INTRODUCTION
Degradation of the natural environment continues at a pace beyond that which the planet can sustain. Governments have historically attempted to curtail environmental degradation using regulation and economic instruments, but such interventions are usually not sufficient to bring about a sustainable development path. Policy instruments are subject to failures, as violations may not be detected, and regulations may fail to be sufficiently enforced. Furthermore, it is often not clear what kind of regulations or interventions will be most effective in an increasingly global and interdependent world (Rugman, Kirton & Soloway, 1997; Rugman & Verbeke, 1998). Market mechanisms are also subject to failure as consumers are largely unable to verify "green" claims by manufacturers, and media exposes of false claims lead all such claims to be suspect.
In the midst of such market and policy failures, the importance of voluntary efforts by firms to improve their environmental performance is increasing. There is growing recognition that "...changes in corporate organization, culture and procedures can yield environmental improvement in ways that a compliance-based [regulatory] approach cannot" (Roht-Arriaza, 1997:294). [1] We contend that a firm's commitment to pro-environmental values and actions is a good predictor of its long run environmental performance. It is thus important for those desiring sustainable development to understand what drives firms to commit to voluntary, pro-environmental efforts.
At least since Thompson's explanation of systems theory in 1967, the idea that firms must be aligned with their task environments has been a dominant assumption of organizational theorists and strategists. Firms that are out of alignment are expected to experience performance decrements, assessments of illegitimacy, and pressures for conformity. With severe misalignments, crisis can be the result. Yet there are different perspectives within the literature as to how these pressures for alignment come about and how they are experienced and responded to by firm members.
Economic explanations for the voluntary greening of firms point to incentives in the economic environment that reward pro-environmental behaviour and penalize behaviour that damages the environment (Baumol & Oates, 1988; Kneese & Schultz, 1975). For example, consumer markets that demand environmental responsibility force firms to adapt or forego sales, and firms that have greater consumer contact have been found to over-comply with environmental legislation (Arora & Cason, 1996). Firms in highly regulated industries have also been found more likely to over-comply with regulation (McKinsey & Company, 1991; UNCTAD, 1993), often in strategic attempts to influence the direction of future regulation such that it is more favourable to the firm (Porter & van der Linde, 1995). Given that firms in the same industry and geographical area often face similar consumer markets and regulatory regimes, we might expect similar environmental stances from these firms.
Sociological explanations from institutional theory also point to convergence among firms in the same industry/geographical space. Firms are expected to reflect the dominant forms, practices and interpretive frames (i.e. institutions) of the organizational field in which they are embedded (Meyer & Rowan, 1977), and these institutions are considered stable, enduring, and largely taken for granted by organizational field members (Scott, 1995). Institutional theory uses the "startling homogeneity of organizational forms and practices" (DiMaggio and Powell, 1983: 148), as its starting point. Norms, values, forms and practices are diffused within organizational fields by interaction among industry members (e.g., within trade associations), consultants who specialize in the industry, the common requirements of financial institutions and capital markets (Hoffman, 1997; Jennings, Zandbergen & Martens, 1997), and executive migration (Kraatz & Moore, 1998). Firms that are closely related in industry and location can al so expect to face similar social pressures from stakeholders such as environmental groups, communities, employees, and others. Thus, both sociological and economic explanations suggest that firms in the same industry and location are likely to act in similar ways.
Yet, empirically, this is not always the case. Canfor and MacMillan Bloedel (MB) are two forestry companies operating in British Columbia that have faced similar regulatory, legal, industry, customer, raw material, economic, social, political and other environments, yet they have responded quite differently to the natural environment. Canfor reacted early to pressures to protect the environment, becoming an industry leader in environmental protection by the mid to late 1980s. By the late 1990s, Canfor retreated somewhat from this leadership position, focusing more on profitability issues.
MB's environmental record was subject to strong public criticism. Targeted by Greenpeace and attracting international boycotts due to the clearcut logging of old growth forests in Clayoquat Sound, MB repeatedly resisted changes in environmental legislation and ignored, discounted, and prosecuted environmental protestors. As recently as 1997, MB leaders claimed publicly that an end to clearcutting would be the end of MB in BC. In 1998, MB graduated from being an environmental 'pariah' to a Greenpeace 'darling', by announcing an end to clearcut logging and a new commitment to sustainability.
These companies illustrate a dilemma for standard sociological and economic approaches. If a firm's economic and social environment determines its responses, why do firms facing similar environments differ so dramatically? Furthermore, if an organization's past largely determines its future through the inertia of stable and enduring institutions (Meyer & Rowan, 1977, Selznick, 1957), what causes these organizations to change their responses to the environment over time?
