Version classiqueVersion mobile

La Théorie de la Régulation au fil du temps

Robert Boyer

Chapitre 7. Dialogue avec les recherches institutionnalistes

7.3 The Economics of Institutions in Japan: a critique to the comparative institutional analysis

Akinori Isogai et Hiroyasu Uemura

Texte intégral

“Restoration” of Institutionalism in Japan

1“Institutionalism” has become magnetism which attracts many economists, and a lot of researchers have recognized that “institutions matter” in understanding socioeconomic phenomena in economics as well as other areas of the social sciences.

  • 474 Hodgson 1989.

2First of all, we will explain how institutionalism has been restored in Japan. Even in Japan situated in the Far East, we have a tradition in studying American institutionalism which is usually represented by Thorstein Veblen, John R. Commons, and Wesley Clair Mitchell. However, they have been discussed mainly as a subject in the history of economic thought except Shigeto Tsuru, who had enough farsightednesses to study contemporary capitalism from the institutionalist point of view. As the “old” institutionalism had a flaw in thinking that they could reject the orthodox economics only by pointing out its unrealistic and illusory character474, most study of the “old” institutionalism in Japan has been a kind of theoretical “anemia”.

  • 475 Aoki (ed.) 1977.

3In a sense, we can find a pioneering “theoretical” work of the economics of institutions in articles in Masahiko Aoki (ed.) The Theory of Economic System475, in which Aoki argued that there are several approaches such as statistical approach, mathematical-economic approach, political-economic (or game-theoretic) approach, and institutional and historical approach to four levels of the economic system, that is, “real”, “informational”, “political”, “cultural (conventional)”, and that we should use them complementarily in analyzing the economic system. He also empathized there that making a bridge between the orthodox analysis of economic mechanism and institutionalist analysis is a key to develop the theory of economic system. A problématique to lead to the Comparative Institutional Analysis (CIA) emerged in Masahiko Aoki’s study of “comparative economic systems” in the 1970s. At that time, however, it was still assumed that each economic system is independent of each other. Although the variety of economic system is recognized, interactions between the different economic systems and cumulative changes in institutions were not considered.

  • 476 Tsuru 1992; Inoue & Yamada 1994; Hirano 1996; Ebizuka, Uemura & Isogai 1997.
  • 477 Shiozawa 1997.
  • 478 Yagi et al. (eds) 1997.

4Non-orthodox economists started to pay attention to the economics of institutions as they introduced heterodox economics such as the Theory of Regulation and the Social Structures of Accumulation (SSA) in the 1980s, and some results of empirical study have been produced from this perspective476. Furthermore, there are a series of studies by Yoshinori Shiozawa who has empathized the importance of “reproduction” and “stationarity” in the Classical framework. He has developed his theory recently so as to integrate “the economics of complexity” into it477. The establishment of “the Japan Association for Evolutionary Economics (JAFEE)” marked a new epoch in the economics of institutions in Japan. Especially, regulationists as well as many other non-orthodox economists participate actively in JAFEE478. In the following, we will summarize critically the core argument of CIA, and will try to compare it with our own theoretical framework of the economics of institutions.

Logical Structure of Comparative Institutional Analysis and Its Problems

  • 479 Aoki 1988.
  • 480 Aoki 1995a.
  • 481 Aoki & Okuno-Fujiwara (eds) 1996.

5Masahiko Aoki was truly remarkable for his contribution to the analysis of the Japanese economic system in the 1980s479. In the 1990s, he formalized the Comparative Institutional Analysis, after the Japanese economy lost its luster and the hybridization of the Japanese system and the American one came to be discussed480, Masahiko Aoki and Masahiro Okuno-Fujiwara (eds.)481. CIA defines institutions as equilibrium rather than the rules of game by using the evolutionary game theory.

Historical path-dependence

  • 482 Aoki 1995b.

