5 Study 1: Symbolic gaps in environmental strategies
Conventional corporate environmentalism has become popular in both the academic literature and corporate boardrooms. Since the 1990s, both firms and governments have been moving away from the traditional command-and-control approach to environmental regulation and instead are increasingly reliant on firms to adopt voluntary green solutions that are not direct legal requirements. As managers adopt voluntary environmental strategies, stakeholders such as shareholders, government, NGOs and other actors determine whether a firm has gained legitimacy. An important outcome of corporate environmentalism schemes is ‘symbolic performance’ – or the extent to which adopting a green solution generates positive social evaluations.
However, as discussed in Chapter 4, a potential problem when firms voluntarily go beyond compliance with environmental regulations is that they might produce more symbolic than substantive green solutions. This leads to a ‘symbolic gap’ – or the difference between substantive environmental improvements and symbolic performance from corporate environmentalism. Symbolic gaps could arise deliberately as firms exert their discretionary power to produce audience-seeking rather than performance-enhancing green solutions or as an unintended consequence of the lower cost of adopting symbolic green solutions. In either case, firms face the tension that whereas they may gain symbolic performance by adopting symbolic green solutions, they may lose it if the difference between symbol and substance is exposed.
Despite the importance of symbolic gaps, insufficient research has focused on the circumstances under which they are more likely to persist. In this chapter, I argue that powerful firms are more likely to seek symbolic performance and generate symbolic gaps. Symbolic gaps are likely to persist when there is relatively weak monitoring of self-regulation schemes. Large firms have more status, resources and political connections, which often gives them systemic power within fields. We therefore can obtain initial insight into symbolic gaps surrounding firms’ environmental strategies by collating all of the available empirical research evidence concerning the relationship between firm size and corporate environmentalism across different contexts.
In this chapter, I develop an argument based on large firms’ decoupling and systemic power to explain when symbolic gaps can persist. A meta-analysis of empirical studies on firm size and corporate environmentalism broadly supports three predictions. First, large firms seek symbolic performance through proactive environmental strategy more than small firms. Second, due to decoupling, large firms exhibit a symbolic gap; that is, the relationship between size and self-regulation is stronger for intended strategy statements than for substantive impacts. Third, because large firms have the systemic power to shape the self-regulatory ‘game’, the extent to which they seek symbolic performance and maintain symbolic gaps is higher in less-monitored fields.
Proactive environmental strategies and power
Firms have adopted corporate environmentalism in response to pressures arising from regulators, NGOs, customers and other actors in institutional surroundings (Bansal Reference Bansal2005; Jennings and Zandbergen Reference Jennings and Zandbergen1995). As previously discussed, corporate environmentalism can take many forms, including a wide variety of firm responses to institutional pressures to act on environmental issues including (but not limited to) altering a firm’s strategic stance to incorporate greener aims, launching more environmentally sound products, altering materials-transformation processes for environmental benefit, implementing pollution abatement technologies or even simply making an environmental policy declaration. To the extent that these green solutions are an attempt to voluntarily control firms’ collective actions on the environment, all of these actions could be examples of industry self-regulation on green issues (King and Lenox Reference King and Lenox2000).
We know that (1) there is considerable variation in environmental self-regulation across firms, even when institutional pressures are similar (Bansal Reference Bansal2005; Russo and Fouts Reference Russo and Fouts1997); (2) legitimacy seeking is a powerful motive for self-regulation (King and Lenox, Reference King and Lenox2000); (3) self-regulation strategies are laden with signalling and symbolic behaviour (Delmas and Montes-Sancho Reference Delmas and Montes-Sancho2010; Phillipe and Durand 2011); and (4) self-regulation at the firm level is interrelated with government regulation at the field level (Reid and Toffel Reference Reid and Toffel2009). However, we do not yet know much about the field-level contingencies that might influence large firms’ self-regulatory strategies or the extent to which they might be symbolic or substantive green solutions.
Scholars have generated an increasing number of empirical studies on the predictors of self-regulation of environmental issues (for a recent review, see King, Prado, and Rivera Reference King, Prado, Rivera, Hoffman and Bansal2012). These studies typically are aimed at uncovering the triggers of proactive corporate environmentalism – that is, strategy choices on environmental issues that go beyond basic legal compliance (Hart Reference Hart1995; Roome Reference Roome1992). Symbolic gaps have been noticed in the context of self-regulation, particularly in environmental reporting (see, e.g., Laufer Reference Laufer2003); in the ceremonial adoption of environmental management schemes, such as ISO 14001(see, e.g., Boiral Reference Boiral2007); and in symbolic participation in voluntary agreements (Delmas and Montes-Sancho Reference Delmas and Montes-Sancho2010; King and Lenox Reference King and Lenox2000).
As noted in Chapter 4, not all green solutions are equal. Researchers are increasingly becoming aware that proactive corporate environmentalism measures based on stated intent, environmental initiative implementation, and actual mitigation of environmental impacts tap into different dimensions of environmental proactivity (Branzei et al. Reference Branzei, Ursacki-Bryant, Vertinsky and Zhang2004; Chatterji and Toffel Reference Chatterji and Toffel2010; Russo and Harrison Reference Russo and Harrison2005). However, the link between recognising symbolic corporate environmentalism and the empirical practice of measuring several of its dimensions has not yet been fully exploited. Exploiting differences between dimensions of corporate environmentalism allows us to examine symbolic gaps as firms attempt to gain and protect symbolic performance; they may state proactive environmental intentions or implement environmental initiatives without reducing their actual environmental impact. My elaboration in this chapter is to explicitly consider the role of firm size in maintaining symbolic gaps in corporate environmentalism across different social fields.
Surprisingly little research attention has been given to when and how symbolic gaps persist within fields. Although research suggests that large, visible firms are more likely to respond to institutional pressures (Pfeffer and Salancik Reference Pfeffer and Salancik1978), little is known about whether and when they are more likely to generate and maintain symbolic gaps. This oversight is particularly problematic because large firms often benefit from high-status social evaluations (Fombrun and Shanley 1990) and therefore have discretionary power in defining appropriate institutional responses (Phillips and Zuckerman Reference Phillips and Zuckerman2001).
