Hostname: page-component-7b7567c8c-tnzcf Total loading time: 0 Render date: 2026-11-12T06:44:34.169Z Has data issue: false hasContentIssue false
  • English
  • Français

Economic institutions: explanations for conformity and room for deviation

Published online by Cambridge University Press:  09 October 2012

DAVID DEQUECH
DAVID DEQUECH*
Affiliation:
Institute of Economics, University of Campinas, São Paulo, Brazil
*
*Email: dequech@eco.unicamp.br
Rights & Permissions [Opens in a new window]

Abstract

Why do economic agents conform with existing institutions? Drawing on economics and organizational institutionalism, this article identifies existing explanations, extends or refines some of them, suggests new ones, classifies these various explanations, and integrates them into an organized framework. One set of explanations focuses on conformity out of habit. The other refers to more conscious thought and behavior and considers coordination and increasing returns to adoption, social sanctions, informational differences, uncertainty, legitimacy, naturality, and lack of power/resources to deviate.

Information

Type
Research Article
Information
Journal of Institutional Economics , Volume 9 , Issue 1 , March 2013, pp. 81 - 108
Copyright
Copyright © Millennium Economics Ltd 2012

1. Introduction

This paper is concerned with explanations for conformity of agents with existing institutional rules in economic life. It draws on, and attempts to add to, contributions from different strands of economics as well as from other disciplines, particularly the new institutionalism in organization studies (which is very different from the new institutional economics). In other words, the main question discussed here is: what leads individual or collective agents – including those in search of pecuniary gain – to follow an existing economic institutional rule? Given this question, the paper focuses on institutions that are not organizations, although many are institutions within or among organizations. As explained below, institutions are broadly understood here as socially shared rules of behavior or of thought.Footnote 1 They include legal norms, which have a formal character, together with conventions and informal social norms, which do not.

Either directly tackling issues of conformity with institutions or not, mainstream economics has provided useful but limited explanations.Footnote 2 Treatments usually examine one of three mechanisms: sanctions, informational differences and some kind of payoff externality (in a different way, also briefly identified by Bernheim, Reference Bernheim1994; Bikhchandani et al., Reference Bikhchandani, Hirshleifer and Welch1998 and, in a more detailed discussion, by Orléan, Reference Orléan, Lesourne, Orléan and Walliser2006). Mainstream economics almost always considers institutions as constraints, incentive mechanisms and/or informational devices, neglecting other, more profound roles that institutions play. In addition, sometimes it adopts a very narrow notion of self-interest and it often overemphasizes the role of self-interest, however narrowly or broadly understood. Furthermore, even its understanding of narrowly self-interested behavior is problematic, in at least two aspects. First, it has not paid enough attention to why people conceive their situations and thus their self-interest in the way they do and to why the agents supposedly share enough aspects of their representation of a situation so as to see it in their interest to conform and to expect others to conform (although a few authors show concern with foundational cognitive issues – e.g. Aoki, Reference Aoki2001, Reference Aoki2011; Gintis, Reference Gintis2009; Greif, Reference Greif2006). Second, mainstream economics also has serious difficulties regarding some types of uncertainty and the introduction of innovations (as distinct from their diffusion), which is often motivated by self-interest. Apart from having other consequences, this affects the explanations for conformity, because the following question is neglected: if agents could deviate by innovating, why do they conform?

Additional explanations for conformity or important insights appear in the writings of some non-mainstream economists and other social scientists. Among the latter, this paper concentrates on the new institutionalists in organizational analysis. Their explanations for the conformity of organizations (as distinct from individuals) seem to correspond to the three mechanisms of isomorphic institutional change in DiMaggio and Powell's typology (Reference DiMaggio and Powell1983): coercive, normative and mimetic. Others have translated these mechanisms as explanations for conformity, and DiMaggio and Powell themselves refer in passing to ‘conformist organizations’ and ‘deviant peers’ (Reference DiMaggio and Powell1983: 154). Alternatively, the organizational institutionalist arguments can be traced back to the related, but somewhat different, distinction established by Scott (Reference Scott2008) between three pillars on which institutions rest: regulative, normative and cultural–cognitive (each emphasized by different approaches in the social sciences).Footnote 3 In their turn, these ideas about conformity outside mainstream economics are certainly helpful, but still need to be further developed and integrated with others.

At the same time, the literature on conformity with institutions is large and varied. Even within the fraction considered here, there is a plethora of concepts and labels. It is not always immediately clear when we are dealing with different arguments or with different forms of the same argument. In addition, valuable ideas that can be translated as implicit explanations appear in works that do not use the word ‘conformity’ or ‘institutions’ (because they focus instead on path dependence, coordination, imitation, diffusion, learning, isomorphism, stability etc.). An effort at classification is required.

For all these reasons, more dialogue and integration are also necessary, both within and across different social science disciplines.

This paper is mainly intended to contribute to this debate by identifying several existing explanations, extending or refining some of them, suggesting new ones, classifying these various explanations and integrating them into an organized framework. This hopefully provides in a consistent way a wider and richer set of possible explanations than one can find in other contributions in each of the separate disciplines and approaches of the social sciences. No claim to exhaustiveness is made here, however, even regarding economics and organizational institutionalism. Outside economics, this and other approaches and disciplines deserve further study.Footnote 4

The objective of the paper is not merely to describe the explanations for conformity that have been proposed or implied and to add others, so as to catalog them. Each of the explanations discussed here is labeled and described in a broad way. In these broad terms, each one of them, often in combination with some of the others, may be useful to explain concrete cases of conformity. Each explanation may, however, appear in different specific forms in the literature, which are not equally plausible or helpful in understanding real-world situations.

While these broad explanations may in several cases be combined, they are presented in such a way that none is always reducible to another or to a combination of the others.

Each of these broad explanations can be applied to economic settings and, more specifically, to markets, including pecuniary behavior. Indeed, a better understanding of markets and this kind of market behavior partly depends on improving and extending arguments regarding institutions. Institutions, including informal ones, are much more widespread and important in markets than economists usually realize, even when it is acknowledged that markets are sets of institutions.

While the focus is on conformity and thus on already existing institutions, the ideas defended here are compatible with, and complementary to, the study of the emergence and change of institutions. In fact, this is in part reflected in the concern with properly leaving room for deviation. Furthermore, at least some of the arguments discussed here can also be used in explaining how institutions emerge through the increase in the number of adopters of a certain rule (although institutionalization and diffusion are not the same). In other words, an increase in adoption may be explained by some of the same reasons that apply when an institutional rule is already established. The same is true of later adoptions, as there is self-reinforcing conformity – a conformity effect, in Young's (Reference Young, Durlauf and Young2001) terms. The focus on conformity results from the need to delimit the scope of the article and does not mean that non-conformity and institutional change are less important or that they have not been discussed in economics or in institutionalist organization studies.

The remainder of the paper is organized as follows. Section 2 contains a short presentation of the concept of institution employed here and the theory of how institutions influence individual behavior and thought. This is part of the foundations underlying the third and main section, which deals with explanations for conformity. There is a considerable difference between following rules consciously and unconsciously. Moreover, for some important approaches to institutions, rules that require much deliberation to be followed may not even be considered institutions at all, so that arguments applying to such rules are not seen as explanations for conformity with institutions. Section 3 thus groups the explanations for conformity into two main sets. The first set consists of variations of the idea that conformity with an institution may be the result of a socially spread habit, instead of a conscious decision to act or think in a particular way. The second set contains a more diverse group of arguments that have in common their applicability to situations in which agents follow institutional rules more consciously. Section 3 also explains that these two sets are not as disconnected as they may initially seem. Concluding remarks follow in Section 4.

2. Institutions: preliminary conceptual and theoretical remarks

Institutions are broadly understood here as socially shared systems of rules of behavior or of thought that have some recurrence (for a similarly broad concept, see Greif, Reference Greif2006: 30). These rules indicate what (not) to do or think in given circumstances. Rules that are not recurrently followed are not institutions (e.g. a law that nobody obeys), but rules are not the same as actual patterns (Lawson, Reference Lawson1997) and could at most be considered as prescribed patterns. They guide actual behavior and thought. In addition, they do not disappear when the actual patterns are momentarily interrupted (Hodgson, Reference Hodgson2006), as long as a tendency remains for several agents to adopt the corresponding actual patterns, and the powers that these rules grant to agents also continue to exist even at moments when they are not actually exerted. Nevertheless, some rules may be only implicit and are inferred from their manifestation in actual patterns. Socially shared means here shared for social reasons, as opposed to purely genetic or other natural causes. The concept of institutions has both a behavioral and a mental dimension. The mental dimension includes not only expectations but also shared mental models [to borrow Denzau and North's (Reference Denzau and North1994) expression]. These models can be of various degrees of specificity or generality.