6The emergence of two distinct organizational forms is analyzed as a result of the strategic choice of agents on the basis of the system of information and the type of skills. The system of information is classified into “assimilated information structure” and “differentiated information structure”, depending on how to cope with both system shocks and individual shocks. The type of skill is classified into “contextual skills” and “functional skills”, depending on whether it is specific to a workplace in a certain firm. This is determined by the type of human capital investment. Agents with bounded rationality play a game in terms of two systems of information and two types of skills. In the framework of evolutionary game theory, strategic complementarity finally dominates, and it produces three evolutionary stable equilibrium which are called “A-equilibrium” (= U.S.), “J-equilibrium” (= Japan) and “P-equilibrium” (= the Pareto optimal equilibrium). Each equilibrium represents a certain type of institutional arrangements with “institutional complementarity”. It is, however, emphasized that “since exogenous parameters do not uniquely determine the selection of equilibrium (thus institutional arrangement), there is no guarantee that institutional arrangements are efficient or converge to an efficient one482”. In other words, the institutional arrangements in an economic system have “historical path-dependence”.

  • 483 Williamson 1975.

7CIA must be a more sophisticated analysis of institutions than that of New Institutionalism483, but we think that there are some crucial problems, both theoretical and empirical. As for a theoretical problem, what we empathize at first here is the fact that the system of information and the type of skills are not separable, because how information flows depends on worker’s skills and supervisory relations. Therefore, the random-matching game on the basis of the separation between information and skills is no more than a theoretical fable. The real process may be a specific hybridization of different economic systems which have been formed in history. Furthermore, pay-off matrix cannot be seen as a stable structure in an evolutionary process with changes in the technology of firm and the perception of agent.

  • 484 Koike 1988; 1991.
  • 485 Nomura 1993.
  • 486 Asanuma 1997; Kaneko 1997.

8As for an empirical problem, Masahiko Aoki’s concept of “contextual skills” depends very much on Kazuo Koike’s empirical study of “intellectual skills” and “firm-specific skills484”. However, it seems to be greatly exaggerated by omitting the existence of “maintenance employees485”. Furthermore, we should pay more attention to the fact that a “group leader” supervising workers in a group plays a role as a conjunction between information flow and the hierarchical chain of command486. More careful empirical studies are being pursued by taking account of a difference in the industry and firm-size.

Comparative Institutional Analysis vs. the Institutional Analysis of Socio-Economic System

9We call our theoretical framework “the Institutional Analysis of Socio-Economic System (IASES)”, and compare its basic concepts with those of CIA.

10Economic agent: Both CIA and IASES think that the rationality of agents is bounded, but IASES argues that not only rationality is bounded because of the limit of the ability of computation and data processing, but also the behavior of agent is constrained in a social context. In other words, an agent tries to keep his or her identity by internalizing complex and over-determined social relations which are not necessarily consistent in the socioeconomic system. Therefore, an individual agent has multiple characters in relation to different social contexts, while he or she internalizes habits and rules with a conflict in mind.

  • 487 Veblen 1899.

11Institution: In CIA “institution” is defined as a self-enforcing equilibrium at which an agent can get benefit from adopting a dominant strategy, that is, “evolutionary stable equilibrium”. In contrast to this, IASES thinks that a following Thorstein Veblen’s definition is appropriate in a broader sense: “The institutions are prevalent habits of thought with respect to particular relations and particular functions of the individuals and of the community487”. In a much narrower sense in regard to social institutions, institution is defined as a social apparatus which guides agents to particular habits of thought and behavior, giving them incentives to lead a conforming behavior and imposing sanctions against a deviating behavior. In this sense, institutions provide a basis for the “socialization” of economic agent.