Prior studies incorporated aspects of power into explanations of symbolic gaps. For example, Westphal and Zajac (Reference Westphal and Zajac2001) found that powerful CEOs were more likely to only symbolically adopt stock repurchase programmes. Tilcsik (Reference Tilcsik2010) examined how the contest for power within an organisation eventually led to the closing of a symbolic gap. More generally, Boxenbaum and Jonsson (Reference Boxenbaum, Jonsson, Greenwood, Oliver and Sahlin-Anderson2008) suggested that power relationships within a field may influence the likelihood of decoupling but they did not say how or why. I respond to their challenge by recognising that institutional pressures arise within a network of social relationships (Bourdieu Reference Bourdieu1990; Wooten and Hoffman Reference Wooten, Hoffman, Greenwood, Oliver and Sahlin-Anderson2008) wherein institutional control is based on routine, systemic power that advantages particular groups (Lawrence Reference Lawrence, Greenwood, Oliver and Sahlin-Anderson2008; Oakes, Townley, and Cooper Reference Oakes, Townley and Cooper1998). Large firms are highly visible and therefore subject to resulting institutional pressures, leading to a tendency to decouple proactive words from actions (Meyer and Rowan Reference Meyer and Rowan1977). We recall from Chapter 4 that these firms face a lower relative cost of audience-seeking activities and therefore tend to invest more in symbolic activities and earn higher levels of social esteem for equivalent levels of overall investment in green solutions. However, large firms also wield status and symbolic power that individuals misrecognise as legitimate authority to shape the rules of the game. Thus, whereas decoupling arguments explain symbolic gaps as ‘rationalised myths’ at the firm level, a more power-based approach examines how myths are formed by the powerful and then misrecognised as legitimate within the broader social system (Bourdieu Reference Bourdieu2005).
In this chapter, I develop a power-based logic to investigate whether and when large firms are more likely to generate and maintain symbolic gaps. This is an important research concern because large firms often are key players in responding to institutional demands and also may have the discretionary power required to separate symbolic from substantive responses (Philippe and Durand Reference Philippe and Durand2011). The extent to which large firms have the power to shape the rules of the game and to benefit from symbolic gaps in different contexts has been under-explored. One reason why the field-level determinants of persistent symbolic gaps among large firms have not been adequately addressed so far may be the challenge of data requirements. I overcome this challenge by using meta-analytic techniques to test hypotheses with data from across a wide variety of fields.
The two overarching research questions in this chapter are as follows: (1) Does separating symbolic from substantive performance provide different answers about drivers of corporate environmentalism in large, powerful firms? (2) When can symbolic gaps persist between the likelihood that large, powerful firms will adopt corporate environmentalism through proactive strategic intent, implementation or impact? I argue that the extent to which large firms seek symbolic performance and can maintain symbolic gaps depends on the extent of monitoring within fields. I hypothesise that in highly monitored fields, large firms are more likely to seek symbolic performance but will have more difficulty maintaining a symbolic gap. I test the hypotheses by conducting a meta-analysis of empirical corporate environmentalism studies that capture data from a diverse range of fields. An ancillary empirical contribution of this chapter is that by conducting an across-field meta-analysis, I provide guidance to corporate environmentalism researchers about when large firms are likely to be more environmentally responsive, in both words and actions.
Explaining symbolic gaps: Decoupling and power
A symbolic gap arises when there is a difference between a firm’s stated intentions designed to shift social evaluations and its substantive environmental impacts on a given issue within a field. We know that firms may reap legitimacy benefits from gaining symbolic performance and maintaining a symbolic gap. However, researchers have focused less attention on how these benefits are threatened if a symbolic gap is exposed. Firms may gain economic benefit from signalling proactive corporate environmentalism policy statements, but these benefits can be lost if the strategy is perceived as merely symbolic. Institutional theory suggests that persistent symbolic gaps can be explained by decoupling at the firm level, whereas a more politicised approach suggests an alternative logic based on power within a field.
Decoupling. Institutional theorists focus on organisational legitimacy, defined as ‘a generalised perception or assumption that the actions of an entity are desirable, proper or appropriate within some socially constructed system of norms, values, beliefs and definitions’ (Suchman Reference Suchman1995: 574). Maintaining legitimacy can enhance firm survival but also can be challenging when firms are faced with conflicting demands and uncertainty about the efficiency of proposed solutions. Institutional theorists describe this as ‘decoupling’, in which organisations adopt new practices only ceremonially without necessarily fully implementing them (Endelman Reference Endelman1992; Meyer and Rowan Reference Meyer and Rowan1977). Within this view, persistent symbolic gaps are explained by decoupling, in which organisations separate their ‘talks, decisions and actions’ to cope with conflicting institutional demands (Brunsson Reference Brunsson1989). Large firms may be more inclined to adopt only ceremonially because they are more visible and need to signal legitimacy (Westphal, Gulati, and Shortell Reference Westphal, Gulati and Shortell1997).
Power. However, the typical decoupling logic gives too little attention to systemic power – that is, power that works through routine and taken-for-granted practices that advantage particular groups (Lawrence Reference Lawrence, Greenwood, Oliver and Sahlin-Anderson2008). High-status firms have the systemic power to shape what is considered to be legitimate within a given field (Deephouse and Suchman Reference Deephouse and Suchman2008). Once new symbols are established, they continue to define what is deemed legitimate and can exert considerable institutional control (Lawrence Reference Lawrence, Greenwood, Oliver and Sahlin-Anderson2008). Dominant organisations wield their systemic power to generate symbolic power that resides in words, language and symbols valued by community members (Bourdieu Reference Bourdieu1991). Once established, these symbols are ‘misrecognised’ as legitimate responses to institutional pressures because the systemic power that privileges high-status organisations is invisible. Thus, persistent symbolic gaps can be explained by misrecognition, wherein less powerful members of a field believe in the legitimacy of symbolic power as well as the legitimacy of high-status organisations that wield it.
Large firms may have more direct political and economic power through their access to more resources, larger numbers of stakeholders (akin to voters) and ability to capture rents from public policy (Hillman et al. Reference Hillman, Keim and Schuler2004). Large firms also are more likely to benefit from higher prestige (Mas-Ruiz and Ruiz-Moreno Reference Mas-Ruiz and Ruiz-Moreno2011); hence, the potential to deviate from or define new social norms (Phillips and Zuckerman Reference Phillips and Zuckerman2001). Larger firms also are better able to resist external stakeholder pressure (Darnall et al. Reference Darnall, Henriques and Sadorsky2010; Meznar and Nigh Reference Meznar and Nigh1995). All of these characteristics suggest that large firms are likely to enjoy higher status and therefore systemic power. In this chapter, I build on the features of large firms and develop a power-based logic to contend that in more-monitored fields, large firms are more likely to seek symbolic performance but have more difficulty in maintaining symbolic gaps.
Hypothesis development
Figure 5.1 illustrates the relationships between firm size and three different dimensions of corporate environmentalism. The first is ‘stated intent’ – that is, environmental policies or corporate announcements that communicate a firm’s intentions on environmental issues. The second is ‘initiative implementation’, wherein a firm invests in changing a particular process, product or organisational practice primarily for environmental reasons (e.g., it introduces a recycling scheme or collects environmental information from suppliers). The third dimension is ‘impact mitigation’, which captures measurable decreases in the substantive impact of the firm on the natural environment (e.g., emissions reductions, waste reductions or reduced energy use).1 We can expect different effect sizes between firm size and each of these proactive responses within and between fields (i.e., the effect sizes A, B and C in Figure 5.1 are not equal). In the next section, I use decoupling and power arguments to predict when large firms are most likely to seek symbolic performance and can maintain symbolic gaps.
Figure 5.1 Relationships between firm size and corporate environmentalism.
