Reputational Imprints: How Public Criticism During Global Crises

REPUTATIONAL IMPRINTS: HOW PUBLIC CRITICISM DURING GLOBAL CRISES

AFFECTS SUSTAINABILITY-DRIVEN INNOVATION

ABSTRACT

Public criticism of firms’ environmental, social, and governance (ESG) practices offers an important source of environmental feedback that can prompt innovation. However, such criticism can also affect organizations’ reputations in ways that constrain innovation, limiting both access to necessary resources for and the potential payoff from such innovation. In this study we examine this tension between public criticism as a source of organizational learning and reputational constraint in the context of global crises. Building on the organizational imprinting literature, we argue that global crises represent pronounced periods of upheaval for firms, during which firms are imprinted by public criticism in ways that over time channel firms’ sustainability–related strategies. We test our hypotheses using a panel data set of 4,738 companies across a 12–year period beginning during the 2007–2009 global financial crisis. Our findings suggest that although the firms initially engaged in similar levels of sustainable practices, firms differed over time in their embrace of innovation in response to ESG incidents depending on how extensively they were criticized during the crisis. Moreover, competition is shown to amplify the resulting virtuous and vicious spirals. Our findings contribute to scholarship on imprinting, corporate reputation, and organizational sustainability.

Keywords: imprinting; public criticism; sustainability; corporate reputation; ESG issues

INTRODUCTION

Organizations’ reputations are increasingly informed by their responses to environmental and social sustainability issues (Barnett & King, 2008; Brammer & Pavelin, 2006). In many cases organizations become acutely aware of such reputational effects following third–party sustainability audits which expose incidents wherein the organizations’ environmental, social and governance (ESG) practices are perceived as deficient. In this paper we conceptualize ESG incidents as public scandals derived from both traditional and non–traditional media sources regarding a firm’s unsustainable practices. Such incidents, for instance, include expressed media and public criticism regarding a firm’s negative contributions to a host of social and environmental issues, including climate change, forced labor, corruption, and controversial products and services, among many others. Prior research suggests that ESG incidents represent substantial reputational threats, oftentimes triggering organizations to respond by way of either symbolic or more substantive reforms in order to protect their reputations (Briscoe & Gupta, 2016; Durand, Hawn

& Ioannou, 2019; Grimes, Williams & Zhao, 2019). In effect, we might expect organizations to learn from public criticism. Yet it is also possible that as ESG incidents accumulate, the reputational damage might undermine even the possibility of such sustainability–related improvements by limiting the respective organization’s capacity to acquire the resources needed to pursue those improvements as well as its ability to capitalize on the improvements (Shea &

Hawn, 2019; Waldron, Navis, & Fisher, 2013). Although recent and longstanding work offers important insight into why firms’ responses to social and environmental issues are likely to vary

(Dowell & Muthulingam, 2017; Durand, et al., 2019; Gupta & Briscoe, 2020; Oliver, 1991), this work has yet to explore how times of global crisis, such as the global financial crisis or the Covid-

19 crisis might contribute to such variation. As noted recently by one impact investor, “Investors

2 who follow ESG criteria are looking closely to see how companies have responded to employees, customers, and other stakeholders during the [Covid-19] crisis. These actions will inform the company’s reputation as an employer and of its brand value for many years to come (Barron’s,

2020)”. In other words, global crises may represent pronounced periods of organizational upheaval, during which investors and other stakeholders are increasingly attuned to ESG-related criticism, with potentially lasting ramifications for the firms, their reputations, and thus we argue their capacity to innovate even after the crisis has subsided.

In order to theorize about the lasting significance of public criticism during a global crisis, we draw from the growing body of research on organizational imprinting, which focuses on how

“external environments exert a powerful influence on organizations and their building blocks during periods of organizational transition, upheaval, and instability” (Marquis & Tilcsik, 2013, p.

200). Although many studies of organizational imprinting have focused on the environment’s external influence over firms in the context of those firms’ foundings or births, more recent scholarship argues and offers evidence that other moments of transition (Dowell & Swaminathan,

2006; Noda & Collis, 2001), disasters (Tilcsik & Marquis, 2013) or crises (Almandoz, Lee, &

Marquis, 2017; Dieleman, 2010; Dixon, Meyer, & Day, 2007) represent important sensitive periods, during which firms can be exposed to lasting imprints (Simsek, Fox, & Heavey, 2014).

