development of institutional theories?

13GH. : Much of the discussion over the definition of an institution has been confused and muddled by the claim that Douglass North does not regard organisations as institutions. While Doug is insufficiently clear on this point, he does not actually say that. A few years ago I corresponded with him on this topic and he acknowledged that organizations are also institutions. (Extracts from this correspondence are published in an article of mine in the March 2006 issue of the Journal of Economic Issues).

14Another controversy concerns when we define institutions as either systems of rules or equilibrium outcomes of a game. I think that the definition of an institution as an equilibrium is inadequate because definitions should focus on essences rather than outcomes. Institutions are relatively durable phenomena and they are equilibria in this sense. However, the nature of the equilibrium is itself defined in terms of rules and strategies, and what lies behind the equilibrium are rule-driven dispositions and rule-framed cognitions of individual agents.

15While definitions themselves do not constitute theoretical explanations, the lack of sufficient consensus over the meaning of words is a severe impairment to theoretical enquiry. That is why definitions are important. My own definitions in this area are as follows:

Social structures include all sets of social relations, including the episodic and those without rules, as well as social institutions.

Institutions are systems of established and embedded social rules that structure social interactions.

Rules in this context are understood as socially transmitted and customary normative injunctions or immanently normative dispositions, that in circumstances X do Y.

Conventions are particular instances of institutional rules. (Note that this definition differs from that of the French école des conventions – including André Orléan and others – who define conventions as entities similar to institutions by my definition.)

Organizations are special institutions that involve

    • (a) criteria to establish their boundaries and to distinguish their members from non-members,

    • (b) principles of sovereignty concerning who is in charge and

    • (c) chains of command delineating responsibilities within the organization.

16RR. : When have you first heard of the French regulation school? Is Regulation theory actively discussed in some circles in Great Britain? If so, which ones?

17GH. : I first came across the French regulation school when I read the 1979 English translation of Michel Aglietta’s A Theory of Capitalist Regulation. I was impressed by the way in which he used Marxist theory in a productive and non-dogmatic manner. These ideas also made an impact on the small group of Marxist economists in British universities. However, many in that group were more dogmatic, and a combination of internal failings, sectarianism and external pressures led to a severe decline in their numbers in the 1980s. Consequently, the French regulation school has had a limited impact among British economists. Its works and ideas are sometimes mentioned, but there is no discernable regulation school in my country. Its impact upon British academic sociologists and political scientists has been greater to some degree, notably with the work of Bob Jessop and a few others.

18RR. : The French Regulation theory is rooted in an open-ended and open-minded Marxism integrating Keynesian and institutionalist insights. What remains of Marx and Keynes in your own approach?

19GH. : I regard Marx and Keynes as extremely important: they both made a major contribution to our understanding of how capitalism works. Part of the enduring significance of Marx lies in his understanding of capitalism as a historically specific system, and his focus on the characteristic relations and mechanisms within this structure. Keynes is important for his emphasis on the key role of radical uncertainty in capitalist markets, and how that may lead to recession and unemployment.

20Nevertheless, while I am sympathetic to Marx’s theoretical approach I think it remains highly limited. Marx never understood the key role of markets in any complex economic system. Behind this lay his inadequate understanding of the way human knowledge is largely tacit, embodied in customs and institutions, and locally parcelised. These flaws and omissions have had enormous negative consequences for Marxism, in theory and practice. Marxism has failed to get to grips with the modern world, and as I argue in my book Economics and Utopia, it is inadequate to deal with complex, knowledge-intensive economies.

21Keynes also remains an inspiration but I am uneasy about the ‘Post Keynesian’ label, partly because it implies that Keynes was the most important watershed in the development of economic theory. In fact – as with many other great writers – several of Keynes’s ideas were prefigured by others, including American institutionalists and members of the German historical school. Some Post Keynesian economists preoccupy themselves with interpreting what Keynes really meant, and there is insufficient self-criticism and innovation in this school of thought. The 1970s project of the Cambridge school and others of building a heterodox alternative around figures such as Marx, Kalecki, Keynes and Sraffa was ill-planned at the outset, as it papered over too many cracks. It has since proved unviable and no longer holds sway in any major university.

22Furthermore, I regard several other economists as being of equivalent importance to Marx and Keynes. I would add the names of Alfred Marshall, Thorstein Veblen, Joseph Schumpeter and Friedrich Hayek, among others. These additional writers have focused on important topics such as habits, institutions, knowledge and technology, which are treated inadequately in the Marxian and Keynesian traditions.

23RR : The Regulation approach, like your own approach, can be viewed as a trend in a larger family of socioeconomic theories characterized by historical institutionalism. Which are, according to you, the most important convergences and differences between your approach and the regulationist one?