Institutions depend on the individuals who reproduce, transform or create them, but they also influence individual behavior and thought in crucial, sometimes constitutive ways.Footnote 5 Perhaps least controversial of all is their restrictive role, as constraints. Their cognitive role is threefold: informational, practical and profound. In addition to providing information (as several economists recognize), in particular about the likely behavior of others, they also incorporate practical or tacit knowledge about how to do things, and, especially as shared mental models, perform a profound cognitive function, by influencing the way in which individuals select, organize and interpret information. In terms of motivations, institutions not only provide incentives, as usually emphasized by mainstream economists, but also influence the very objectives that people pursue in their instrumental actions and the obligations that individuals attribute to themselves in their non-instrumental actions. These can be called the simple and the profound variants of their motivational role, respectively.Footnote 6 Finally, institutions may also perform a less studied emotional role, through which they influence the individuals’ emotional state.

The profound cognitive and motivational types of influence of institutions on individuals – which can happen through either habituation or more conscious processesFootnote 7 – lead us away from at least an extreme form of methodological individualism, sometimes called atomism, as individuals are in part explained by institutions, rather than being the sole starting point of social inquiry.

There are different types of institutions. Socially shared norms prescribe what an individual should do or think, bring with them the possibility of external sanctions, but, in the case of at least some individuals, are internalized. Formal or legal norms are ultimately enforced by the organizations of the legal system, while informal social norms are linked to sanctions by other people in the relevant group or community. Conventions have at least two characteristics that other socially shared rules (be they informal or formal) may not have: (1) when followed consciously, a convention is followed at least in part because other people (are expected to) follow it, and not, or not only, because there is an external pressure to comply; (2) a convention is to some degree arbitrary, in the sense that a hypothetical alternative that is not clearly inferior to the prevailing rule is conceivable.Footnote 8 These two characteristics may be respectively called conformity with conformity – or conformity with the (expected) conformity of others – and arbitrariness. Unlike a norm that is not internalized, a convention that is not internalized may be followed without external sanctions, just because of the conformity of others; in contrast, a norm that is internalized can still be followed if others do not follow it, while a convention cannot.

We may now consider a few differences between the concept of institution adopted here and other treatments.

This way of conceptualizing and understanding institutions does not have the precision that many economists associate with mathematical formalization, but it has the advantage of leaving considerable room for deviation, as well as for different relevant explanations of conformity, avoiding some limitations of a game-theoretic definition, the most eminent example of mathematical treatment of institutions. This is an important sense in which the concept is broad.

Some concepts of institutions in economics and in sociology include actual patterns or even restrict institutions to them. For the purpose of merely identifying conformity, this point does not matter much: conformity with institutional rules takes place through the adoption of, or conformity with, the actual pattern dictated by the rule. In theoretical terms, however, it does matter. Rules are not the same as actual patterns, as already pointed out, and the above-mentioned types of influence on individual thought and behavior are exerted by institutional rules, which thus help explain the adoption of actual patterns. Admittedly, it is understandable if authors who wish to benefit from, and combine, insights from different approaches adopt an inclusive concept of institution (e.g. Scott, Reference Scott2008: 48. This was also the spirit in Dequech, Reference Dequech2009). At the present state of the art, with a great disparity among approaches to institutions, this inclusiveness may in some cases facilitate dialogue and avoid antagonizing authors who treat institutions as one or the other, but we must be aware that rules and actual patterns are conceptually and theoretically two different things. More important than the terms used is the distinction between actual patterns and the underlying mechanisms that explain them.

For some proponents of new institutional organizational analysis, not every socially shared prescribed pattern of behavior or of thought is an institution. An additional characteristic deemed necessary is sometimes described as a ‘rulelike’ status (DiMaggio and Powell, Reference DiMaggio, Powell, Powell and DiMaggio1991; Meyer and Rowan, Reference Meyer and Rowan1977; Zucker, Reference Zucker1987). These authors mean something that is perceived as objective, exterior, natural or the like. (The term ‘rulelike’ is not helpful for interdisciplinary communication, since economists often define institutions as rules or systems of rules – in North's Reference North1990 analogy to rules of the game, as well as in more heterodox circles, e.g. Hodgson, Reference Hodgson2006). This requirement is restrictive and, if taken too far, would exclude deviation and endogenous change. However, the difference in relation to the concept adopted here is smaller than it may seem. Organizational institutionalists generally treat acts as being more or less institutionalized, in degrees that change in a process of institutionalization or deinstitutionalization (Zucker, Reference Zucker1977). This implies that socially shared rules may be institutionalized to some degree, even if not a full one – they are institutions, in my sense. In addition, both for me and for the organizational institutionalists (e.g. Tolbert and Zucker, Reference Tolbert, Zucker, Clegg, Hardy and Nord1996: 182), the process of institutionalization includes an increase in the number of adoptions. This is also compatible with the broader definition of institutions by other exponents of that approach (e.g. Scott, Reference Scott2008: 48). In any case, organizational institutionalists who understand institutions in the more strict sense could translate the explanations for conformity with institutions in the broader sense used here as explanations for institutionalization or for the emergence of institutions. Moreover, the approach defended here converges with DiMaggio and Powell's description of sociology as very encompassing ‘with respect to the sorts of things that may be institutionalized’ (Reference DiMaggio, Powell, Powell and DiMaggio1991: 9).

Similarly, for some original institutional economists in the tradition of Veblen, only a specific type of social rule or injunction qualifies as an institution, namely rules that have become habitualized, with habits understood as dispositions or propensities to engage in certain kinds of behavior or thought in given circumstances (Hodgson, Reference Hodgson2006; see also Camic, Reference Camic1986, on habits and Fleetwood, Reference Fleetwood2008b, on institutions, as he excludes even laws and regulations). Here again, one can think of degrees of institutionalization, together with degrees of habituation, depending on how close or far a set of rules is from being followed unconsciously. ‘A rule may be considered, acknowledged, or followed without much thought,’ as Hodgson (Reference Hodgson2006: 3) puts it, although fully conscious deliberation about everything is not viable. Even in the broader sense used here, institutions, by being recurrent, already imply that at least the initial stage of habituation has been achieved. Moreover, as explained below, the explanations for conformity with institutions in this broader sense could also be translated as explanations for institutionalization or for the emergence of institutions in the more restricted sense of habitualized rules.

3. Conformity

An institution is not always or necessarily followed consciously. It is useful therefore to distinguish between two main sets of explanations for conformity.

The first set consists of variations of the argument that conformity with an institution may be the result of a socially spread habit, instead of a conscious decision to act or think in a particular way.

The identification of social habits as the foundations of institutions – or sometimes as institutions themselves – is a major feature of the original institutional economics (e.g. Hodgson, Reference Hodgson2007). It also appears in the Austrian wing of the new institutional economics (e.g. Langlois, Reference Langlois and Dennis1998), at least in part under the influence of its precursor (Hayek, Reference Hayek1973). In contrast, mainstream economists have not paid enough attention to habits when discussing conformity or more generally. At least North (Reference North1990, on informal constraints), Denzau and North (Reference Denzau and North1994) and Greif (Reference Greif2006) have opened more room for the lack of conscious thinking and for habits in new institutional economic analysis. Closer to their older, Veblenian economic counterpart, the new institutionalism in organization studies has focused on ‘pre-conscious processes and schema, as they enter into routine, taken-for-granted behavior’ (DiMaggio and Powell, Reference DiMaggio, Powell, Powell and DiMaggio1991: 22; the Veblen connection is noted by Colyvas and Powell, Reference Colyvas and Powell2006). The cognitive–cultural pillar of institutions, highlighted by this approach, ‘rests on preconscious, taken-for-granted understandings’ (Scott, Reference Scott2008: 60; also p. 50). This does not correspond exactly to the mimetic part of DiMaggio and Powell's typology, since for them both the mimetic and the normative types of isomorphism ‘involve managerial behaviors at the level of taken-for-granted assumptions rather than consciously strategic choices’ (Reference DiMaggio and Powell1983: 149, n. 5). Influenced by Berger and Luckmann (Reference Berger and Luckmann1966), some proponents of this approach even see ‘habitualization’ as a crucial part of the process of institutionalization (Tolbert and Zucker, Reference Tolbert, Zucker, Clegg, Hardy and Nord1996).

The process of habit formation occurs through repetition of behavior and thought. In turn, this may stem from unconscious imitation or from the repetition of conscious behavior and thought (including deliberate imitation).

The latter possibility already implies that, at least for some time, many rules of action or of thought are followed consciously. Several of these rules are institutions. This does not necessarily mean that the person or group following an institutional rule is aware that it is an institution. In other words, one may be conscious of what one is doing and of similar behavior or thought by others, without being aware that their joint rules of behavior or thought constitute an institution. The situations in which an institutional rule is followed consciously, with or without awareness of its institutional character, are covered by the second main set of explanations for conformity.

These two sets are less separate than they may appear, for three reasons. First, the corresponding distinction between unconscious and conscious is just a simplifying approximation to the actually more complex coexistence of several levels of consciousness. Second, at any point in time, doing some things less consciously frees up mental resources for doing other things more consciously. In this sense, habitual conformity with some institutions allows a more conscious conformity with, or deviation from, other institutions. Third, when considered dynamically, these two sets may partly intersect: as habits originate from repetition, repeated conscious conformity with an institution and repeated unconscious imitation of conformers may become a social habit; explanations for conscious conformity thus become explanations for the development of habits and for future unconscious conformity. These three arguments also imply that actual thought and behavior are a mixture of more conscious and less conscious aspects. They are probably never fully unconscious (at least when people are awake) or fully conscious.