Symbolic performance
The quest for legitimacy has been shown to be a primary driver of proactive corporate environmentalism (Bansal Reference Bansal2005; Jennings and Zandbergen Reference Jennings and Zandbergen1995). Larger organisations are likely to be subject to more intense public scrutiny and therefore more likely to signal responsiveness to institutional pressures (Pfeffer and Salancik Reference Pfeffer and Salancik1978). Furthermore, dominant players in fields generate and protect symbolic performance, so we would expect large, high-status, established firms to seek to achieve social gains through audience-seeking activities and signalling conformity with social demands. As large firms navigate new institutional pressures, they generate new symbols to signal their conformity, including environmental policies (Henriques and Sadorsky Reference Henriques and Sadorsky1996), codes of ethics (Stevens et al. Reference Stevens, Steensma, Harrison and Cochran2005), certification schemes (Jiang and Bansal Reference Jiang and Bansal2003), and industry voluntary agreements (King and Lenox Reference King and Lenox2000). These symbols are imbued with value by stakeholders seeking improved environmental performance and thus become legitimate and tradable green solutions. The largest firms are best positioned to use their status to gain symbolic performance through proactive corporate environmentalism because they are perceived as dominant players in a field (Bourdieu Reference Bourdieu2005).
Several empirical studies of industry self-regulation suggest a positive relationship between firm size and corporate environmentalism. Henriques and Sadorsky (Reference Henriques and Sadorsky1996), for example, argued that visible firms are more susceptible to public scrutiny or may be called on to act as industry leaders; therefore, they are more likely to possess environmental plans. Similarly, Sharma and Nguan (Reference Sharma and Nguan1999) suggested that larger firms are subject to greater media scrutiny and are forced to adopt a leadership stance on biodiversity conservation to protect their reputation. Hettige et al. (Reference Hettige, Huq, Pargal and Wheeler1996) argued that in local economies, large plants are more visible and therefore more susceptible to pressure for clean-up. In a contemporary variant, Delmas and Montes-Sancho (Reference Delmas and Montes-Sancho2010) argued that ‘big players’ – or the largest few firms in each market – were more visible and therefore more likely to be early entrants into a voluntary environmental agreement.
These explanations for a positive relationship between firm size and proactive corporate environmentalism are based on symbolic performance and the potential effect on brand name and corporate reputation of negative environmental information (Konar and Cohen Reference Konar and Cohen1997). The search for legitimacy leads firms to abide by the rules, regulations and norms of the field of which they are a part (Dacin et al. Reference Dacin, Oliver and Roy2007) and to shape their communications so as to demonstrate symbolic performance (Fuller and Tian Reference Fuller and Tian2006). Large firms seek to gain and maintain their symbolic performance by performing well in third-party environmental rankings, making positive environmental statements and being awarded environmental certifications (e.g., ISO 14001).
Large firms also are expected to be more proactive in corporate environmentalism because they may have more resources to devote to environmental issues (Russo and Fouts Reference Russo and Fouts1997; Sharma and Henriques Reference Sharma and Henriques2005); they can draw on a broader range of relevant capabilities (Zhu et al. Reference Zhu, Sarkis and Lai2007); or they simply benefit from economies of scale in environmental investments (Gray and Deily Reference Gray and Deily1996; Hettige et al. Reference Hettige, Huq, Pargal and Wheeler1996). In larger firms, the investments required to develop symbolic performance in the environmental domain – whether based on changed business practices, new technologies or reporting on environmental issues in a new way – can be distributed across many activities. Larger firms thus are more likely to invest in symbolic performance through proactive corporate environmentalism or, stated differently:
Hypothesis 5.1: There is an overall positive relationship between firm size and proactive corporate environmentalism.
Symbolic gaps
As described previously, symbolic gaps arise from the difference between symbolic and substantive actions on a given issue, operationalised here as between stated intent on environmental issues and either the implementation of particular environmental initiatives or the mitigation of actual environmental impacts. The typical explanation for symbolic gaps is decoupling, in which new practices are adopted only ceremonially without corresponding action. Large firms face complex and conflicting demands from a range of stakeholders on environmental issues, leading them to not only be more likely to respond to institutional pressures but also to increasingly codify their responses (Meyer Reference Meyer, Greenwood, Oliver, Suddaby and Sahlin2008; Meyer and Rowan Reference Meyer and Rowan1977). Among firms that decide to adopt a new practice, the larger and more visible firms may be more inclined to conform symbolically to gain legitimacy but disinclined to change practices inside the firm (Westphal et al. Reference Westphal, Gulati and Shortell1997). Within the broader economy of esteem, large firms have lower audience-seeking costs and generally invest more in symbolic activities (see Chapter 4). Large firms cope with uncertainty about the efficiency of new organisational practices by decoupling and maintaining a symbolic gap.
Researchers have begun to focus attention on the potential for symbolic gaps in a range of corporate environmentalism contexts. For example, there is an apparent gap among the acquisition of the ISO 14001 certification symbol, implementation of the environmental management standard and actual environmental performance (Boiral Reference Boiral2007; Yin and Schmeidler Reference Yin and Schmeidler2009). Firms that join voluntary environmental programmes have been shown to be no more likely – and, in some cases, they may be less likely – to improve their environmental performance than non-members (Delmas and Montes-Sancho Reference Delmas and Montes-Sancho2010; Rivera and de Leon Reference Rivera and de Leon2004). Firms use impression management techniques to appear proactive on their website and in their annual reports to influence social evaluations, which may diverge from the substance of their actions (Bansal and Kistruck Reference Bansal and Kistruck2006; Laufer Reference Laufer2003). A core finding in this literature is that the motivation to be legitimate apparently affects the symbols of corporate environmentalism more than the substance of implementation or impact (Rhee and Lee Reference Rhee and Lee2003).
Thus, even if firm size is positively related with proactive corporate environmentalism, as captured by stated intent (see A in Figure 5.1), there is not necessarily a relationship between firm size and either the implementation of environmental initiatives (B) or the mitigation of environmental impact (C). I use decoupling to argue that large firms exhibit a symbolic gap, wherein the relationship between size and environmental proactivity is stronger for intended corporate environmentalism statements than for more action-oriented dimensions of corporate environmentalism (i.e., A > B and A > C). Thus:
Symbolic performance across fields
Symbolic performance is valued differently across social systems or fields (Heugens and Lander Reference Heugens and Lander2009). To the extent that large firms attempt to gain and maintain symbolic performance through acceptable environmental strategies, we can expect the intensity of their environmental efforts to vary across fields. The question then becomes when, or in which fields, are large firms more or less likely to adopt proactive environmental strategies.
