Herd behavior refers to the phenomenon of people following a crowd for a given period, sometimes "even regardless of individual information suggesting something else" (Banerjee 1992, 798). The phenomenon of herd behavior was among the first topics studied in social psychology (Van Ginneken 1992). Early economists like Thorstein Veblen (1899) and sociologists like Georg Simmel ( 1957) applied it to sudden shifts in consumer behavior such as fashions and fads. More recently, issues relating to herd behavior have again caught the eye of economists and management scholars. In 2003, Marlene Fiol and Edward O'Connor, for instance, maintained that the decisions of individuals to do whatever anyone else is doing can be applied to decision-making processes in organizational settings. In the economic approach, however, the concept is turned into something which differs from herd behavior in social psychology. As a result, researchers in the social sciences these days face inconsistent conceptions of herd behavior.
Herd behavior, however, cannot be fully understood from a single perspective alone. What was argued by John R. Commons (1934) for economics and psychology in general could also be applied to the study of herd behavior. Although both disciplines ask what herd behavior is, the economic perspective primarily is to focus on long-term effects, to study the value of (partaking in) herd behavior and how much one can benefit from it. The motivations underlying herd behavior are viewed in terms of the choices they produce. With respect to this mainstream economics approach, Geoffrey Hodgson (1993) came to the conclusion that, in doing so, many economists have taken individual motivations and preferences as given, because "the essential aspects of human personality and motivation are conceived of as independent of the social relations with others" (236). The psychological perspective, on the other hand, is to account for the subjective value of herd behavior per se. By asking "why" and "when" it occurs, the motivations underlying herd behavior are more broadly viewed in terms of the processes involved. With respect to the mainstream psychological approach, however, Jaap van Ginneken (2003, 2004) concluded that, in doing so, psychologists have hardly paid any attention to the influence which institutional settings may have. Especially when studying herd behavior in the business environment, one needs to answer the "why" and "when" of herding on both the individual and institutional level in order to understand the "how" and "how much" of it.
In this paper, first the problem of herd behavior is made clear via the historical development of the concept in economics and social psychology. Accordingly, the focus is on differences between the approaches. Second, an integrated economic psychological approach to herd behavior is proposed, attempting to overcome theoretical and methodological differences. To demonstrate its importance and to illustrate the theoretical and methodological problems that need to be overcome, the framework is applied to decision making in the presence of groupthink, a form of herd behavior that tends to be more problematic within institutional settings.
The History of Herd Behavior
Herd Behavior in Economics
In early economics, herd behavior gained the attention of researchers like Veblen that studied sudden shifts in consumer behavior such as fads and fashions. In his famous 1899 study on The Theory of the Leisure Class, Veblen wrote of patterns of conspicuous consumption in which people engage in actions by making comparisons with similar people who are slightly better off in order to express pecuniary strength. Consumption, however, was considered to be a passive activity without real value for society and the economy and was not given much attention in the field of economics (Dolfsma 2000). As a consequence, neither was the phenomenon of herd behavior.
In the early 1950s, Harvey Leibenstein introduced the social psychological bandwagon metaphor in economics. Remarkably, the bandwagon phenomenon, which originally referred to a wagon full of playing musicians that was followed by a big crowd, was not defined in terms of what it was, but in terms of its effects. Leibenstein defined a bandwagon as "the extent to which demand for a commodity is increased due to the fact that others are also consuming the same commodity" (Leibenstein 1950, 189). Drawing from psychological perspectives on social influence, he explained the motivations underlying herd behavior as "the desire of people to purchase a commodity in order to get into 'the swim of things'; in order to conform with the people they wish to be associated with; in order to be fashionable or stylish; or, in order to appear to be 'one of the boys.'" For decades this article and its treatment of herd behavior in consumption remained a much referred to and often-cited exception in the field of economics (Dolfsma 2000). (1)
Remarkably, for decades sociologists instead of economists further developed the topic of herd behavior. Economists, however, showed a great willingness to integrate sociological frameworks into their discipline. Two research streams, one related to diffusion of innovations and the other related to social network analysis, became particularly influential in economics. As in economics, in research on the diffusion of innovations the focus was on the extent to which innovations were adopted or not, and with what effect (Rogers  1995). Everett Rogers ( 1995) defined diffusion as "the process by which an innovation is communicated through certain channels over time among the members of a social system" (5). As innovations mostly diffused via S-shaped curves and the social system gave regularity and stability to human behavior, diffusion processes could be predicted to a large degree. Rogers offered a comprehensive and extensive overview of diffusion research. Among other things, opinion leadership in diffusing innovations and innovativeness of members of a social system were presented as main independent variables in diffusion research. Remarkably, no psychological research into diffusion, contagion, and leadership processes was included. Nonetheless, the study was a classic work and the ideas developed in the book were applied in disciplines other than psychology, such as sociology, and especially connected to social network analyses of collective behavior.
In the 1970s and 1980s, a social network approach to collective behavior became particularly influential in economics. In 1973, addressing more fundamental aspects underlying collective behavior, Mark Granovetter stressed the importance of weak ties in diffusion processes. In general, the strength of interpersonal ties in large networks of relations could be defined as "a combination of the amount of time, the emotional intensity, the intimacy, and the reciprocal services which characterize the tie" (1361). A weak tie, someone who was not too close and similar to oneself, often served as a bridge between different groups of people, which implied that "whatever is to be diffused can reach a larger number of people and traverse greater social distance when passed through weak ties rather than strong" (1366). Herd behavior, for instance, relating to the adoption of innovations, was thus more likely to occur via the weak ties of relations with people that were rather dissimilar to oneself.
In addition, in 1978 Granovetter also proposed a threshold model to collective behavior. In this model, people had a binary choice to engage in an action or not. This decision depended on perceived costs and benefits to the person in relation to a threshold, or "the number or proportion of others who must make a decision before a given actor does so; this is the point where net benefits begin to exceed net costs for that particular actor" (1420). If someone reached one's threshold in a given situation, bandwagon effects would occur when the person decided to join the herd. Again, the number of other people already partaking in a given action were thought to motivate herd behavior. The assumptions underlying the model were that of individuals as rational actors and the existence of complete information.
In the 1980s, Ronald Burr (1982, 1987) developed another argument for the diffusion of innovations. Drawing from several grand theories in social psychology (Festinger, Schachter, and Back 1950; Homans 1950, 1961), he argued that in ambiguous situations people turned to other people that served as a reference group in order to come up with a solution that made sense in that particular context. In such situations, people would not follow others due to exchange of information. People would rather join a crowd as a result of the observation that people that were part of their reference group had already adopted it. More precisely, herd behavior occurred if someone "who could replace him in his role relations if he were removed from the social structure" engaged in a particular action (1987, 1294). Therefore, innovations merely diffused via people that were "structurally equivalent" to each other, meaning that they occupied "the same position in the social structure and [were] so proximate to the extent that they [had] the same pattern of relations with occupants of other positions" (1291).
In the mid 1980s, the bandwagon metaphor of herd behavior was again connected to an economic line of research, related to the newly introduced concept of increasing returns. Whereas bandwagon effects interacted with traditional market forces such as price to create distinct consumption patterns on the micro level, people like Paul David (1985) and Brian Arthur (1988, 1989, 1994) stressed that, on the macro level, bandwagon effects could result in increasing returns and network externalities in consumption, leading to the more or less stable evolution of average consumption preferences...