This paper addresses these questions by focusing on individuals as interpreters of their environments. Individuals in organizations focus on different aspects of the firm's external and internal environments, depending on the cognitive frame through which they look at the world. Cognitive frames are mental representations of a particular aspect of the world (Fiske & Linville, 1980), that are used by individuals to interpret and make sense of their world. Frames can come to be collectively held within organizations (Kitayama et al., 1997; Daft & Weick, 1984), especially through the influence of the organizational leader.
In this paper, we review the literature on frames and on the influence that a leader's frame has on the organizational frame. We then review previously published empirical work about organizations' responses to the environment, and use it to conceptualize a typology of frames that leaders of organizations hold about the environment. By focusing on leaders' cognitions, we can go beyond merely describing behaviours as they exist to focus on how change takes place. We predict pathways along which leaders can move from one frame to another. We propose that movements in leaders' frames correspond to changes in organizational responses to the environment, through direct actions and resource commitments by the leader, and through movement in the collective frame over time.
We end by returning to the stories of MacMillan Bloedel and Canfor, to examine how our concept of frames and frame dynamics holds up in light of their experiences. We conclude with a discussion of limitations of our model and directions for further research.
LITERATURE REVIEW
Framing In Organizations
Cognitive frames, cognitive maps, causal maps, schemata, and scripts all refer to conceptually similar constructs, but different terms and slight variations in focus are used by different researchers (Fiske & Linville, 1980; Laszlo, Artigiani, Combs, Csanyi, 1996; Laukkanen, 1998; Weick, 1995; Walsh, 1995). We will focus on cognitive frames. In general, these refer to the way knowledge is abstracted from examples and stored cognitively. Frames arise because repeated encounters with complex issues and conflicting pressures from the external environment to deal with such issues in certain ways result in both learning and the need to simplify complexities (Fiske & Linville, 1980). The abstracted knowledge in frames can consist of facts, affective responses, behaviours, attitudes, beliefs and values, and their interrelationships. The interrelated material adheres as a category, allowing simplified, heuristic decision making and speeding problem solving (Fiske & Taylor, 1984).
Frames are used to make sense of stimuli coming from the natural, economic and social environments (Weick, 1995). They signal what is to be attended to (White & Carlston, 1983), and allow individuals to anticipate what they are likely to see. When information is ambiguous or missing, frames provide default values based on past experience (Langer & Abelson, 1974). New information is perceived (or not perceived) and interpreted according to how it fits with existing cognitive frames. Frames therefore guide both search and perception, and not only represent individual realities, but also help to create them (Laszlo, et al., 1996). While not all cognition is accomplished via cognitive frames (theory-driven or top-down approaches), Louis and Sutton (1991) claimed that theory-driven processing may dominate for all but the most novel situations.
Shared frames develop within organizations (Daft & Weick, 1984; Weick, 1995). Kitayama et al. (1997) stress that the way a situation is defined and understood by group members depends on communication. One member (e.g., the leader) communicates a definition of the situation either by words or by actions, and the others understand and interpret the definition, confirming, modifying or challenging it by their responses. A common frame of reference emerges. As frames become enacted and reinforced by organization members, they become institutionalized into organizational knowledge, routines, values and norms (Daft & Weick, 1984; Weick, 1995).
Organization leaders play a powerful role in the development of the organization's shared frame: they make sense of issues for their members (Daft & Weick, 1984; Thomas, Clark & Gioia, 1993), legitimize them through attention and communication (Gioia & Thomas, 1996; Sharma, Pablo & Vredenburg, 1999) and shape collective frames and behaviour through strategies, budgets, and reward and control systems (Gioia & Thomas, 1996; Portugal & Yukl, 1994). By examining leaders' frames, we can understand firms' resource allocations and the shared frames that develop among employees. While the leader's frame does not solely determine the shared frame of the organization, in general we can expect the leader's frame to have the most prominent influence on the organization's collective frame. For example, Winn (1995) found that employees other than the CEO may also act as environmental issues leaders, however top management commitment to environmental improvements results in faster and more comprehensive diffusion of policy changes and a greater impact on the organization as a whole. Environmental issue leaders often work through the CEO to promote organizational greening.
Shared frames in organizations can be inferred from organizations' actions and communications. The communications of organizational leaders, and those who have leadership responsibility for a particular area (in this case, the relationship between the business and the natural environment) are especially important.
It is important to note that frames are not the same as actions. Frames indicate the cognitions behind actions, including the values, beliefs, intentions and assumptions. While we can examine actions and infer cognitions, different cognitions can motivate the same short-term behaviours. For example, a firm that is threatened …