Visible firms experience a higher level of scrutiny from external constituents, which is one reason why visibility is such an important antecedent of organisational behavior (Delmas and Montes-Sancho Reference Delmas and Montes-Sancho2010; Fiss and Zajac Reference Fiss and Zajac2006). Signalling is most likely when there are active constituents such as third-party rating agencies, ENGOs, well-resourced regulatory agencies and activist shareholders to receive and interpret it. As discussed in Chapter 4, these actors play a key role in sharing information about green solutions. Monitoring is an important complementary asset that can increase the benefits of adopting green solutions to the next incremental adopter. Active stakeholders can directly share learning about green solutions and can underpin network externalities in diffusing green initiatives. Therefore, we expect more attempts to gain and maintain symbolic performance when individuals pay attention to corporate actions. Large, visible firms are more likely to seek and maintain symbolic performance on environmental issues where the potential for monitoring their behavior is high. This leads to the next hypothesis:
Hypothesis 5.3: The positive relationship between firm size and proactive corporate environmentalism is stronger in closely monitored fields.
Symbolic gaps across fields
Firms may gain symbolic performance through the adoption of symbols and language that are recognised and valued by community members. However, the value of this symbolic performance is maintained only as long as community members misrecognise the symbols – and the hierarchical system that produced them – as substance. The ability of large firms to maintain a symbolic gap – wherein the relationship between size and environmental proactivity is stronger for intended corporate environmentalism statements than for environmental impact mitigation (i.e., A > C in Figure 5.1) as hypothesised in H5.2 – will vary across fields. Because the benefits from symbolic gaps are threatened if the discrepancy is exposed, the question becomes: In which fields are symbolic gaps most likely to persist?
The most common answer from the decoupling literature is that symbolic gaps are more likely when there is widespread adoption of a new organisational practice as the new practice gains social legitimacy (Tolbert and Zucker Reference Tolbert and Zucker1983). Other explanations include the extent to which firms are connected with external networks and coalitions (Westphal and Zajac Reference Westphal and Zajac2001), the stringency of stakeholder demands (Oliver Reference Oliver1991), and other organisational-level variables (Rhee and Lee Reference Rhee and Lee2003; Stevens et al. Reference Stevens, Steensma, Harrison and Cochran2005). As responses to institutional pressures become standardised, it is increasingly likely that firms can disconnect their words and actions (Brunsson Reference Brunsson, Brunsson and Jacobsson2000; Jiang and Bansal Reference Jiang and Bansal2003). According to this argument, symbolic gaps are most likely to persist in highly monitored fields because large firms cope with complex and conflicting institutional pressures by codifying their environmental responses.
A more power-based approach implies that symbolic gaps between words and actions are most likely when dominant players have the most power to shape the game. The same forces that drive decoupling – that is, contested environmental expectations of firms and uncertainties as to how to address them – are also likely to encourage misrecognition of corporate authority to define appropriate environmental strategies. Large firms are often the high-status, dominant players who seek to develop early voluntary agreement rules, environmental regulations and expectations (Delmas and Montes-Sancho Reference Delmas and Montes-Sancho2010; Hoffman Reference Hoffman1999). Given the potentially wide power disparity between large firms and their stakeholders, including regulatory agencies in some cases, there is strong potential for large firms to influence the rules of the new corporate environmentalism game. Large firms also have the resources to enable them to become politically engaged (Schuler and Rehbein Reference Schuler and Rehbein1997). Large, dominant firms ‘have the capacity to set the tempo of transformation’ (Bourdieu Reference Bourdieu2005: 202). This can extend beyond political lobbying to influencing which environmental problems are discussed in the public domain and the range of acceptable environmental solutions. These symbols are misrecognised as legitimate in a system that privileges the economic or corporate decision criterion (Ozbilgin and Tatli Reference Ozbilgin and Tatli2005).
Large firms can maintain symbolic gaps because of their high cultural literacy (Boiral Reference Boiral2007). They benefit from social prestige, mutual forbearance and limited rivalry (Mas-Ruiz and Ruiz-Moreno Reference Mas-Ruiz and Ruiz-Moreno2011), and they have the power to influence the accepted symbols of corporate environmentalism (Banerjee Reference Banerjee2008). Large firms developed the ‘beyond compliance’ language in environmental policies, build and maintain voluntary codes of conduct, and often control the entry rules for industry environmental certifications and voluntary agreements (Hoffman Reference Hoffman1999; Rivera and de Leon Reference Rivera and de Leon2004). This effect can be reinforced as leaders in large firms interact with peers within the same high-status group, making them more likely to support one another’s strategic positions (Mas-Ruiz and Ruiz-Moreno Reference Mas-Ruiz and Ruiz-Moreno2011). It is also strengthened because prestigious firms face relatively lower audience-seeking activity costs and invest more in symbolic green solutions, reinforcing their higher esteem.
As Oakes et al. (Reference Oakes, Townley and Cooper1998) noted, the most effective forms of power are those that are associated with little or no visible conflict. Thus, in less-monitored or less-mature fields, the large, high-status firms have higher power over the language and symbols of corporate environmentalism and can invent the grammar of their environmental response. Also, in less-mature heterogeneous fields, resources are dispersed – leading not only to symbolic bargaining to cope with conflicting demands but also to a wide power disparity between firms with high and low status. There is strong potential for high-status firms to influence the rules of the emerging environmental game in less-mature fields because there is an absence of strong monitoring, and stakeholders are actively complicit in accepting corporate solutions to environmental problems as legitimate. High-status firms can wield more institutional control through systemic power in less-monitored fields (Lawrence Reference Lawrence, Greenwood, Oliver and Sahlin-Anderson2008). In more-monitored fields, monitoring such as shareholder actions and threatened regulatory attention on one firm can influence others in the field to improve their environmental performance (Reid and Toffel Reference Reid and Toffel2009).
The potential for large firms to manage the rules of the game and to maintain a symbolic gap is highest in less-monitored contexts, in which active complicity and misrecognition are high. Thus, we should expect the symbolic gap (i.e., A–C in Figure 5.1) to vary systematically, as follows:
Methods
I tested the hypotheses by meta-analysing existing research on the predictors of proactive corporate environmentalism. Testing all four hypotheses within one empirical study would place considerable demands on data collection: I would have needed to gather data on environmental stated intent, initiative implementation and impact mitigation on a large sample of companies across countries, industries and organisational locations during a sufficiently long period to capture issue maturity evolution. Instead, I followed Heugens and Lander’s (Reference Heugens and Lander2009) approach of conducting an across-field meta-analysis. To test H5.1 and H5.2, a research assistant2 and I gathered all of the available evidence from the empirical corporate environmentalism studies and estimated the aggregate associations of firm size with different dimensions of proactive corporate environmentalism. We then analysed whether there were systematic differences in effect sizes across different data samples, which we assumed to be proxies for different types of fields (i.e., testing H5.3 and H5.4). Meta-analysis enabled us to deploy data from a wide range of fields because it is ‘the statistical analysis of the summary findings of many empirical studies’ (Glass and McGaw Reference Glass and McGaw1981: 21).