The explanations for conscious conformity are quite diverse. Some of them seem to have been neglected by most economists; others have been neglected by other social scientists. Some arguments have been considered by the organizational institutionalists when dealing with habitualized acts and ideas, but the present paper suggests – and this is part of its intended contributions – that they should also be more clearly and strongly applied to conscious behavior and thought, before it becomes habitualized or when there is a period of dehabituation (or deinstitutionalization), which may be reverted into rehabituation. Such a suggestion is congruent with the new organizational institutionalists’ concern, after the early formulations, with agency, intentionality, contestation and justification. This does not imply that conscious behavior is always instrumental or rational.

In addition to not being so separate from the first set of explanations, the second one is not necessarily more relevant, but it requires a lengthier discussion, to which I now turn. All the following subsections refer to this second set.

Coordination (effect) and increasing returns to adoption

One possible explanation for conformity involves the idea that in several instances it is in an agent's interest to coordinate with others.Footnote 9

In some cases, this happens in such a way that one benefits from doing the same as others, but the benefit does not increase with the number of other people adopting the same option, at least after some level has been reached, or the benefit is conditional on everybody doing the same thing. A formal illustration is a simple game of pure coordination with more than one equilibrium.

In other cases, there are increasing returns to adoption, although this is not always translated in such terms. The notion of increasing returns to adoption was first introduced by Arthur (Reference Arthur1989), who applied it to technologies. Related notions include network externality or effect and strategic complementarity (as well as the broader concepts of self-reinforcing mechanism and positive feedback), but a discussion of their conceptual relation will have to be made elsewhere. Returns may increase with the number of adoptions until they more than compensate an intrinsic preference for another alternative, thus explaining a choice at a given moment in time. In turn, this may lead to new adoptions and so on. The same dynamics may involve expected adoptions (when returns also depend on them), leading to self-fulfilling expectations. If an individual expects many others to adopt a given option, he or she will also adopt it; if a sufficient number of individuals expects the same the resulting collective opinion becomes self-fulfilling.

The explanation for conformity based on coordination and increasing returns to adoption has been incorporated into mainstream economics. We should, however, broaden its scope. Let us consider two ways in which this can be done: by widening the notion of increasing returns to adoption; and by enlarging the notion of self-interest associated with increasing returns to adoption.

The object of choice involving increasing returns to adoption does not have to be a physical one or a technology. Others have already extended the notion of increasing returns to adoption to deal with the adoption of institutional forms such as laws and constitutions. For North (Reference North1990: 94–95), Arthur's ideas about self-reinforcing mechanisms in the adoption of technologies can be applied to the study of institutional change. It can be suggested more explicitly that the object of choice may be an informal behavioral rule or a set of ideas, like an opinion, an expectational system of rules, or a more general model. Then we can relate increasing returns to adoption to informal institutions as rules not only of behavior but also of thought.

In the latter case, the connection between increasing returns to adoption and self-fulfilling expectations becomes more direct, as the object of adoption itself can be an expectation or a model used to generate expectations. In situations where the average opinion is expected to be a self-fulfilling expectation, the more an opinion is expected to gain supporters, the more it tends to really gain supporters and thus to become a representative average opinion and a convention. Keynes's comparison of the stock exchange to a newspaper beauty contest provides a famous example of such a convention (Reference Keynes1936: 156). Several individuals may adopt what they expect to be the average opinion while finding it unfounded from an isolated individual viewpoint. Each may require a different threshold of adoption by others to abandon their personal views.

The concept of increasing returns to adoption can be usefully understood in a very general sense, as referring to the case in which the benefit that an agent gets from choosing an option depends positively on the number of other agents choosing the same option. In this sense, the concept is not restricted to any specific type of object of choice. The more one generalizes the idea of increasing returns to adoption beyond its common use in economics, by including a varied array of objects of choice, the more one opens room for alternative sources of increasing returns to adoption and the more important it becomes to identify these sources in each specific case. This identification will render the argument more precise, in addition to facilitating communication, since to many economists the term ‘increasing returns to adoption’ calls forth ideas about what they already consider as objects of choice (such as goods and technologies) and sources of these returns. Coordination effects are only one such source.

Another relevant issue here is the notion of self-interest. The concept of increasing returns to adoption has (even if implicitly) been associated with the traditional economic notion of self-interest as strict selfishness. This does not, however, need to always be the case, especially if one is not restrictive about the object of choice involving increasing returns to adoption. An alternative notion of self-interest allows one's well-being to depend on the well-being of others or on the common good of the participants of a collective entity (an organization, a group etc.). The individual may adhere to a collective rule because he or she considers this to be good for the group in general;Footnote 10 if the individual expects the benefit for everyone involved to increase with the number of other people who conform (at least after a minimum number has been reached), then there are increasing returns to adoption.

Social sanctions

Formal and informal social norms imply the possibility of negative or positive sanctions: some sort of punishment against dissenters or reward for those who comply. In some cases, a certain institutional rule is consciously followed by some individuals in order to avoid negative social sanctions or to receive positive ones. In such cases the institution is (at least in part) followed out of egoistic self-interest, if self-interest is defined in a broad enough sense to incorporate the negative consequences of disobeying a given social norm and the positive consequences of following it (the focus here is on sanctions that are external to each individual;Footnote 11 the possibility of internal sanctions is considered below, within a broader discussion of internalization).

In approaches centered on utility, such as neoclassical economics, the meaning of utility may also be broadened by including these consequences among the arguments of the utility function (e.g. Akerlof, Reference Akerlof1980; Becker, Reference Becker1996; Bernheim, Reference Bernheim1994; from a public choice perspective, Kuran, Reference Kuran1987a).

In organizational institutionalism, social sanctions correspond, in one possible reading, to the coercive mechanism in DiMaggio and Powell's (Reference DiMaggio and Powell1983) typology and to what Scott (Reference Scott2008) calls the regulative pillar. These authors refer to both informal and formal pressures (DiMaggio and Powell, Reference DiMaggio and Powell1983: 150; Scott, Reference Scott2008: 52), although there are passages in their works that may have led some interpreters to associate the coercive mechanism and the regulatory pillar only with legal sanctions, perhaps with informal sanctions relating to the normative mechanism/pillar.Footnote 12

Scott (Reference Scott2008: 51–52) and Powell and Colyvas (Reference Powell, Colyvas, Clegg and Bailey2008) rightly point out ‘expedience’ as the basis of compliance in this case (but we should add that expedience is also important in other plausible explanations of conscious conformity).

Here again, in the case of social sanctions, we may broaden the scope of this explanation for conformity. Both economists and sociologists seem to usually think of values as having a moral or political character. By considering the possible existence and importance of other types of values (as illustrated below in the discussion of legitimacy), one allows different types of norms to exist, together with a greater number of instances of possible sanctions.

Those who conform to a norm only because of external sanctions do not find the norm legitimate and have not internalized it. In contrast, those who impose informal sanctions have internalized it or, in the case of at least a few people, may enforce it to signal to others that they find the norm legitimate, when in fact their own conformity is due to external pressure. The latter case is sometimes called ‘false enforcement’ in the literature. If at least some individuals have to accept the legitimacy of a norm before they impose sanctions on others, the explanation based on external sanctions is not completely independent from that based on legitimacy, but the individuals to which these explanations apply are not the same.

It is possible to have social sanctions combined with a special case of increasing returns to adoption, when the value of the rewards for conformity (or the value of the punishment for nonconformity) increases with the number of people who conform.Footnote 13

When they are possible, external sanctions may explain the conformity of some, but not all, agents. External or even internal sanctions may be unnecessary.Footnote 14

Informational differences

Sometimes an agent consciously imitates others because of a belief that they may be better informed. The private beliefs that the agent would otherwise hold may be abandoned in favor of what is deemed to be the opinion of others.

Keynes (Reference Keynes1937) used this idea to justify the conformity of different individuals to the average or majority opinion, as part of what he called a ‘practical theory of the future’ in situations of uncertainty not covered by neoclassical economic theory. Several post-Keynesian economists have approvingly referred to Keynes on this, while the conventionalist André Orléan has studied this phenomenon of ‘informational mimetism’ in more detail (e.g. Orléan, Reference Orléan, Lesourne, Orléan and Walliser2006).

Within mainstream economics, there have been models of information cascades and herd behavior due to informational differences (for an overview, see Bikhchandani et al., Reference Bikhchandani, Hirshleifer and Welch1998). At least two subsets of this literature are noteworthy.

In the first, herding and cascades can be based on little information; although several authors refer to learning, the behavior of others may quickly become uninformative, and the blind lead the blind.Footnote 15 Like in Keynes's theory, the resulting convention has a precarious stability, as a small piece of news may break it. On the other hand, while many economists in this literature see imitation due to a belief in informational differences as a short-term phenomenon, an anomaly that tends to be corrected, an alternative approach could see it as a more recurrent factor in the explanation of conformity. Fundamental uncertainty is ever present in some decision environments, even if the news and the specific contents of the convention being followed change.