Building on this research, we examine how global crises increase firms’ sensitivity to their external environments, and we argue that the mechanism by which the environment exerts influence over firms during these periods of crisis is that of public ESG criticism, which then shapes firms’ sustainability-driven innovation in perpetuity. By sustainability-driven innovation we refer specifically to the “integration of ecological and social aspects into products, processes, and organizational structures” (Klewitz & Hansen, 2014; p. 57). Understanding how public criticism

3 regarding ESG incidents intersects with moments of global crises is theoretically important, given growing scholarly attention to the role of activism and social movements as agents of change, pressuring firms to behave in socially responsible ways (Briscoe & Gupta, 2016; den Hond & de

Bakker, 2007; McDonell & King, 2013; King, 2008; King & Soule, 2007; Waldron, Navis,

Aronson, York, & Pacheco, 2019). Yet it is also of contemporary significance given the increased salience of several global crises as well as the increased media-based scrutiny of firms’ actions during these crises.

Our analysis addresses the following question: how does public criticism of firms’ ESG practices during a global crisis enable and constrain those firms’ capacity to engage in sustainability–driven innovation? We introduce and test a conceptual model of how ESG incidents experienced during a period of economic and social crisis perpetually condition firms’ sustainability–driven innovation by way of reputational imprints. We also theorize and test how competitive dynamics further exacerbate these reputational effects. To test this conceptual model, we empirically examine changes in ESG practices across 4,738 unique firms in 76 countries and

62 industries over a 12–year period beginning during the 2007–2009 global financial crisis.

Consistent with our interest in sustainability–driven innovation, ESG practices in this case refer to a company’s policies, programs, and strategies designed to manage its ESG risks and opportunities. Our sample ranges from firms that were not exposed at all to ESG incidents to those which received extensive public criticism across both traditional and social media on a comprehensive set of 28 different ESG issues.

Our findings demonstrate that differences in how firms are criticized during a period of global crisis are likely to imprint upon the firms’ reputations, thereby enabling and constraining sustainability–driven innovation in an ongoing manner following the crisis. At the start of the global financial crisis the two groups of firms exhibited similar ESG practices. Despite such initial 4 similarities in ESG practices, the two groups were exposed to substantial differences in public criticism of those practices during the crisis. We observe that firms which received less extensive

ESG criticism during the crisis demonstrate the ongoing capacity to be positively responsive to and learn from such criticism, increasing their ESG–driven innovation following future ESG incidents. Alternatively, firms which received more extensive ESG criticism during the crisis become negatively responsive to future incidents, decreasing their ESG–driven innovation over time even after the global crisis wanes. These findings thus suggest that such virtuous and vicious reputational spirals were not due to different initial capabilities or other organizationally determined inertial forces, but once again shaped by the imprinting effect of public criticism received during a time of global crisis. Building on these findings, we further demonstrate that competitive pressures from the product market intensify the divergent reputational spirals. Taken together our findings challenge and extend prior research on organizations’ responses to normative pressures, most notably by highlighting the important ways that periods of global crises can serve to both enable and constrain firms’ capacity to learn from public criticism and engage in sustainability-driven innovation.

THEORETICAL MOTIVATION and HYPOTHESES

The Increasing Importance of Public Criticism of Firms’ Sustainability Reputations

Over the recent decade, public awareness and public criticism of corporations’ sustainability practices has risen precipitously, such that it has become an essential feature of those firms’ reputations. Here we draw on prior definitions of a firm’s reputation as “observers’ collective judgments of a corporation based on assessments of the financial, social, and environmental impacts attributed to the corporation over time” (Barnett et al., 2006; page 34;

Barnett & Hoffman, 2008). Whereas an organization’s image is comprised of observers’ general

“impressions” of the firm, reputation refers to the public’s active “judgement” of that firm, based on acute events or compounding and externally–observable chronic organizational actions (Barnett et al., 2006).