24GH. : I find much that is inspiring and useful in the regulation school approach. Writers in this tradition have developed an enlightening framework for understanding different phases of capitalist development and types of capitalist formation. This framework addresses employment relations, forms of competition, financial institutions, modes of consumption, forms and degrees of state intervention and international structures. Several different types of combination of these factors are explored. This acknowledgement of different varieties of developed capitalism is an important advance, both on neoliberal economics and traditional Marxism, who by contrast emphasise one type of capitalism and one path of capitalist development.

25On the other hand, regulation school writers have paid relatively little attention to important microeconomic aspects of analysis, including how agents process information and how knowledge is formed and transmitted. Unlike the original institutionalism of Veblen and others, there is no developed and widely accepted psychological analysis to underpin regulation theory. But there is no reason why this omission could not be rectified, and this creates an important opportunity for dialogue between institutional and regulation school economists.

26RR. : The Regulation theory has stressed the decisive role of political conflicts and compromises, the centrality of crises in economic evolution. Do you think this perspective on economic history is rather complementary or at odds with your Darwinian conception of economic evolution?

27GH. : Darwinism does not imply that evolution is always slow or at constant speed. As Stephen Jay Gould and others have shown, biological evolution has been ‘punctuated’ by spurts of more rapid change. This observation is perfectly compatible with a Darwinian perspective. What Darwin himself did emphasise is that all evolution occurs in linked chains of cause and effect, and no ‘leaps’ from one state to another are possible without multiple, intermediate, incremental stages.

28I agree very much with the regulationist emphasis on politico-economic compromises, and the way in which they sometimes break down and in turn lead to new politico-economic accords. Indeed, I have emphasised the role of such disruptions in leading to change in several of my own publications. Institutions are often resistant to change: without major turbulence or disruption they remain inert. As in biological evolution, change has an up-side and a down-side. On the one hand, major disruptions are enormously wasteful and damaging, and often add to human misery. On the other hand, disruptions create opportunities for innovation and radical change towards a better future.

29However, improvement is by no means assured, and only the test of time and practice can validate the outcomes. In highly complex systems the complete, rational design of a superior future system is impossible. Conflict and disruption are important because they can create the opportunity for experiment, but not a guarantee of success.

30RR. : Let us take an example. It seems that if one adopts – as you brilliantly do – the Veblenian-Darwinian standpoint, legal rules are more or less a by-product, an outgrowth of shared habits. Moreover, you quoted recently Edmund Burke’s idea that “custom reconciles us to everything”. Is there a place for conflict and crises in the emergence and evolution of institutions? What is political in your economy?

31GH. : In my work I have emphasised that prominent legal and other rules, to become established and effective, have to be grounded on shared habits. However, because of the extent and complexity of legal systems, not all individual laws can be grounded on habit. What is important for the functioning of modern legal systems is habituated respect for the law in general.

32Furthermore, the causal history is not always in one direction, from habits to laws. On the contrary, in a recent (not yet published) paper I criticise a prevalent view that law always emerges historically from habit, custom and tradition. Neoliberal writers such as Hayek make this mistake. On the contrary, law is a creature of the state, and historically it was the concentration of power in a state apparatus that made law proper possible. Law is very much an outcome of conflict or dispute. If it were not so, custom and tradition would be sufficient, without courts and other legal enforcement mechanisms.

33I quoted Burke to emphasise that we tend to invest customs and laws with normative properties. As children we are taught the right and wrong way of doing things. I believe that this mechanism of normative investment is universal in human cultures. However, this does not mean that existing customs or laws are necessarily valid, according to some acceptable moral code. Instead, my point is that institutions always carry normative or moral overtones, and these are often under-acknowledged in utilitarian or relativist discourse.

34None of my arguments undermine the importance of politics and power. On the contrary, to understand how systems of power work, we need to focus on both the use of force and coercion and the much more subtle mechanisms of customary acknowledgement and habituation that affect the dispositions and preferences of actors. In 1974 the social scientist Steven Lukes published an excellent book on Power. He emphasised that power has several dimensions, including mechanisms that lead actors to change their preferences and beliefs to those more in line with the sentiments of their rulers. This book had a major impact on me and I still recommend it to others.

35RR. : How can one use your approach as a basis for practical prescriptions (management of a firm, economic policy etc.)?

36GH. : In my writing I have prioritised the task of understanding how the world works, as a necessary precondition for any policy prescriptions, rather than focusing primarily on policies themselves. Nevertheless, I feel that some of my ideas do have fairly immediate policy implications and I have devoted specific attention to some policy issues in some essays. For example, the insistence in my 1998 Economics and Institutions book and elsewhere that markets are historically specific institutions undermines the idea that markets are the natural or normal ether of human interaction. Not only are there sometimes alternatives to the market, but also different types of market function in different ways, and may be evaluated differently. Furthermore, as multiple experiences in both transitional and developing countries demonstrate, markets rarely emerge spontaneously and typically have to build up and engineered over time, with the aid of other enforcing institutions, notably the state.