In greater contrast with such an alternative view, a second subset of this literature has allowed imitation to be based on much more knowledge. Some models even assume evidence that the behavior adopted has previously yielded good or desirable results. In the latter case, there surely is imitation, but it is not clear why this case of social learning should be described, from the viewpoint of these economists, as ‘informational conformity’ (Young, Reference Young, Durlauf and Young2001, emphasis added).Footnote 16

Uncertainty perception and uncertainty aversion

Conformity may also be explained by uncertainty.

In some cases, an agent may conceive an alternative to an existing institution and may expect this alternative to yield a high payoff, but still refrains from implementing it, because of the fundamental uncertainty involved in path creation, to borrow Garud and Karnøe's (Reference Garud, Karnøe, Garud and Karnøe2001) expression. The degree of uncertainty about the results of the unconventional action may be perceived as too high and/or the strength of the agent's disposition to face this uncertainty (the courage to deviate despite it) may not be high enough, so that the agent is unwilling to act on the basis of the expectations about the payoff of the unconventional alternative. In Keynesian terms, expectations may be very positive, but, at the same time, confidence may be too low (Dequech, Reference Dequech1999).

In this case, uncertainty as perceived and disliked by the agent only constitutes a separate explanation for conformity if the agent perceives more uncertainty in deviation than in conformity. This may not be true of uncertainty about absolute payoffs. On the other hand, at least conformity may indeed involve less uncertainty about one's relative position in comparison with others who also supposedly conform.

Here again one may speak of conformity out of self-interest, which may or may not be egoistic. As conformity may occur even without social sanctions, this is self-interest in more a traditional sense.

In contrast, the type of uncertainty involved in some kinds of unconventional action – such as the introduction of innovations – is not the one traditionally considered by most economists who refer to uncertainty aversion or uncertainty discounting. The act itself of introducing an innovation – instead of imitating one that has already been introduced – seems to have so far been barred from the formal apparatus of contemporary mainstream economics (e.g. game theory and axiomatic utility maximization theory).

Other authors who also do not have a traditional economic notion of uncertainty have suggested a different variety of conformity explanation based on uncertainty aversion. They do not refer to the case in which a specific unconventional alternative is conceived, considered and then discarded. This version of the argument that conformity is a safe or defensive option seems to implicitly apply to a previous stage of the decision-making process, when some agents decide against even searching for a new alternative or against continuing this search, if they started it at all (see Dequech, Reference Dequech2003, for some references from the post-Keynesian and Austrian economic literature).

The same applies to DiMaggio and Powell (Reference DiMaggio and Powell1983: 151), with some differences. Their analysis of an unconscious mimetic mechanism in principle fits better in the first broad set of explanations, as mentioned above; on the other hand, DiMaggio (Reference DiMaggio and Zucker1988) is more suggestive of deliberation.Footnote 17 Moreover, in their account, organizations tend to be modeled after those that their decision-makers perceive as more legitimate or successful (p. 153). One should note that this may happen before the appearance of an institutional rule: if the proportion of adopters is not sufficiently large, the pattern being imitated is not an institutional one. It may, however, become so, if imitation continues to spread – as assumed by DiMaggio and Powell (Reference DiMaggio and Powell1983), who see this mimetic process as a source of isomorphism among organizations.

Also related to uncertainty perception is a different proposition, according to which conformity is a practical response to the problem of complexity (in relation to the agents’ capabilities). What seems to matter is not aversion to this procedural type of uncertainty, but rather the cost and the trouble of searching for alternatives (again, see Dequech, Reference Dequech2003, for some references); the argument can be extended to search under fundamental uncertainty, which is also costly. Perhaps this should be treated as a separate explanation, but only if something different from the explanations considered here accounts for the choice of conformity with an institutional rule over other rules of thumb.

Legitimacy

Another explanation for conformity with an institution lies in its legitimacy, which may make it more acceptable than a deviant alternative.

In broad terms, legitimacy can be understood as a socially accepted compatibility with some values. The legitimate is that which is considered right, appropriate or the like, in accordance with those specific values. Dynamically, the legitimate is that which either remains unquestioned or survives legitimacy tests, i.e. that which, after some doubts have been raised or criticisms have been made, is still considered right, appropriate etc.Footnote 18

Legitimacy may be of different types, just like the values on which it is based. In the discussion of institutions, at least three types of legitimacy are especially interesting (Dequech, Reference Dequech2009). An institutional behavioral rule or an institutional mental model may have moral legitimacy. Moral values include honesty, justice, goodness and the like. Also possible is political legitimacy (which has been usually associated with formal institutions). Examples of political values, sometimes overlapping with moral ones, include civism, democracy, freedom, equality, social justice, representativeness etc. Moral and political legitimacy are probably the most commonly emphasized types of legitimacy in the social science literature.Footnote 19 Other types are also important, however. For example, an institutional set of rules of thought may be the dominant, most prestigious model among a certain group and possess epistemic legitimacy, i.e. a socially recognized appearance of well-foundedness.Footnote 20 Examples of epistemic values are compatibility with the empirical evidence, internal consistency, rigor and relevance.Footnote 21 Epistemic values are cultivated in academic life, but they are also held by agents outside academia, regarding both their own mental models and ideas about the world and those of other people.

The epistemic legitimacy mechanism is different from informational mimetism: it is broader in that it refers not simply to information, but to a cognitive structure that can be used to select, organize and interpret information and to ideas that result from this mental model.

Sometimes, there is belief distortion and self-deception, which epistemic legitimacy helps explain. In several other cases of epistemic legitimacy, however, these problems are not clearly present.Footnote 22

By not restricting legitimacy to its moral and political varieties, the scope of this explanation increases greatly. In particular, accepting ideas because they are seen as right in the sense of having the equivalent of epistemic legitimacy also seems very frequent and relevant in everyday life. We can refer to epistemic norms that can be internalized (or followed because of sanctions, so that the scope of the sanctions argument is also expanded, as mentioned aboveFootnote 23).

In turn, the adoption of a set of ideas because of their epistemic legitimacy helps explain many institutional behavioral practices that are adopted in part because of these ideas and in part because of other factors (such as the objectives pursued and the perception of constraints, when behavior is instrumental).

The legitimacy of an institution may be seen as a factor leading several individuals to internalize the institution as a norm, thus complying with it regardless of the possible external sanctions. Internalization arises from a sense of what is right – morally, politically, epistemically etc.

Legitimacy as a factor of conformity may be interpreted in instrumental or in non-instrumental terms.

The non-instrumental interpretation of legitimacy applies to the situations in which an agent follows the institutional rule out of duty, obligation, appropriateness or (perhaps especially in the case of the epistemic legitimacy of ideas) simply because this is seen as right. In these situations, internalization is not linked to a concern with consequences. In this specific sense, it is related to Scott's normative pillar, which involves a logic of appropriateness and social obligation;Footnote 24 on the other hand, his description refers to ‘morally governed’ behavior (Scott, Reference Scott2008: 51), while we may also include other types of underlying values and respective legitimacies.Footnote 25

In turn, the instrumental interpretation of legitimacy can be built in two different ways. The first also involves internalization, and again there is a sense of what is right, but now internalization is associated with objectives (as distinct from obligations) and internal sanctions. The agent is seen as complying with an existing institutional rule in order to avoid internal negative sanctions (such as a feeling of shame that would arise in the case of deviance) or to obtain internal rewards (such as a feeling of pride). Illustrating this interpretation, neoclassical economists and like-minded rational-choice scholars have a way of relating legitimacy with utility maximization, via internalization. The utility function may reflect the negative and positive internal sanctions, possibly in addition to the external sanctions mentioned above (e.g. Akerlof, Reference Akerlof1980; Akerlof and Kranton, Reference Akerlof and Kranton2000).Footnote 26 It should be noted, however, that the utility maximization hypothesis is not always linked to an assumption that behavior is instrumental or that people are aware of their motivations.

There is a second way of interpreting legitimacy in instrumental terms and seeing legitimacy as helping to make it in the individual's interest to conform to an institution. According to the economics of conventions, an interdisciplinary approach originated in France, the successful coordination of different individuals requires that they be concerned with the collective entities of which they are part and that they share a representation associated with a satisfactory working of the relation between them (Bessy and Favereau, Reference Bessy and Favereau2003; Favereau et al., Reference Favereau, Biencourt, Eymard-Duvernay, Lazega and Favereau2002; Storper and Salais, Reference Storper and Salais1997). It is on the basis of this social representation that conventional rules of behavior are interpreted and may be deemed legitimate. This approach thus can be seen as combining the coordination and the legitimacy arguments. The issue of accountability is part of this combination, since the conventionalists maintain that coordination requires that actions be seen as ‘justifiable’ (e.g. Thévenot, Reference Thévenot and Fullbrook2002: 183). Biggart and Beamish (Reference Biggart and Beamish2003: 444) briefly link conventions, accountability and coordination in a broadly similar way.