The criterion for inclusion in our study was ‘any empirical paper which reported on the relationship between firm size and any aspect of proactive corporate environmentalism’. As I described previously, we used a broad interpretation of corporate environmentalism; therefore, our criterion yielded papers focusing on a range of green solutions including investing in specific initiatives, introducing general corporate environmental programmes, being adjudged to have a proactive corporate environmentalism according to third-party ratings, and reducing emissions. Our criterion did not require the firm size and corporate environmentalism relationship to be the main focus of the study; rather, included papers merely reported on data pertaining to that focal relationship (Hunter and Schmidt Reference Hunter and Schmidt1982).
Meta-analytic techniques
Literature search and inclusion in analysis. We searched the Web of Science, ANBAR, Emerald, ABI Inform Global, JStor, Business Source Premier and Wiley Inter-Science journal indices for articles using the search terms environmental, environmental performance, environmental responsiveness, green and size. We then tried to partially overcome the ‘file-drawer problem’ (Glass and McGaw Reference Glass and McGaw1981) in which only positive results tend to be published. We sent emails to networks of scholars interested in environmental issues, asking if they had any published or unpublished work that might fit our criterion for inclusion. We sent these requests to the ONE, Social Issues in Management, Business Policy and Strategy, and Organisation and Management Theory listservs of the Academy of Management. In addition, we sent individual emails to all authors identified by our initial searches to solicit further potential articles. Finally, to identify any other conference papers, doctoral theses, or working papers, we searched the Social Science Research Network Working Paper database, author webpages, and the ProQuest Dissertations and Thesis database.
Our entire literature search process identified more than three hundred abstracts of studies that seemed to contain appropriate information. We required three pieces of information to evaluate each study for inclusion in our analysis: (1) some measure of corporate environmentalism, (2) some measure of firm size, and (3) a Pearson correlation coefficient between the two measures – or data that would allow us to calculate such a correlation (Hunter and Schmidt 1982). Corporate environmentalism and firm size often were seen as joint predictors of another variable (usually financial performance). Researchers typically did not specifically examine the relationship between corporate environmentalism and size (but see Darnall et al. Reference Darnall, Henriques and Sadorsky2010 for an exception); instead, size was included as a control variable as a conventional empirical research practice (see, e.g., Branzei et al. Reference Branzei, Ursacki-Bryant, Vertinsky and Zhang2004, Delmas and Toffel Reference Delmas and Toffel2008, and Phillipe and Durand 2011). The primary reasons for eliminating studies from our analysis were a lack of descriptive statistics and definitions that fell outside the criterion for inclusion in this study. Table 5.1 lists the studies included in our meta-analysis. The effect size (i.e., Pearson correlation) reported in the studies varied from –0.22 to 0.46.3
Table 5.1 Studies included in the meta-analysis
Coding. We read and extracted data from the papers using several dimensions: basic identification details; study characteristics (i.e., main aim of study, base discipline, motivation for including size, and measure of corporate environmentalism); methodological characteristics (i.e., date of data collection, sample details, operationalisations used, analysis techniques used, and reliability of measures); and relevant results.
We found a wide range of proactive corporate environmentalism measures, including perceptual measures of environmental performance (see, e.g., Sharma Reference Sharma2000); the existence of corporate environmental policies (see, e.g., Henriques and Sadorsky Reference Henriques and Sadorsky1996 and Russo and Fouts Reference Russo and Fouts1997); the implementation of environmental initiatives (see, e.g., Bowen Reference Bowen2000b, Klassen Reference Klassen2000, and Theyel Reference Theyel2000); voluntary participation in environmental schemes such as ISO (see, e.g., Potoski and Prakash Reference Potoski and Prakash2005); and decreasing toxic emissions (see, e.g., King and Lenox Reference King and Lenox2000). My research assistant and I independently coded whether the corporate environmentalism measure was based on stated intent, implementation or impact according to the definitions outlined in the previous discussion about Figure 5.1 (see Table 5.1). We classified this information based on the methods used in a paper, often referring to questionnaires or the precise wording of third-party rankings. In some cases, studies incorporated more than one type of corporate environmentalism measure (see, e.g., Branzei et al. Reference Branzei, Ursacki-Bryant, Vertinsky and Zhang2004 and Russo and Harrison Reference Russo and Harrison2005), in which case we separately recorded effect sizes based on the different measurement types. In other cases, several published studies seemed to be based on the same underlying dataset (see, e.g., Darnall, Henriques, and Sadorsky Reference Darnall, Henriques and Sadorsky2008 and Darnall, Seol, and Sarkis Reference Darnall, Seol and Sarkis2009), in which case we recorded the relevant effect size only once.
We were particularly concerned with asking empirically whether the estimated effect size between firm size and corporate environmentalism differs across different datasets (i.e., fields) – particularly fields that are relatively more or less monitored. We distinguished highly-monitored from less-monitored fields based on national institutional context, organisational level, issue centrality and issue maturity. The two assessors coded each study separately, making judgements on the field and issue context, as follows.
National institutional context. Corporate environmentalism is less monitored in developing than in developed countries. In developing countries, governments frequently do not give clear direction to environmental regulators, regulatory agencies lack the capacity to enforce and monitor compliance with environmental regulations, broader stakeholder pressures for environmental improvements are perceived to be weak, and local industry associations encourage less environmental proactivity than in developed countries (Rivera Reference Rivera2010; Shah and Rivera Reference Shah and Rivera2013). Auditing and certification practices can be weaker in developing countries (Christmann and Taylor Reference Christmann and Taylor2006). For both formal monitoring by regulatory agencies and informal monitoring by environmental groups, the media and professional associations are weaker in less-developed national institutional contexts (Hettige et al. Reference Hettige, Huq, Pargal and Wheeler1996). We therefore can expect a bigger gap between the rhetoric and reality of environmental strategy in developing countries (Rhee and Lee Reference Rhee and Lee2003). We coded the national institutional context of each effect size using the World Bank’s definition of developed and developing countries.
Organisational level. Corporate environmentalism is less monitored at the subsidiary level than at the corporate level. Most environmental reporting and monitoring, including that required for socially responsible investing, is conducted at the corporate rather than the subsidiary level (Laufer Reference Laufer2003). The responsibility for developing, maintaining and protecting symbolic performance through environmental reports and industry-club membership related to environmental issues usually resides at the corporate level rather than at each subsidiary. We coded the organisational location of each effect size relative to whether the sample was based exclusively on entire firms (i.e., the corporate level) or on plants/subsidiaries of firms (i.e., the subsidiary level) (Bowen Reference Bowen2000a).
Issue centrality. Corporate environmentalism is monitored more in fields in which environmental issues exhibit high centrality, such as the chemicals and resource-extraction industries, than in other industries. Burke and Logsdon (Reference Burke and Logsdon1996) defined centrality as a measure of the closeness of fit between a corporate social policy or programme and the firm’s mission and objectives. When environmental issues are highly central to a firm’s activities, we can expect more stakeholder and regulatory attention to environmental concerns (Chatterji and Toffel Reference Chatterji and Toffel2010; Reid and Toffel Reference Reid and Toffel2009) and more attempts to protect a firm’s environmental reputation (Halme and Huse Reference Halme and Huse1997; Phillipe and Durand 2011). It is difficult to achieve legitimacy through purely symbolic actions in stronger institutional fields such as highly polluting industries (Berrone and Gomez-Mejia Reference Berrone, Gelabert and Fosfuri2009). Furthermore, service industries with low issue centrality are ‘informationally opaque’ in the sense that their responsiveness is difficult to monitor (Ramchander, Schwebach, and Staking Reference Ramchander, Schwebach and Staking2011). We coded the issue centrality of each effect size relative to whether the sample was based on a high- or low-pollution industry (Halme and Huse Reference Halme and Huse1997).