37My emphasis on the routine-based and habit-based nature of knowledge and production has major implications for management practice. It means that any attempt to manage an organization, or bring about internal change, must address and build up routines, including the development of habit-based skills. Less effective routines must be dismantled and new ones must be established. These processes are lengthy and often difficult, especially when organisations are ossified or set in their ways.

38Evolutionary ideas are also helpful for dealing with policy questions. For example, many policy makers believe that unfettered and flexible markets are a general solution to policy problems. However, evolutionary analysis of firms and labour markets shows that competitive selection is operating on multiple levels. What may be most effective in regard to one level of evolutionary selection may be less effective overall. In particular, while free and flexible labour markets may lead to more competitive wage levels, the result would be a greatly enhanced turnover of labour within firms, undermining learning, teamwork and the construction of routines. From the point of view of competition between firms, especially in the modern, knowledge intensive economy, some labour stability within the firm is necessary to bring about learning and interpersonal cooperation. Entirely free and flexible labour markets would make firms less competitive.

39The existence of institutional complementarities, especially between firms and their institutional environment, also undermines the idea of an optimal economic configuration based on free markets. Such complementarities mean that effective possible adjustments in one sphere are limited, given the related institutional configuration. Systems cannot search indefinitely through the space of evolutionary possibilities – they are much confined by history and the structures that they have inherited from their past.

40Generally, I have argued for a mixed, experimental approach to economic problems. Evolutionary theory stresses the importance of variety. Institutional theory recognises the coexistence of both designed and spontaneous institutions, the coevolution of different types of institution, and stresses the general role of institutions as repositories of knowledge.

41From this perspective, the familiar polar ideas of a collectively planned economy and an economy ruled by the market, are both found wanting. Twentieth century attempts to move towards either of these extreme solutions have been unsuccessful. Markets will always be necessary to some extent in a complex developed economy. Obversely, all modern economic systems require some economic guidance and intervention from the state. Within the copious middle ground between these polar extremes, we have a large number of options. Among existing cases I find the Scandinavian model most attractive, because it combines high levels of economic performance with relatively low inequality of income and a strong welfare state. Although the Scandinavian model will require some modification when applied to other contexts, it would be a good approach to promote in the European arena, as an alternative to both neoliberalism and excessive statism.

42RR. : Robert Boyer stressed the point that economics should scrutinize much more the macro-institutional foundations of microeconomics - not denying by the way that there should be micro foundations to macroeconomics. What do you think of such a view?

43GH. : I understand that by macro-institutional foundations, Boyer refers primarily to national institutions that promote related management structures or routines within business firms. He uses this idea to examine possible institutional complementarities and varieties of capitalism. I find his recent work on this to be very interesting and useful, and I agree with most of his arguments.

44In response I have one reservation and some suggestions for further analysis. My reservation is that regulation theorists sometimes overstate the way in which business units follow macro-institutional parameters and underestimate the coexisting variety of different forms of business organisation. For example, empirical studies have shown that neither Taylorism nor Fordism were universal during the ‘Fordist’ era. Paternalist, authoritarian and other forms of business organisation were also widespread and coexisted with Taylorism and Fordism. Furthermore, the restructuring of capitalism in the 1970s and 1980s did not mean that Taylorism and Fordism disappeared. Even if a new phase of capitalist development emerges, business organisations are often resistant to change and take time to change their ways. Furthermore, competition is a blunt instrument that does not always promote advanced over less advanced firms. The lines between different phases and varieties of capitalism are not as clear cut as some regulationist accounts suggest.

45My suggestion for further analysis is to examine more closely the mechanisms of ‘downward causation’ from macro to micro, where they exist. Some of my own work has this aim. What must be avoided is the structuralist mistake of simply assuming that the higher level structures shape lower level entities. Such assumptions with inadequate supporting reasons are widely criticised in the social sciences. Furthermore, the philosophy of science shows that there are different versions of ‘downward causation’ some of which are unacceptable. Downward causation can only work through mechanisms and forces at the lower-level. In particular, it is only at the level of the individual psyche and through structured relations with other individuals that the preferences or mentalities of individuals are changed. As I argue in my recent book Economics in the Shadows of Darwin and Marx, a major mechanism in this process is the formation of new habits.

46RR. : What do you think of the newly flourishing neo-classical research on institutions and institutional performance, which ignores overall the earlier achievements of heterodoxy in this field?