This legitimacy-cum-coordination argument is different from the one based on sanctions, in at least two respects: first, the conforming individual shares the perception of legitimacy, instead of conforming just because of sanctions;Footnote 27 second, a non-egoistic type of self-interest is involved.

The legitimacy explanation – either interpreted in instrumental terms or (in connection with the logic of appropriateness) in non-instrumental ones – can be said to subsume the argument that conformity is due to a sense of identity. For some authors, this might include the special case of status-seeking conformity with significant others, but the latter is more clearly related to the sanctions explanation. Identity may be the actual factor behind the veil of what some mainstream economists consider as another possibility: conformity for its own sake, apparently labeled as conformity preference or preference interaction.Footnote 28

Dynamically, the legitimacy explanation may partly reflect other explanations. Some individuals may have perceived the institution as legitimate since the beginning of their conformity, but others may have found it illegitimate at first and changed their minds over time. For example, the perception of legitimacy may vary with the number of other individuals who conform (perhaps together with those who, by falsely enforcing a norm, appear to conform). Some scholars in economics have argued that sanctions may also cause a change in individuals, with ‘adaptive preferences’ instead of fixed ones.Footnote 29

Naturality or inevitability

Another possible explanation for conformity is that the adoption of a certain pattern of behavior or of thought by many people may contribute to making it seem natural, inevitable, self-evident and the like. (One could also use the term ‘taken for granted’ in a similar sense, but this is not the only way the term has been used.Footnote 30) In this case, rather than being considered less legitimate than the prevailing institution, an alternative is not considered at all. People who find the institutional rule natural or inevitable are certainly not aware of its institutional character.

Neither legitimacy nor naturality stimulates the search for alternatives. In addition, the widespread and continuous use of an institution may lead to a lack of contact with other ideas and practices, increasing the chances of the institution being seen as legitimate or inevitable. The lack of variety and the institution's legitimacy or naturality tend to mutually reinforce one another (see also Tolbert and Zucker, Reference Tolbert, Zucker, Clegg, Hardy and Nord1996: 181, on the enhancement of exteriority by new adoptions).

In a sense, naturality or inevitability could be regarded not as separate from legitimacy, but as an extreme form of legitimacy. This would bring us closer to institutional organizational theory and to the economics of conventions. Although there seems to be no consensus about the concept, some organizational institutionalists treat legitimacy as coming in degrees and involving, at least when its degree is high enough, an absence of questioning. The same applies even to taken-for-grantedness (Aldrich and Fiol, Reference Aldrich and Fiol1994; Tolbert and Zucker, Reference Tolbert, Zucker, Clegg, Hardy and Nord1996), which Suchman (Reference Suchman1995) considers as one among several types of legitimacy.Footnote 31 Some conventionalists have embraced Douglas's (Reference Douglas1986) argument that legitimation occurs through naturalization (e.g. Orléan, Reference Orléan and Orléan2004), very much like Johnson et al. (Reference Johnson, Dowd and Ridgeway2006: 56). At the same time, the conventionalist approach has emphasized the need for legitimate conventions to pass recurrent legitimacy tests, during which they are confronted with alternatives.

Regardless of the exact terms used, conceptually and also theoretically, there is a difference between considering an alternative less legitimate than another and not considering another alternative at all. This may be relevant even for the very understanding of the process by which an institution becomes taken for granted. For example, having more legitimacy than other alternatives may at least in some cases lead to the abandonment of these other alternatives and therefore may be a step toward becoming the only alternative considered and being seen as inevitable.Footnote 32 In turn, if a certain rule is seen as inevitable, this may make further reflection unnecessary and habituation easier. After habituation, we would fall in the first set of explanations. From time to time, however, someone may be questioned and then consciously reply that there is no alternative.

Whatever their precise conceptual relation, legitimacy and naturality combined help to account for the fact that people many times do not conceive of a sufficiently appealing alternative to an existing institution (as pointed out regarding conventions in Dequech, Reference Dequech2003). This is often neglected by economists.

A suggestion may be added to broaden the scope of the idea of inevitability: in some cases, a limited set of alternatives, instead of a single option, may be perceived as inevitable, exhausting all the possibilities. This may help explain why possible alternatives outside that set (or ‘outside the box’) are not considered and thus why people adopt a particular alternative within the set. This idea may then be used to reinterpret economic models that assume that agents take the set of alternatives or strategies as given. These models could be seen as implicitly assuming an extreme degree of institutionalization.

Either regarding a single alternative or a larger set, the perception of inevitability requires that individuals neglect options whose consideration would involve the ability to imagine a future that is, in some relevant aspect, structurally different from the present. In Dequech (Reference Dequech1999), this ability is called creativity, while its lack is identified as one among other factors that may lead to conventional behavior.Footnote 33 When creativity is weak, it is easier to perceive a set of alternatives or a single existing institution as inevitable – or as legitimate, for that matter. Creativity itself is not strictly individual and reflects the stimuli and influences of the institutional environment (e.g. Stark, Reference Stark2009).

Insufficient power or resources to deviate or to implement institutional change

There often exist situations in which someone would prefer to behave differently from others or would prefer that everybody in the group behaved differently, but ends up doing things in the usual way. In this case, an alternative is not only conceived but also preferred (even after discounting for uncertainty), although not actually implemented. The reason is a lack of practical means to pursue the alternative route.

For example, within an organization, an individual or group of individuals who has devised an innovation and is willing to implement it very often needs to convince other members of the same organization that this is a desirable and viable course of action. Convincing others may be difficult. The proposed innovative action may be seen by others as contrary to their private interests, as violating social norms, as insufficiently profitable, as involving too much uncertainty and/or as illegitimate. These other members of the organization may prevent the innovation from being introduced either by being clearly opposed to it (even without criticizing or punishing its proponents), by holding without sufficient confidence an expectation that the innovation may succeed, or by not reacting at all. As Garud and Karnøe (Reference Garud, Karnøe, Garud and Karnøe2001: 14) put it: ‘Most deviations are met with apathy at best and resistance at worst’.

Even if internal obstacles are overcome, factors operating outside the organization may still prevent the innovation from taking place. If external funds are necessary and a loan application or a partnership offer is made to a bank, for instance, the bank's decision-makers may not agree to finance the innovative project. Even if the necessary funds are or become available, some additional external agreement may not be granted: for instance, a government license to sell a new product, a patent application approval etc.

Other cases of blocked deviation may not involve the attempt to introduce an innovation. They may reflect a conflict of interests and ideas regarding the existing situation. For example, some individuals in a group may question the legitimacy of the prevailing institutional rule. The recognition of, say, moral legitimacy is not indisputable. There are different moral values, as mentioned above, and each may be seen from different viewpoints. The same applies to other forms of legitimacy. If the dissenting group publicly criticizes the status quo and defends a different alternative, it may fail to persuade others, including more powerful members. If not hopeful or courageous enough, the dissatisfied individuals may not even voice their discontentment anymore. In the latter case, deviation becomes restricted to thought.

An even more blatant lack of practical means to deviate may exist when the intended institutional change would involve several organizations at the same time or other large parts of society.

In terms of the traditional economic jargon, one could say that neither the set of alternative options nor the set of constraints is necessarily conceived in the same way by every agent. The potential deviant agent(s), in particular, may imagine a different feasible set. In practice, however, this imagined feasible set may turn out not to be really feasible, and the deviant action may not be implemented for lack of the necessary means. Analysts should not always take the feasible set as given, as agents themselves often do not. The definition of the feasible set is not static: rather, it is the object of negotiation and possibly of conflict of interests and ideas; and it may reflect creativity.Footnote 34

4. Concluding remarks

This paper is an attempt to contribute to a broader and better understanding of conformity with institutions, combining insights from different disciplines and suggesting some new ideas and a useful way of organizing the discussion. It tries to show that the set of possible explanations for conformity with institutions is wider and richer than is implied by other treatments of institutions in the literature, especially by the game-theoretic and utility-maximization approaches that have been developed in mainstream economics.

Contrary to what many economists do, we should not neglect the role of habits and subconscious processes, which are manifested in many instances of conformity with institutions. Acknowledging this does not imply treating economic agents as slaves of habits, certainly not all agents and not all the time. Habits often result from the repetition of conscious action and thought. Moreover, habits may indeed be a conservative factor, but they also liberate attention that may be used in conscious decisions to deviate (or to conform).

On the other hand, people do often follow some institutions consciously, in many cases without being aware of the institutional character of their behavioral and mental rules.

It is regarding conscious behavior in accordance with existing institutions that mainstream economists have made their most distinctive contributions to the study of conformity. In particular, they have discussed coordination effects and, even when not concerned with institutions, have introduced the notion of increasing returns to adoption and the like. These ideas have been useful to understand behavior that may be conventional in relation not only to consumption goods and some technologies but also to institutional forms. It has been argued here that the notion of increasing returns to adoption may be broadened, especially by not being restrictive about the object of choice. In turn, this allows for the existence of more situations and sources of increasing returns to adoption than many economists realize. At the same time, it becomes necessary for us to be more explicit about these sources and analyze them more carefully. Indeed, in its most general sense, the notion of increasing returns to adoption goes only so far in explaining conformity. It indicates that conformity may be in someone's interest if others also conform, but does not specify why.