Issue maturity. Corporate environmentalism is more stringently monitored by experienced stakeholders in mature fields. A firm’s environmental strategies evolve and change with altering coercive, mimetic and normative pressures (Bansal Reference Bansal2005; Hoffman Reference Hoffman1999). Industry norms and values evolve over time, changing what are considered acceptable activities and bringing in new incentives for appropriate behaviour (Delmas, Russo, and Montes-Sancho Reference Delmas, Russo and Montes-Sancho2007). As environmental issues mature as a legitimate concern for firms, environmental performance becomes more salient and increasing numbers of firms are drawn into the corporate environmentalism decision domain (Bansal Reference Bansal2005). Furthermore, the extent of monitoring by third-party rating agencies (e.g., KLD Research and Analytics, Inc.) expands over time, and firms that were poorly rated in the past attempt to improve their environmental impact (Chatterji and Toffel Reference Chatterji and Toffel2010). We coded each effect size for issue maturity relative to whether the sample was based on ‘early’ data collected before 2000 or ‘late’ data collected after 2002. This cut-off date was somewhat arbitrary but we used it because there were no effect sizes for 2001 (and, hence, a convenient gap between studies). Experimenting with other dates did not significantly change the results.
We also coded for three study artefacts: (1) the measure of firm size, (2) whether a study was published or unpublished, and (3) whether the study relied on a large sample (defined as more than one standard deviation above the mean sample size). The studies included in the meta-analysis spanned a range of both developing and developed countries (e.g., the United States, Canada, China, Brazil, Bulgaria, Norway, and South Korea) as well as industries (e.g., forestry, chemical, petrochemical, electricity, and manufacturing); however, not all studies focused on a single country or industry. We coded studies that had only unambiguous information on moderators rather than forcing all studies into a subgroup.
Meta-analysis procedure. We conducted the meta-analysis in accordance with guidelines provided by Hunter and Schmidt (Reference Hunter and Schmidt1990) and Lipsey and Wilson (Reference Lipsey and Wilson2001). This is an effect-size meta-analysis technique that corrects for sampling and measurement error. The main statistics required to undertake the analysis are the Pearson correlation coefficient of the relationship (r) and the number in the sample (n). Our results are based on 70 effect sizes gathered from within 45 studies, with a total sample size of more than 85,000.
We calculated the 95 per cent confidence and credibility intervals for each analysis. Confidence intervals are a familiar method for estimating the extent to which sampling error remains in the weighted effect size, and they can be interpreted using significance-testing methods. Credibility intervals, conversely, indicate whether moderators are operating, which in turn indicates whether the population should be categorised into subgroups (Whitener Reference Whitener1990). We followed Hunter and Schmidt (Reference Hunter and Schmidt2004), who argued that credibility intervals are more important than confidence intervals in meta-analysis, and we based our inferences primarily on credibility intervals.
We conducted moderator analyses in three basic ways. First, we calculated the Q statistic, which is an index of how much variability there is within the results (Hunter and Schmidt 1990; Sagie and Koslowsky Reference Sagie and Koslowsky1993). Significant Q statistics indicate the possible presence of moderators. We then developed meta-analytic summary data for each moderator subgroup (see Table 5.2) and examined the credibility intervals. Wide credibility intervals that include zero indicated that moderators are likely, which impacted the effect-size estimate (Sagie and Koslowsky Reference Sagie and Koslowsky1993; Whitener Reference Whitener1990). However, a more precise method of statistical analysis is weighted least squares multiple regression (Steel and Kammeyer-Mueller Reference Steel and Kammeyer-Mueller2002). Consequently, we weighted each moderator analysis by sample size in a two-stage multiple regression analysis. In the first stage, we controlled for study artefacts, field context and the measure of corporate environmentalism used. In the second stage, we entered the appropriate interaction effects, determining if they explained significant incremental variance.
Results
Table 5.2 shows the overall findings of the meta-analysis by categorising each effect size according to whether the data related to corporate environmentalism as stated intent, initiative implementation or impact mitigation. The overall relationship between firm size and environmental strategy, which includes all of the effect sizes available, is weak but positive (r = 0.02) and significantly differs from zero at the 95 per cent confidence interval (0.01–0.04).
Table 5.2 Meta-analysis overall results
K = number of effect sizes.
N = total sample size across all included effect sizes.
Mean r = sample‐size weighted Pearson correlation coefficient across included effect sizes.
Table 5.2 shows that studies were distributed across the three different dimensions of corporate environmentalism (i.e., K = 16 for intent, K = 28 for implementation, and K = 26 for impact), demonstrating that there is adequate coverage of effect-size data across the three dimensions. Of note, however, is the average sample size within each type of study, which ranges from 187 for implementation-based effect sizes to 708 for intent-based studies to a high of 2,700 for impact-based studies. This range reflects data availability for the different dimensions – particularly the dominance of impact studies based on Toxic Release Inventory data from the US Environmental Protection Agency (EPA). Based on the confidence intervals, we found support for H5.1’s positive relationship between firm size and corporate environmentalism for the dimensions of stated intent (r = 0.13) and implementation (r = 0.10) but not for impact (r = −0.01). We also found support for H5.2, in that the relationship between firm size and proactive stated intent (r = 0.13) is significantly stronger than between firm size and mitigating environmental impact (r = −0.01).
However, the more interesting result was the credibility intervals, which are wide and include zero for all three dimensions; this indicates that the effect size between firm size and each dimension of corporate environmentalism may vary from negative to strongly positive. This variability provided the impetus to test our hypotheses on how the effect size between firm size and corporate environmentalism is influenced by the strength of monitoring in different fields.
Table 5.3 shows the results of our weighted least squares analysis of the 70 effect sizes, each weighted according to sample size. Model 4 shows significantly smaller effect sizes between firm size and corporate environmentalism for studies measuring impact mitigation than in the base case of stated intent. This adds further support for H5.2, which predicted a stronger effect of firm size on symbolic stated intent than on substantive impact.
H5.3 predicted that the positive relationship between firm size and proactive corporate environmentalism is stronger in closely monitored fields. The field-context variables in Table 5.3 show support across the models for smaller effect sizes in studies undertaken in developing compared with developed countries as well as for studies based at the subsidiary rather than the corporate level. H5.3 is not supported for high- compared with low-pollution industries or for later studies compared with earlier studies. Thus, H5.3 seems to be supported only when monitoring strength is based on field-level characteristics (i.e., national institutional context and organisational location) and not for issue-level characteristics (i.e., issue centrality and maturity). This was a surprising finding that requires further exploration.