47GH. : Economists often rediscover earlier ideas with inadequate attribution of their original authorship. This is not confined to neoclassical theorists. For example, Marx did not acknowledge that several of his ideas on the falling rate of profit were taken from the work of John Stuart Mill. Keynes did not give adequate attribution of the precursors of his own ideas, particular among the German historical school and the American institutionalists. Austrian thinkers such as von Mises and Hayek ignored a much earlier critique of socialist planning by the German historical school economist Albert Schäffle. A similar pattern of neglect is found among mainstream theorists, compounded by the tragically growing ignorance of the history of economic thought among economists in general

48Nevertheless, I welcome the rediscovery of the importance of institutions by mainstream economists. This offers a major opportunity for dialogue and the reinsertion of some earlier ideas. Some mainstream thinking about institutions is very crude, but other contributions are more sophisticated and important. While it has limitations, the development of the game theoretic analysis of institutions carries important insights.

49Overall, however, the rational choice model that lies behind much mainstream thinking on institutions has revealed its limitations. Either rationality ends up being defined rather vacuously as consistency of preferences (it is vacuous because no two choice situations are ever identical and hence strict consistency is never tested) or it is shown to be inadequate in dealing with questions concerning the nature and evolution of institutions. Alternative approaches, particularly involving a habit-based psychology, have an opportunity to prove their superiority.

50RR. : Let us have an institutionalist view on economic science itself. How could we develop supporting rules and organizations in order to promote alternative indicators and interpretations of the economic world? You are very active in the EAEPE-network and editor-in-chief of JOIE. What could be done further in Europe to give more institutional and organizational flesh to historical institutionalism?

51GH. : In Europe there are several important, overlapping networks that help to sustain and stimulate evolutionary and institutional approaches. However, these are less influential in the Anglophone world, particularly Britain and America. In these two countries, economics has been severely damaged by the post-1960s preoccupation with formalism over substance. A narrower version of economics, centred on models purporting to be contributions to a ‘science of choice’, has become part of the institutional fabric of academia. People like myself, who try to understand real economic phenomena using other methods, are simply dismissed as not being economists.

52Similar tendencies are already found in Continental Europe, although they are less developed. If they spread further, then the global outlook for economics as a discipline is dire.

53We may have to abandon the economics label and find an alternative disciplinary description. Such labels are important to establish territory and legitimacy within university structures. The problem is that there no universally accepted or unproblematic label. ‘Political economy’ has been hijacked by neoclassical public choice theorists. I once submitted a grant application using the term ‘evolutionary economics’. It was submitted to an evolutionary game theorist as a referee, who successfully recommended rejection because there was no evolutionary game theory in my research proposal.

54In this regard, the most important problem for institutional and evolutionary economists within the institutions of academia is to find an academic home. Many like myself have become refugees in business schools, but I do not regard that as an optimal outcome.

55RR. : Is it a relevant strategy to elaborate a more or less unified approach among institutionalist theories, in an attempt to challenge the mainstream’s leadership?

56GH. : In some respects it might seem nice if we all agreed, but is not going to happen anyway. On deeper reflection, variety is the fuel of scientific innovation and we should aspire to any kind of monolithic science. Nevertheless, while being open to persuasion, we should develop our own ideas the best we can, and try to convince others of their worth.

57Heterodox economics as a whole, and the institutional and evolutionary streams within it, are all highly varied internally. There is a huge variety of ontological preconceptions and methodological approaches, even within any one approach. A successful challenge to the mainstream will not come from premature attempts at unification that must inevitably slur over these differences. Instead it will come from a sufficiently powerful and persuasive alternative theoretical approach. Heterodox economists should play much more attention to the priority of developing a good alternative theory.

58RR. : One last word on the relations to other social sciences: to what extent should the toolkit of the institutional economist integrate sources and tools relevant in sociology or history like for instance archive work, qualitative interviews or even participant observation?

59GH. : The boundaries between economics and sociology, and the defining identities of these two disciplines, are no longer clear. Although the Robbinsian definition of economics as the ‘science of choice’ still has a strong influence, it has been undermined by the admission of behavioural approaches and the award of the Nobel Prize in economics to dissenters such as Herbert Simon and Amartya Sen. Furthermore, since the break up the Parsonian consensus within sociology, it has become a fragmented disciplinary terrain, lacking a unified approach or common definition of that subject.

60I have no problem in incorporating ideas from sociology because I find no good reason to draw a boundary between these disciplines. I also think that we should experiment with a variety of analytical tools and methods, including archival research, bibliometrics, citation analysis, text analysis, and so on. Let us experiment, and see what works best.

61RR. : We wish to thank you for participating in this interview in such an inspiring way.

Haut de page

Pour citer cet article

Référence électronique

Geoffrey Hodgson, Agnès Labrousse et Julien Vercueil, « Fostering Variety in Economics. Interview with Geoffrey Hodgson », Revue de la régulation [En ligne], 2 | Janvier / January 2008, mis en ligne le 30 janvier 2008, consulté le 29 mai 2016. URL :

Haut de page


Geoffrey Hodgson

Agnès Labrousse

Articles du même auteur

Julien Vercueil

Articles du même auteur

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page