The presence of external sanctions is another possible reason for conformity with an institution, which in this case is a formal or informal social norm. This explanation has also had its scope of possible application broadened here, by the emphasis on epistemic and possibly other types of values in addition to moral and political ones.

In these two explanations, an individual guided by self-interest conceives an alternative to the institution, but prefers to conform. The same may happen when uncertainty perception and/or uncertainty aversion is the reason for conformity, if a search is undertaken and does generate one or more alternatives. Uncertainty may, however, even discourage the search for alternatives. With or without this search, at least in some cases this explanation leads us away from mainstream economics. Most notably, this occurs when the deviation from which one refrains consists in the introduction of an innovation (or even the search for this new alternative), as this involves a fundamental kind of uncertainty with which mainstream economics has had an overwhelming difficulty to deal.

Similarly, if conformity is explained by informational mimetism, then sometimes the explanation also involves fundamental uncertainty. If so, this kind of imitation may be more recurrent than usually acknowledged, instead of a short-term anomaly.

Another reason for conformity with an institution is its legitimacy. In this case, one or more alternatives are perceived as less legitimate than the prevailing rule. This is another explanation whose scope should be widened, with the variety of values underlying legitimacy and particularly with the introduction of the notions of epistemic values and epistemic legitimacy. This reinforces the view of the process of cognition as profoundly institutional in nature. It also reinforces the possibility of using the notion of legitimacy to identify several cases of non-instrumental conformity – with prevailing ideas. In turn, ideas that are accepted because of their epistemic legitimacy may guide instrumental conformity with prevailing rules of behavior.

While sometimes an alternative to conformity may be considered less preferred and/or less legitimate, it also often happens that no alternative is considered at all, as the institutional way is seen as natural or inevitable.Footnote 35 This argument has been broadened here to a set of alternatives that is seen as inevitable, a box outside of which people do not think.

The comprehension of legitimacy and naturality as reasons for conformity is much improved once one recognizes the profound motivational and cognitive roles that institutions perform and how they also influence the conception of legitimate ways of pursuing goals and fulfilling obligations, all of which leads us beyond most of mainstream economics.

The whole line of reasoning based on the ideas of legitimacy and naturality or inevitability converges with contributions from the new institutional organizational analysis. On the other hand, as Hirsch and Lounsbury (Reference Hirsch and Lounsbury1997: 415) point out, cognitive arguments may be just as capable as moral ones of falling into the mistake of treating individuals as ‘cultural dopes’. We must not treat all individuals like that, all the time. While the new organizational institutionalists have tended to associate legitimacy and naturality – as well as uncertainty leading to mimetism – with preconscious understandings, it has been argued here that they should also be considered as important factors in the explanation of conscious conformity (including pecuniary behavior). This reinforces the proposition that ‘taken-for-grantedness is the outcome of purposive action, the refining of skills, and the development of reflexivity. (. . .) The institutionalization of principles and practices initially requires the mindful engagement of individuals in organizations’ (Colyvas and Powell, Reference Colyvas and Powell2006: 308, 310). Moreover, responses to institutions may vary for different individuals and organizations (as also noted by Lawrence and Suddaby, Reference Lawrence, Suddaby and Clegg2006: 238). Room has been allowed here not only to more conscious thought and action, but also to a larger perception of alternatives – some of which may be discarded because of uncertainty, while others may still be considered appealing – and to attempts to deviate.

This last point leads us to a different type of explanation for conformity, one that is not usually considered as such: the lack of sufficient resources or power to implement a preferred alternative, which could even be an alternative involving fundamental uncertainty, like the introduction of an institutional or technological innovation. This is an explanation of conformity only when the potential deviant, unable to make the preferred alternative viable, ends up doing the same as the others. Of utmost importance here is the fact that many significant economic decisions are made within organizations, by individuals who are not autonomous to decide as they please. In addition, even when an agreement is reached within an organization, implementing the agreed-upon plan may be prevented by people in other organizations. In any case, exploring this explanation involves trying to understand why other people are insufficiently confident, simply apathetic or openly opposed in relation to the proposed alternative.

Behind this and other explanations is power as a factor inducing conformity. The lack of power by some agents to implement an alternative is the other face of the power that other agents have to prevent changes. Similarly, the possibility of sanctions leads some agents to conform because other agents have the power to impose these sanctions, enforcing the rules. Institutions are therefore in part maintained by power relations. At the same time, especially through their presumed legitimacy or inevitability, institutions help maintain power relations. As Dobbin (2004: 6) notes, some authors see power itself as the ability to influence how people understand the world and their interests. From this point of view, the perceived legitimacy or inevitability of institutions and power mutually reinforce one another.

Having highlighted the profound cognitive and motivational influence of institutions on individuals, we can not only extend the set of explanations in comparison with those that have been developed by mainstream economists, but also deepen all kinds of different explanations in which people conform to institutions out of self-interest. Institutions, especially as shared mental models, are an integral part of the answer to a question that economists do not ask frequently enough, if at all: why do individuals see their situation as they do, including the perception of what is in their interest?

On the other hand, recognizing the influence of institutions and the factors that explain many instances of conformity must not prevent us from adopting a concept and a theory of institutions that allow room for deviation, especially without implying that deviating individuals are somehow deficient or always subject to sanctions. This is hopefully the case here, but developing this argument is another story.

Footnotes

1 Institutions help provide order and stability to social life. Discussing explanations for conformity in turn is crucial to understand why institutions themselves continue to exist, given that institutions are reproduced by conforming agents. In addition, explanations for conformity allow us to identify more precisely the specific ways in which rules and the associated actual patterns are socially shared. In turn, some of these ways generate more stability in a given institution than others. The study of conformity is also useful for us to better understand the different forms of rationality or reasonableness of economic behavior and their limits. From a more dynamic perspective, explanations for conformity illuminate the process by which the degree of institutionalization and the ensuing stability increase, while properly leaving room for deviation is essential for an adequate understanding of how institutions change, as institutions are transformed (at least in degree, but sometimes in kind) or destroyed by non-conforming agents.

2 As defined by Dequech (Reference Dequech2007: 281), ‘mainstream economics is that which is taught in the most prestigious universities and colleges, gets published in the most prestigious journals, receives funds from the most important research foundations, and wins the most prestigious awards.’

3 Scott (Reference Scott2008: 51, 133) himself associates each of his pillars with one of DiMaggio and Powell's mechanisms, but in some respects the correspondence is less than exact, as indicated below.

4 Regarding sociology, in particular, the literature explicitly mentioned here is much less diverse than in the case of economics, because of the focus on organizational institutionalism; the paper does not cite authors such as Weber, Durkheim, Parsons, Bourdieu, Giddens, and many others, but at least some of them appear here indirectly, through their influence on organizational institutionalism (see, for example, DiMaggio and Powell, Reference DiMaggio, Powell, Powell and DiMaggio1991, and Scott, Reference Scott2008) and on some branches of economics (such as the economics of conventions, in the case of Weber and Durkheim).

5 This paragraph extends and refines a similar discussion in Dequech (Reference Dequech2006).

6 Greif (Reference Greif2006: 8) refers to motivation in a different sense, by equating it with ‘incentives broadly defined to include expectations, beliefs, and internalized norms’.

7 See also Fleetwood (Reference Fleetwood2008a: 189–193) on intentions.

8 In Dequech (Reference Dequech2009), it is only implicit that the alternative must not be clearly inferior to the prevailing rule for the latter to be arbitrary.

9 Young (Reference Young, Durlauf and Young2001) labels this case ‘instrumental conformity’, but several other explanations discussed below assume instrumental behavior. See also the comments on ‘expedience.’

10 According to the economics of conventions, this kind of consideration occurs in situations where individuals need to coordinate with each other and come to think that a successful coordination involves a shared image of the common good of the group. See the discussion on ‘legitimacy’ below.

11 These external sanctions may be endogenous to a model that explains them, together with the corresponding norm. In economics, this is the case in some game-theoretical treatments of norms.

12 Scott (Reference Scott2008) also refers to ‘external pressures’ coming from the ‘normative expectations’ of other agents when discussing the normative pillar (pp. 54–55), so that the normative view emphasizes ‘shared norms that are both internalized and imposed by others’ (p. 130). See also DiMaggio and Powell (Reference DiMaggio and Powell1983: 148, 153) on ‘normatively sanctioned’. If sanctions by other actors are also related to part of the normative pillar, the latter is not as different from the regulative pillar as it may seem.

13 Kuran (Reference Kuran1987a) considers ‘preference falsification’, which occurs when a preference that an individual expresses publicly differs from the one he would express secretly (see also Posner, Reference Posner1998, on signaling). Conformity is due to (reputational) sanctions whose value in Kuran's model increases with the number of other individuals expressing support for the norm, which in turn includes false enforcement. After a point, these sanctions more than compensate for the loss in utility caused by compromising one's integrity.