The interaction effects in Table 5.3 were designed to test H5.4: that is, the symbolic gap between stated intent and either initiative implementation or mitigation impact is larger in less-monitored fields. Consistent with our finding on H5.2, we did not find systematic evidence for symbolic gaps for the implementation dimension (i.e., no significant coefficients on implementation interaction effects). Also consistent with our findings on H5.3, we did not find evidence for symbolic gaps based on issue-level characteristics (i.e., no significant coefficients on issue centrality or maturity interaction effects). However, we did find support for H5.4 when symbolic gaps were based on the impact dimension, particularly at the subsidiary level. Thus, symbolic gaps between stated intent and impact mitigation are wider at the less-monitored subsidiary level.
Discussion and implications
The results show broad support for our predictions that large firms seek symbolic performance through proactive corporate environmentalism. Our results showed that large firms exhibit a symbolic gap, wherein the relationship between size and corporate environmentalism is stronger for intended statements than for substantive environmental impact mitigation. The extent to which large firms seek symbolic performance and maintain symbolic gaps is moderated by the extent of monitoring within fields.
Our findings confirmed decoupling expectations of the existence of symbolic gaps but extended contemporary explanations of when symbolic gaps persist. This analysis advances emerging findings about ceremonial adoption in environmental strategies (see, e.g., Boiral Reference Boiral2007 and Laufer Reference Laufer2003) and when organisations decouple (cf. Boxenbaum and Jonsson Reference Boxenbaum, Jonsson, Greenwood, Oliver and Sahlin-Anderson2008). The meta-analysis confirms that it does matter whether we measure corporate environmentalism along intent-, implementation- or impact-based dimensions.
This study also made three extensions to the basic decoupling argument. First, the findings indicate that the gap between symbolic and substantive responses is moderated by firm size: that is, large firms are more likely than small firms to state a symbolic responsive intent but are no more likely to substantively mitigate their impact. This effect was included as a control variable in some studies (see, e.g., Westphal et al. Reference Westphal, Gulati and Shortell1997) but so far has been held as a widespread assumption rather than a robust empirical finding. The finding provides initial support for one of the core arguments of this book: larger, more powerful firms are more likely to exert their discretionary power to drive a wedge between symbolic and substantive green solutions. This finding also suggests that the generation of substantive and symbolic green solutions is driven by different factors. It turns out that firm size is a determinant of only symbolic and not substantive green solutions.
Second, this analysis expanded the more typical two categories of ‘espoused’ and ‘implemented’ (Fiss and Zajac Reference Fiss and Zajac2006), ‘rhetoric’ and ‘reality’ (Rhee and Lee Reference Rhee and Lee2003), or ‘adoption’ and ‘implementation’ (Kennedy and Fiss Reference Kennedy and Fiss2009) in responses to institutional pressures to the three categories: intent, implementation and impact.4 The effect sizes between firm size and implementing green solutions differed significantly from substantive impact mitigation but did not differ from stated strategic intent. This finding implies that the likelihood of large firms responding through initiative implementation is closer to symbolic intent than to actual substantive impact. Thus, the findings add weight to recent decoupling research that reminds us that the process of implementation itself may be symbolic (Boiral Reference Boiral2007; Christmann and Taylor Reference Christmann and Taylor2006) and not the end point in the institutionalisation process (cf. Fiss and Zajac Reference Fiss and Zajac2006 and Kennedy and Fiss Reference Kennedy and Fiss2009). These findings also reinforce my contention in Chapter 4 that it may be easier for firms to move the private cost curve for symbolic rather than substantive green solutions and that this may be a natural outcome of profit-maximising strategic corporate environmentalism (see Figure 4.2). We need a more nuanced understanding of proactive corporate environmentalism: we must focus more attention not only on whether firms adopt ‘beyond compliance’ strategies but also on whether and how they are implementing them symbolically (Bromley and Powell Reference Bromley and Powell2012; Delmas and Montes-Sancho Reference Delmas and Montes-Sancho2010; Philippe and Durand Reference Philippe and Durand2011).
Third, this analysis extended conventional decoupling arguments by giving more attention to power dynamics within fields. For example, Lawrence (Reference Lawrence, Greenwood, Oliver and Sahlin-Anderson2008) encouraged more consideration of ‘institutional side effects’. Decoupling proactive corporate environmentalism implies (1) less environmental improvement than might be expected based on stated corporate intentions, or (2) a negative side effect on the natural environment and vulnerable people who might rely on threatened natural resources. Using a power-based logic questions who has the systemic power to gain from decoupling and how this might explain persistent symbolic gaps. We know that less-monitored firms with greater discretionary power are more likely to disconnect proactive strategic intent from implementation actions (Philippe and Durand Reference Philippe and Durand2011) and that decoupling can protect local power relationships (Bromley and Powell Reference Bromley and Powell2012). Large firms have the misrecognised authority to define acceptable responsiveness as proactive environmental intent rather than impact. Large firms are more likely to adopt symbolic green solutions, setting the norms for other less-prestigious firms to follow. Explanations for symbolic gaps change as we focus more attention on the misrecognition of the symbolic power and status of large firms. I present a deeper analysis of this phenomenon in Chapter 6.
Standard institutional arguments suggest that early adopters do so for efficiency reasons, whereas late adopters are more likely to maintain symbolic gaps as they try to gain social legitimacy (Tolbert and Zucker Reference Tolbert and Zucker1983). Kennedy and Fiss (Reference Kennedy and Fiss2009) countered this logic by showing that the lines between efficiency and legitimacy logics are not so clear. I took a different approach and linked decoupling at the firm level with power at the field level. I argued that not only are large firms more likely to adopt symbolic green solutions to gain legitimacy but also that they are the high-status, dominant players in developing institutional responses and expectations. Large firms seek legitimacy by gaining and maintaining symbolic performance; they also can use their status, credibility and connections to shape the rules of the game and protect symbolic gaps between their intentions and substantive performance, particularly in less-mature monitoring contexts. Thus, in early periods, large firms have the latitude to design the symbols of their response. We saw in Chapter 2 that in the early days of environmental awareness, firms such as Chevron gained reputational benefits from environmental advertising campaigns that today would be dismissed as greenwashing. In later periods, as monitoring intensifies, the symbolic gap between the propensity of large firms to state proactive intentions and their impacts is less socially sustainable. The more monitored the field, the more that large visible firms are held accountable for their actions and not their words and the less valuable is symbolic performance. This might explain the intriguing strategy among large firms of addressing substantive institutional pressures without communicating about these new organisational practices (Fiss and Zajac Reference Fiss and Zajac2006). This effect is more likely in highly monitored fields. However, as discussed in the shift from greenwashing to symbolic corporate environmentalism (see Chapter 2), closing obvious gaps is relatively easy compared with the more nuanced task of identifying and exposing costly, misrecognised symbolic aspects of everyday activities.