12Firm organization: In the analysis in CIA, information structures and incentive mechanisms within a firm organization are considered. The comparative analysis of prototypical models of firm organization is made on the basis of the understanding of symmetrical patterns of such three levels as employment contract, financial contract and decision-making on management which are established in an institutional context specific to each country. This is radically different from the concept of firm in the neoclassical economics which regards firm as a point without internal organization. We think, however, that the CIA’s concept of firm is still static within an equilibrium framework, so that we should develop the analysis of firm from following perspectives. First, firm is a dynamically moving organization which integrates both monetary and real factors. Second, information structures play a role as not only the transmission of information but also the chain of command. In this sense, it should be empathized that information structures and power structures are not separable. In other words, hierarchy within a firm organization promotes the “social integration” of workers by imposing sanctions. Thirds, a firm pursues innovation, while it is changing its environments in both inter-firm relations and the labor market.

  • 488 Hollingsworth & Boyer (eds) 1997.

13Institutional complementarity: In both CIA and IASES, “institutional complementarity” is very important concept for the analysis of the stability of the economic system. In IASES, we think that the concept of complementarity should be understood in more macrodynamic and social framework, that is, “structural compatibility488”, especially, in regard to following three causal relations. First, we should investigate whether institutions are complementary in enhancing productivity growth and demand formation. As Michal Kalecki empathized, one’s costs are identical to one’s expenditures, and they are also other’s incomes. This produces the dynamics of demand formation. As for productivity-enhancing effects, not only such “micro foundation” as technological and organizational innovation but also such “macro foundation” as increasing return to scale in a stable growth process are important. Second, there is over-determined causation in the system of complementary institutions, so the strength of the structural effect of institutions may change drastically in a certain phase of growth. Third, the problem of complementarity should be analyzed in a much wider social context. In other words, even if the economic system is consistent, its compatibility with such social systems as family and community cannot be necessarily ensured. The incompatibility of institutions with social systems has repercussions on macro-dynamics in the long run.

  • 489 Shiozawa 1997.

14Micro-macro loop: Interactions between micro level and macro level are not considered well in CIA which is based on “methodological individualism”. In IASES, we analyze “micro-macro loop”, that is, the loop of interaction between micro behaviors and the macro dynamics of the system489. In the loop, micro behaviors produce regularity in a macro process through complementary coordinating mechanisms, on the one hand, macro dynamics forms environments and imposes various constrains on micro behaviors, on the other.

15Path-dependence: CIA makes much of path dependence, but CIA analyses only by using the framework of multiple equilibrium. Therefore, this can be characterized as the equilibrium-theoretic understanding of path-dependence. On the contrary, IASES pursues the non-equilibrium-theoretic and more dynamic framework of analysis. In this regard, following two points are important. First, path-dependence should be analyzed in the framework of “historical time” (Joan Robinson) or irreversible time rather than equilibrium. Second, the socioeconomic system has a mechanism to promote productivity growth through learning and technological innovation, and this produces cumulative causation (or positive feedback) which makes the economic system path-dependent.

16CIA is being gradually accepted by orthodox economists in Japan. However, facing the internationalization of economic activities and the globalization of the financial market, we should empathize from the viewpoint of IASES as follows. Possible institutional changes will be neither a movement toward a Pareto-equilibrium in a global domain nor changes in the combination of existing institutional components. They will surely imply the creation of new institutions and the newly emerging problems of compatibility with social systems and the reproduction of society at local, national and global levels.



Aoki Masahiko (éd.) 1977, The Theory of Economic Systems: an economic foundation, Tokyo, Toyokeizai Shinposha (in Japanese).

—, 1988, Information, Incentives, and Bargaining in the Japanese Economy, Cambridge, Cambridge University Press.

—, 1995a, An Evolving Diversity of Economic System, Tokyo, Toyokeizai Shinposha (in Japanese).

—, 1995b, « Towards a comparative institutional analysis: motivations and some tentative general insights », The Japanese Economic Review, 47/1, p. 1-19.

Aoki Masahiko & Ronald Dore (eds.), 1994, The Japanese Firm: the sources of competitive strength, Oxford, Oxford University Press.

Aoki Masahiko & Masahiro Okuno-Fujiwara (eds.), 1996, Comparative Institutional Analysis: a new approach to economic system, Tokyo, University of Tokyo Press (in Japanese).