14 One of the reasons for this has already been discussed: the occurrence of coordination effects and increasing returns to adoption. Thus, although Scott rightly recognizes increasing returns as a possible basis for the process of institutionalization, I would not link them to the regulative pillar, as Scott does (Reference Scott2008: 121–123). There are differences between the explanation based on increasing returns to adoption and the one based on sanctions, as well as between them and other explanations that also share the assumptions of instrumental behavior and self-interest (not always selfish).

15 Kuran and Sunstein (Reference Kuran and Sunstein1999) combine this informational cascade argument with the reputational sanctions approach developed by Kuran (Reference Kuran1987a), which in turn, as mentioned above, includes increasing returns to adoption through sanctions.

16 What is the institution with which one is conforming and in what sense would these authors call it an institution?

17 DiMaggio (Reference DiMaggio and Zucker1988: 7–8) recognizes that ‘institutionalist theorists frequently imply or invoke individual and collective interests’, a fundamental example of which is ‘the assumption that humans have a preference for certainty and predictability in organizational life’. A similar argument is made by Keynes, Hayek and other economists, but would exclude innovations if generalized to all individuals at all times (Dequech Reference Dequech2003: 150, 161).

18 This dynamic view of legitimacy is inspired by the new institutional organizational analysis and by the economics of conventions (regarding the absence of questioning and the tests, respectively).

19 Greif (Reference Greif2006: 148) is a rare mainstream economist to argue that legitimacy is ‘central to institutional development’ (he seems to have political legitimacy and institutions established by decree especially in mind). He complains that ‘contemporary students of institutions in economics, political science, and economic sociology have little to say about it’. This statement does not, however, do justice to economists such as the conventionalists and to social scientists outside economics, like the proponents of the new institutionalism in organizational analysis (although Greif does cite some of their works positively).

20 I have also called this ‘cognitive legitimacy’, but have more recently found out that the same expression is used in a different sense by Aldrich and Fiol (Reference Aldrich and Fiol1994) or Suchman (Reference Suchman1995) and other organizational scholars after them, apart from the fact that, unlike them, I do not refer to legitimacy exclusively or mainly as an attribute of organizations or new ventures.

21 New institutionalists in organizational analysis have emphasized the importance of comprehensibility or intelligibility (sometimes as the criterion for a variant of ‘cognitive legitimacy’). This could also be seen as an example of epistemic value, but it is a prerequisite for the other ones. Intelligibility is a necessary, but not a sufficient, condition for epistemic legitimacy. Ideas that we understand very well may be considered unfounded.

22 Belief distortion and self-deception are involved in cases of groupthink, a concept that was originally developed in psychology and has recently begun to make its way into mainstream economics (Bénabou, Reference Bénabou2009; Shiller, Reference Shiller2002). The groupthink explanation of conformity could be said to combine the arguments based on sanctions and on epistemic legitimacy or informational mimetism.

23 On a more basic level, we can also refer to a norm of intelligibility, with its associated sanctions. An example is what Zuckerman (Reference Zuckerman1999) calls the categorical imperative, the need to fit minimally into an existing category.

24 The social obligation could be informed by the normative expectations of others and still be attributed to oneself and guide non-instrumental action, so that those expectations need not be felt as external pressures. There may be some tensions, however, in Scott's normative pillar, as he writes: ‘Normative controls are much more likely to be internalized than are regulative controls, and the incentives for conformity are hence likely to include intrinsic as well as extrinsic rewards’ (2008: 61, emphases added). In addition, as seen above, he also includes external pressures in the normative pillar, and not only in the regulative one. Mentioning incentives, rewards and pressures (sanctions) resembles an instrumental reading of internalization and legitimacy, associated with the logic of consequences.

25 The inclusion of epistemic legitimacy allows, among other things, a better reading of the role of professionalization in DiMaggio and Powell's (Reference DiMaggio and Powell1983) normative argument.

26 Kuran (Reference Kuran1987a) considers a different case, in which there is a gain in utility from not compromising one's integrity, instead of choosing an option because of external sanctions.

27 In contrast, the legitimacy that is part of DiMaggio and Powell's (Reference DiMaggio and Powell1983: 150) coercive mechanism is in the eyes of others.

28 This is not to be confused with Orléan's (Reference Orléan, Lesourne, Orléan and Walliser2006) concept of preferential imitation, which corresponds to, or at least includes, coordination and payoff externalities.

29 Klick and Parisi (Reference Klick and Parisi2008) consider this mechanism of adaptation to describe ‘conformity as an evolutionary strategy’. While they refer to a psychological literature that is suggestive of habituation, they emphasize a more conscious choice. In terms of the present framework, their approach to conformity as a process combines sanctions and legitimacy, starting with the former and then ending with the latter. It is one way of arriving at the legitimacy explanation for conformity at a point in time.

30 In the new institutional approach to organizations, taken-for-grantedness is an important concept, but, in Jepperson's (Reference Jepperson, Powell and DiMaggio1991: 147) assessment, ‘an ambiguous and underanalyzed one’. Some authors seem to restrict the term to a rule that is followed unconsciously. In this latter sense, the term should be avoided in the present section, which considers different explanations for the conscious adoption of a certain institutional rule (but it has been used in the above discussion of non-fully conscious behavior or thought). For the same reason, the term is not adequate here in broader senses that include, among other things, the possibility of unconscious adoption of a rule. This is the case if ‘taken for granted’ is used as synonymous with ‘unquestioned’. One may fail to question a rule either because one is not consciously aware of it or because one consciously treats it as the only possibility.

31 In his work with Deephouse, Suchman (Reference Deephouse, Suchman, Greenwood, Oliver, Suddaby and Sahlin2008: 53) equates taken-for-grantedness with an absence of questioning. As explained in a previous footnote, this means something broader than the conscious perception of inevitability. Scott considers legitimacy as possibly based on his different pillars, one of which involves ‘preconscious, taken-for-granted understandings’ (Reference Scott2008: 61), as already mentioned.

32 Although mainstream economics often attempts to naturalize institutional phenomena instead of pointing out their appearance of naturality as leading to conformity, Kuran (Reference Kuran1987b: 658) is an exception.

33 Garud and Karnøe (Reference Garud, Karnøe, Garud and Karnøe2001: 11) illustrate this in their discussion of technological innovations and existing practices and meanings in a complex environment: ‘in our quest to find simplicity in all this complexity, these meanings and practices become taken-for-granted. . . . [A]n entrepreneur may become so deeply embedded in these technological fields that a vision of the future that is different from the present is difficult to muster’.

34 There are a few interesting parallels between some of the different explanations for conformity with institutions discussed here and the factors that Beckert (Reference Beckert1999: 790–792) identifies as underlying institutional stability: namely, efficiency, legitimacy and power.

35 In the case of a rule that establishes privileges, the perception of either legitimacy or naturality may explain the conformity of all those who are privileged by the rule. If not, perhaps the preservation of privileges should be added as another possible explanation.