The meta-analytic research design enabled me to include empirical data across fields by accumulating effect sizes from many different types of datasets. This enabled the exploration of symbolic performance and symbolic gaps at a relatively ignored level of analysis: that is, across fields. Institutional theorists typically conduct agentic studies at the individual or organisational level and non-agentic studies at the field, industry or national level of analysis (Boxenbaum and Jonsonn Reference Boxenbaum, Jonsson, Greenwood, Oliver and Sahlin-Anderson2008). This chapter shows an agentic study at the cross-field level by looking at effect sizes across many organisational fields. An advantage of this approach is that meta-analytic techniques can be deployed as a method to test previously untested hypotheses rather than only to aggregate other research studies (Heugens and Lander Reference Heugens and Lander2009). This raised the level of analysis from the firm-level preferred by conventional corporate environmental researchers to the field level preferred by the critical view.
However, this method is based on the key assumption that each effect size is indeed drawn from an identifiable field. Clearly, this method departs from contemporary definitions of fields as defined by issues rather than by industry classification codes (Wooten and Hoffman Reference Wooten, Hoffman, Greenwood, Oliver and Sahlin-Anderson2008). Each effect size in the meta-analysis did not include all ‘organisations that partake of a common meaning system and whose participants interact more frequently and fatefully with one another than with actors outside the field’ (Scott Reference Scott1995: 56) as would be required in domain-based definitions of fields. The samples within the meta-analysis were drawn from populations of firms and did not typically include other actors within a given field, such as NGOs, professional organisations, regulators and the media. Instead, I adopted a more relational conception of a field (Wooten and Hoffman Reference Wooten, Hoffman, Greenwood, Oliver and Sahlin-Anderson2008) and focused on the relative status of organisations within a given social population (Bourdieu and Passeron Reference Bourdieu and Passeron1977). Schuler (Reference Schuler1996) showed that it may be a firm’s relative rather than absolute size that predicts its political activity. Meta-analysis was an appropriate empirical technique for these across-field hypotheses because in any given sample, larger firms tend to have higher social evaluations within the sampled firms (Fombrun and Shanley 1990). This permits the examination of how firm size drives corporate environmentalism across different contexts.
The findings also suggest contingencies related to corporate environmentalism researchers’ conventional use of size as a control variable in predicting proactive corporate environmentalism (cf. Branzei et al. Reference Branzei, Ursacki-Bryant, Vertinsky and Zhang2004 and Delmas and Toffel Reference Delmas and Toffel2008). Given the paucity of large-sample corporate environmentalism and performance data, it is understandable why researchers routinely include firm size as a control variable. However, this meta-analysis shows that including firm size in all types of corporate environmentalism studies is misleading. This evidence suggests that firm size predicts corporate environmentalism only for intent- and implementation-based dimensions of greening. Future research should control for size only in studies of predictors of symbolic green solutions. Firm size may be irrelevant for predicting substantive green solutions that, after all, are driven by the costs and benefits of the green solutions available (see Chapter 3).
Empirical summary
My answer to the first empirical question posed in this chapter – Does separating symbolic from substantive performance provide different answers about the drivers of corporate environmentalism in large powerful firms? – is ‘yes’. The results suggest that larger firms are more likely to attempt to gain and maintain symbolic performance by announcing proactive corporate environmentalism intent in an attempt to gain legitimacy, but they are not more likely to achieve substantive environmental improvements. I also asked: When can symbolic gaps persist between the likelihood that large firms will adopt corporate environmentalism through proactive strategic intent, implementation or impact? The results indicate support for the power of large firms to shape the rules of the game in less-monitored fields. Evidence for such persistent symbolic gaps is a cautionary tale for stakeholders seeking to influence corporate environmentalism who might misinterpret symbolic green solutions as substantive environmental improvement. I argue that large firms can separate their words and actions on environmental issues based on a decoupling logic at the firm level and systemic power at the field level. Large firms may attempt to gain social legitimacy through symbolic performance and decoupling. However, symbolic gaps can persist only when large firms have sufficient systemic power for field members not to question the difference between symbol and substance.
Corporate environmental strategy and the social energy penalty
This chapter focuses on the role of powerful firms in driving symbolic performance and symbolic gaps in proactive environmental strategies. Although the analysis accumulated evidence across many different research studies on one of the drivers of symbolic gaps, I did not address what the consequences of these gaps might be. What do these findings tell us about the social costs and benefits of corporate environmental strategy?
Figure 5.2 illustrates the difference between more- and less-monitored fields in the social energy penalty model introduced in Chapter 4. In less-monitored fields, the social benefit curve is smoother and reaches its limit at a lower level. There is less potential for network externalities because less-monitored fields lack the complementary assets of stakeholder knowledge and awareness that drive the need for standardised self-regulatory schemes. Cumulative adoption does not generate substantial additional benefits to adopting the self-regulatory practice unless there is an audience to legitimise it. Furthermore, there is less force exerted by regulatory agencies, and less intense NGO or media attention driving institutionalisation in less-monitored fields. Less attention means that firms can gain less benefit from sheltering within self-regulatory schemes. Lower monitoring weakens the demand for green solutions and decelerates diffusion.
Figure 5.2 Impact of monitoring within fields on the social energy penalty.
The net result of a smoother and lower social benefit curve is that the social energy penalty outweighs the potential benefits of standardising green solutions at a relatively lower level of cumulative output (i.e., XLM instead of XHM in Figure 5.2). All else being equal, the social energy penalty is likely to be more of a problem in less-monitored than in more-monitored institutional fields. It is worth noting that this is the case even when firms do not deliberately exert their power over voluntary self-regulation schemes (i.e., when the social cost curve remains the same). Even if firms do not exercise their discretionary power to limit the supply of greening solutions, the point at which the social energy penalty outweighs social benefits is farther to the left for less-monitored fields. Thus, even advocates from the conventional view of corporate environmentalism should seriously consider the potential for social losses from self-regulation in less-monitored fields. The higher likelihood of a social energy penalty should not be ignored simply because it is an unintended consequence of profit maximisation. Control over green solutions generates symbolic gaps and for large, powerful firms, the best strategic approach may be to exploit them if monitoring is weak.
Advancing the power-based arguments, we might expect the social cost curve to be steeper than shown in Figure 5.2. As discussed in Chapter 4, the steepness of the cost curve is primarily a function of the discretionary behaviour of powerful firms. If large firms strategically exert their systemic power, then the social energy penalty might outweigh social benefit at cumulative adoption levels even farther to the left of XLM. Symbolic gaps are most likely when dominant players have the most power to shape the game. All else being equal, the social energy penalty is likely to be more of a problem when powerful firms choose to exercise that power – regardless of the extent of monitoring within a field.
Thus, the social energy penalty can become harmful at lower levels of cumulative adoption of voluntary environmental strategies in less-monitored fields and when powerful firms exploit their systemic power – regardless of the state of monitoring in the field. This begs the question of when and how powerful actors can exert this discretionary power. It is to that question that I turn in the next chapter.