Asanuma Banri, 1989, « Manufacturer-Supplier relationships in Japan and the concept of relationship-specific skill », Journal of the Japanese and International Economics, 3/1, p. 1-30.

—, 1997, Japanese Firm Organisation: the mechanism of innovative adaptation, Tokyo, Toyokeizai Shinposha (in Japanese).

Ebizuka Akira, Uemura Hiroyasu & Isogai Akinori, 1997, « L’hypothèse de “la relation hiérarchisée marché-firme” et l’économie japonaise d’après-guerre : une analyse en termes d’incitation et de flexibilité », L’Année de la régulation : Économie, Institutions, Pouvoirs, vol. 1, p. 297-315 [En ligne :].

Hirano Yumeka, 1996, Japanese Institutions and Economic Growth, Tokyo, Fujiwara Shoten (in Japanese).

Hodgson Geoffrey H., 1988, Economics and Institutions: a manifesto for modern institutional economics, Cambridge, Polity Press.

Hollingsworth Roger J. & Robert Boyer (eds.), 1997, Contemporary Capitalism: the embeddedness of institutions, Cambridge, Cambridge University Press.

Inoue Yasuo & Toshio Yamada, 1994, « La théorie de la régulation au Japon », La Lettre de la Régulation, 11, p. 1-2 [En ligne :].

Isogai Akinori, 1996, « Towards an institutional analysis of the socioeconomic system: a new approach to the economics of institutions », Annals of the Society for the History of Economic Thought, 34, p. 1-13 (in Japanese).

Kaneko Masaru, 1997, Market, Ownership, and the Political Economy of Institutions, Tokyo, University of Tokyo Press (in Japanese).

Koike Kazuo, 1988, Understanding Industrial Relations in Modern Japan, Londres, Macmillan.

Nomura Masami, 1993, Toyotism, Kyoto, Minerva Shobo (in Japanese).

Shiozawa Yoshinori, 1997, An Introduction to the Economics of Complex Systems, Tokyo, Seisansei Shuppan (in Japanese).

Tsuru Tsuyoshi, 1991, « Unit labour costs, the reserve army effect, and the collective bargaining system: a U.S.-Japan comparison », in Toshiyuki Mizoguchi (éd.), Making Economies more Efficient and More Equitable, Tokyo, Kinokuniya LTD & Oxford University Press, p. 407-435.

Uemura Hiroyasu, Isogai Akinori & Akira Ebizuka, 1997, « Money and labour and “the economics of institutions”: towards an institutional analysis of the socioeconomic system », in Kiichiro Yagi et al. (eds), Papers of the First Annual Conference of the Japan Association for Evolutionary Economics, Kyoto, 28-29 Mars.

Veblen Thorstein, 1899, Theory of Leisure Class, London, G. Allen & Unwin.

Williamson Oliver E., 1975, Markets and Hierarchies: analysis and antitrust implications, New-York, The Free Press.

Yagi Kiichiro et al. (eds.), 1997, Papers of the First Annual Conference of the Japan Association for Evolutionary Economics, Kyoto, 28-29 Mars.


474 Hodgson 1989.

475 Aoki (ed.) 1977.

476 Tsuru 1992; Inoue & Yamada 1994; Hirano 1996; Ebizuka, Uemura & Isogai 1997.

477 Shiozawa 1997.

478 Yagi et al. (eds) 1997.

479 Aoki 1988.

480 Aoki 1995a.

481 Aoki & Okuno-Fujiwara (eds) 1996.

482 Aoki 1995b.

483 Williamson 1975.

484 Koike 1988; 1991.

485 Nomura 1993.

486 Asanuma 1997; Kaneko 1997.

487 Veblen 1899.

488 Hollingsworth & Boyer (eds) 1997.

489 Shiozawa 1997.

© Éditions des maisons des sciences de l’homme associées, 2018

Conditions d’utilisation :


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search