References

Akerlof, G. (1980), ‘A Theory of Social Custom, of Which Unemployment May be One Consequence’, Quarterly Journal of Economics, 94 (4): 749–775.Google Scholar
Akerlof, G. and Kranton, R. (2000), ‘Economics and Identity’, Quarterly Journal of Economics, 115 (3): 715–753.Google Scholar
Aldrich, H. and Fiol, M. (1994), ‘Fools Rush in? The Institutional Context of Industry Creation’, Academy of Management Review, 19 (4): 645–670.CrossRefGoogle Scholar
Aoki, M. (2001), Toward a Comparative Institutional Analysis, Cambridge, MA: MIT Press.CrossRefGoogle Scholar
Aoki, M. (2011), ‘Institutions as Cognitive Media between Strategic Interactions and Individual Beliefs’, Journal of Economic Behavior and Organization, 79 (1): 20–34.CrossRefGoogle Scholar
Arthur, W. B. (1989), ‘Competing Technologies, Increasing Returns and Lock-In by Historical Events’, Economic Journal, 99 (394): 116–131.Google Scholar
Becker, G. (1996), Accounting for Tastes, Chicago, IL: University of Chicago Press.CrossRefGoogle Scholar
Beckert, J. (1999), ‘Agency, Entrepreneurs, and Institutional Change. The Role of Strategic Choice and Institutionalized Practices in Organizations’, Organization Studies, 20 (5): 777–799.Google Scholar
Bénabou, R. (2009), ‘Groupthink: Collective Delusions in Organizations and Markets’, NBER Working Paper No. 14764.Google Scholar
Berger, P. and Luckmann, T. (1966), The Social Construction of Reality, New York: Doubleday.Google Scholar
Bernheim, B. D. (1994), ‘A Theory of Conformity’, Journal of Political Economy, 102 (5): 841–877.Google Scholar
Bessy, C. and Favereau, O. (2003), ‘Institutions et Économie des Conventions’, Cahiers d'Économie Politique, 44: 1–46.Google Scholar
Biggart, N. and Beamish, T. (2003), ‘The Economic Sociology of Conventions: Habit, Custom, Practice, and Routine in Market Order’, Annual Review of Sociology, 29: 443–464.CrossRefGoogle Scholar
Bikhchandani, S., Hirshleifer, D., and Welch, I. (1998), ‘Learning from the Behavior of Others: Conformity, Fads, and Informational Cascades’, Journal of Economic Perspectives, 12 (3): 151–170.Google Scholar
Camic, C. (1986), ‘The Matter of Habit’, American Journal of Sociology, 91 (5): 1039–1887.Google Scholar
Colyvas, J. and Powell, W. (2006), ‘Roads to Institutionalization: The Remaking of Boundaries between Public and Private Science, Research in Organizational Behavior, 27: 305–353.Google Scholar
Deephouse, D. and Suchman, M. (2008), ‘Legitimacy in Organizational Institutionalism’, in Greenwood, R., Oliver, C., Suddaby, R., and Sahlin, K. (eds.), The Sage Handbook of Organizational Institutionalism, Thousand Oaks, CA: Sage.Google Scholar
Denzau, A. and North, D. (1994), ‘Shared Mental Models: Ideologies and Institutions’, Kyklos, 47 (1): 3–31.Google Scholar
Dequech, D. (1999), ‘Expectations and Confidence under Uncertainty’, Journal of Post Keynesian Economics, 21 (3): 415–430.Google Scholar
Dequech, D. (2003), ‘Conventional and Unconventional Behavior Under Uncertainty’, Journal of Post Keynesian Economics, 26 (1): 145–168.Google Scholar
Dequech, D. (2006), ‘The New Institutional Economics and the Theory of Behaviour Under Uncertainty’, Journal of Economic Behavior and Organization, 59 (1): 109–131.Google Scholar
Dequech, D. (2007), ‘Neoclassical, Mainstream, Orthodox, and Heterodox Economics’, Journal of Post Keynesian Economics, 30 (2): 279–302.Google Scholar
Dequech, D. (2009), ‘Institutions, Social Norms, and Decision-Theoretic Norms’, Journal of Economic Behavior and Organization, 72 (1): 70–78.Google Scholar
DiMaggio, P. (1988), ‘Interest and Agency in Institutional Theory’, in Zucker, L. (ed.), Institutional Patterns and Organizations, Cambridge, MA: Ballinger.Google Scholar
DiMaggio, P. and Powell, W. (1983), ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Sociological Review, 48 (2): 147–160.Google Scholar
DiMaggio, P. and Powell, W. (1991), ‘Introduction’, in Powell, W. and DiMaggio, P. (eds.), The New Institutionalism in Organizational Analysis, Chicago, IL: University of Chicago Press.Google Scholar
Douglas, M. (1986), How Institutions Think, Syracuse, NY: Syracuse University Press.Google Scholar
Favereau, O., Biencourt, O., and Eymard-Duvernay, F. (2002), ‘Where Do Markets Come from? From (Quality) Conventions!’, in Lazega, E. and Favereau, O. (eds.), Conventions and Structures in Economic Organization, Cheltenham, UK: Edward Elgar.Google Scholar
Fleetwood, S. (2008a), ‘Structure, Institution, Agency, Habit and Reflexive Deliberation’, Journal of Institutional Economics, 4 (2): 183–203.Google Scholar
Fleetwood, S. (2008b), ‘Institutions and Social Structures’, Journal for the Theory of Social Behaviour, 38 (3): 241–265.Google Scholar
Garud, R. and Karnøe, P. (2001), ‘Path Creation as a Process of Mindful Deviation’, in Garud, R. and Karnøe, P. (eds.), Path Dependence and Creation, Mahwah, NJ: Lawrence Erlbaum Associates.Google Scholar
Gintis, H. (2009), The Bounds of Reason, Princeton, NJ: Princeton University Press.Google Scholar
Greif, A. (2006), Institutions and the Path to the Modern Economy, Cambridge: Cambridge University Press.Google Scholar
Hayek, F. (1973), Law, Legislation and Liberty, Chicago, IL: University of Chicago Press.Google Scholar
Hirsch, P. and Lounsbury, M. (1997), ‘Ending the Family Quarrel – Toward a Reconciliation of “Old” And “New” Institutionalisms’, American Behavioral Scientist, 40 (4): 406–418.Google Scholar
Hodgson, G. (2006), ‘What Are Institutions?’, Journal of Economic Issues, 40 (1): 1–25.Google Scholar
Hodgson, G. (2007), ‘Institutions and Individuals: Interaction and Evolution’, Organization Studies, 28 (1): 95–116.Google Scholar
Jepperson, R. (1991), ‘Institutions, Institutional Effects, and Institutionalism’, in Powell, W. and DiMaggio, P. (eds.), The New Institutionalism in Organizational Analysis, Chicago, IL: University of Chicago Press.Google Scholar
Johnson, C., Dowd, T., and Ridgeway, C. (2006), ‘Legitimacy as a Social Process’, Annual Review of Sociology, 32: 53–78.Google Scholar
Keynes, J. M. (1936), The General Theory of Employment, Interest and Money, London: Macmillan.Google Scholar
Keynes, J. M. (1937), ‘The General Theory of Employment’, Quarterly Journal of Economics, 51: 209–223.Google Scholar
Klick, J. and Parisi, F. (2008), ‘Social Networks, Self-Denial, and Median Preferences: Conformity as an Evolutionary Strategy’, Journal of Socio-Economics, 37 (4): 1319–1327.Google Scholar
Kuran, T. (1987a), ‘Chameleon Voters and Public Choice’, Public Choice, 53 (1): 53–78.Google Scholar
Kuran, T. (1987b), ‘Preference Falsification, Policy Continuity and Collective Conservatism’, Economic Journal, 97 (387): 642–665.Google Scholar
Kuran, T. and Sunstein, C. (1999), ‘Availability Cascades and Risk Regulation’, Stanford Law Review, 51 (4): 683–768.Google Scholar
Langlois, R. (1998), ‘Rule-following, Expertise, and Rationality: A New Behavioral Economics?’, in Dennis, K. (ed.), Rationality in Economics: Alternative Perspectives, Boston, MA: Kluwer.Google Scholar
Lawrence, T. and Suddaby, R. (2006), ‘Institutions and Institutional Work’, in Clegg, S.et al. (eds.), Handbook of Organization Studies, 2nd edn, London: Sage.Google Scholar
Lawson, T. (1997), Economics and Reality, London: Routledge.Google Scholar
Meyer, J. and Rowan, B. (1977), ‘Institutionalized Organization: Formal Structure as Myth and Ceremony’, American Journal of Sociology, 83 (2): 340–363.Google Scholar
North, D. (1990), Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press.Google Scholar
Orléan, A. (2004), ‘L'économie des Conventions: Définitions et Résultats’, in Orléan, A. (ed.), Analyse Économique des Conventions, 2nd edn, Paris: PUF.Google Scholar
Orléan, A. (2006), ‘Mimetic Interactions’, in Lesourne, J., Orléan, A., and Walliser, B. (eds.), Evolutionary Microeconomics, Berlin: Springer.Google Scholar
Posner, E. (1998), ‘Symbols, Signals, and Social Norms in Politics and the Law’, Journal of Legal Studies, 27 (2): 765–798.CrossRefGoogle Scholar
Powell, W. and Colyvas, J. (2008), ‘The New Institutionalism’, in Clegg, S. and Bailey, J. (eds.), International Encyclopedia of Organization Studies, Thousand Oaks, CA: Sage.Google Scholar
Scott, W. R. (2008), Institutions and Organizations: Ideas and Interests, 3rd edn, Thousand Oaks, CA: Sage.Google Scholar
Shiller, R. (2002), ‘Bubbles, Human Judgment and Expert Opinion’, Financial Analysts Journal, 58 (3): 18–26.Google Scholar
Stark, D. (2009), The Sense of Dissonance, Princeton, NJ: Princeton University Press.Google Scholar
Storper, M. and Salais, R. (1997), Worlds of Production, Cambridge, MA: Harvard University Press.Google Scholar
Suchman, M. (1995), ‘Managing Legitimacy: Strategic and Institutional Approaches’, Academy of Management Review, 20 (3): 571–610.Google Scholar
Thévenot, L. (2002), ‘Conventions of Co-ordination and the Framing of Uncertainty’, in Fullbrook, E. (ed.), Intersubjectivity in Economics, London: Routledge.Google Scholar
Tolbert, P. and Zucker, L. (1996), ‘The Institutionalization of Institutional Theory’, in Clegg, S., Hardy, C., and Nord, W. (eds.), Handbook of Organization Studies, Thousand Oaks, CA: Sage.Google Scholar
Young, H. P. (2001), ‘The Dynamics of Conformity’, in Durlauf, S. and Young, H. P. (eds.), Social Dynamics, Cambridge, MA: MIT Press.Google Scholar
Zucker, L. (1977), ‘The Role of Institutionalization in Cultural Persistence’, American Sociological Review, 42 (5): 726–743.Google Scholar
Zucker, L. (1987), ‘Institutional Theories of Organization’, Annual Review of Sociology, 13: 443–464.Google Scholar
Zuckerman, E. (1999), ‘The Categorical Imperative: Securities Analysts and the Illegitimacy Discount’, American Journal of Sociology, 104 (5): 1398–1438.Google Scholar