Longstanding research on corporate reputation illustrates how positive collective judgments of a firm can serve as one of the greatest strategically–important resources for firms, aiding in their ability to achieve superior financial and social performance and sustained competitive advantage (Barney, 1986; Barney, 1991). For instance, a good reputation can enhance consumer brand image (Ariely, Bracha & Meier, 2009; Brammer & Pavelin, 2006), customer loyalty (Bhattacharya & Sen, 2003; Fombrun & Shanley, 1990), product differentiation

(McWilliams & Siegel, 2001), and process innovation (Eccles, Ioannou & Serafeim, 2014; Porter

& van der Linde, 1995), thereby increasing the firm’s competitiveness and allowing for a premium pricing of its products (Besley & Ghatak, 2005). A number of studies highlight that such reputational benefits may be increasingly prevalent in the context of sustainability–related judgments, wherein a positive sustainability reputation can be used to attract, engage, and retain talent (Turban & Greening, 1997) and even used as a defense against knowledge spillovers

(Flammer & Kacperczyk, 2019).

Perhaps most crucially, to the extent that a firm is perceived as aligning with stakeholders’ expectations, this increases stakeholders’ willingness to collaborate with the firm (Brammer &

Pavelin, 2006). Aligning with audiences’ sustainability–related expectations, for example, has been shown to increase the perception of the firm’s transparency and accountability, thereby mitigating external stakeholders’ perceptions of asymmetric information or risk (Cheng, Ioannou

& Serafeim, 2014; Dhaliwal, Tsang & Yang, 2011). This can, in turn, reduce firms’ resource constraints by facilitating access to capital and improving negotiation with suppliers (Fombrun &

Shanley, 1990). Moreover, such reputational benefits appear to accrue over time, even serving as

“insurance” that buffers the firm against future harmful events (Flammer, 2013).

Although the reputational benefits associated with firms’ sustainability–related practices may be increasing, public criticism of those practices also appears to be increasing. In many cases, such public criticism can affect the associated firm’s reputation in lasting way. For example, Nike had been growing rapidly, but its profits collapsed in late 1997 due to public outrage over labor practices. In 1997, there were more than a thousand media mentions of Nike & “sweatshops” or

“child labor”, or “exploitation” (Locke, 2003). This represents an example of criticisms on cross– cutting supply chain issues and the social (“S”) dimension of ESG incidents. More recently, BP had been ferociously criticized on environmental (“E”) issues (e.g., global pollution, local pollution, climate change, impacts on landscape) from the Deepwater Horizon oil spill disaster in

2010. Similarly Rolls–Royce has been criticized on governance (“G”) issues (e.g., corruption, fraud, bribery) since 2017. Apart from these highly observable criticisms, public criticisms across social media that may be “flying under the radar” (George et al., 2016) have also been shown to have important reputational effects, as internet activists move quickly and exert increasing pressure on corporations’ sustainability practices (Luo, Zhang, & Marquis, 2016).

How Firms Seek to Manage Reputations through Sustainability–Driven Innovation

The managerial challenge of responding to sustainability issues and the importance of innovation for addressing those challenges has received increasing attention from both practitioners as well as academics (Durand, et al., 2019). One of the critical assumptions within the existing literature is that public criticism offers potential opportunities for organizations to demonstrate learning and innovation. For instance, the strong public environmental criticisms on

Dow Chemical Company, since 1992, have triggered the company to introduce massive innovation

7 including waste reduction, new solar–cell products, and processes for improved corporate governance (Eccles & Serafeim, 2013). Earlier this year, the UNGC launched the “Sustainable

Development Goals (SDG) Ambition” to redirect global companies’ actions at an unprecedented scale toward integrating SDGs into their core businesses by 2030, citing the importance of innovation in making this shared success a reality (Accenture & UNGC, 2020). Existing scholarship has also documented how innovation in response to reputational threats, socially incompatible expectations and changes in normative pressures are critical for firm competitive advantage as well as long–term firm survival (e.g. Greenwood et al., 2011; Greenwood, Oliver,

Sahlin, & Suddaby, 2012; Hrebiniak & Joyce, 1985)

Most research on strategic responses to normative pressures has focused on exploring both the internal and external factors that shape those responses (Durand et al., 2019; Waldron, et al.,

2019). Greening and Gray (1994), for example, highlight how external factors including public interest group pressure, media exposure and crises, as well as internal factors including firm size and organizational commitment influence firms’ issue management. This early empirical work examines how firms within three different industries manage reputational threats by way of substantive conformity. Durand et al. (2019) build on these and other similar early insights, proposing a conceptual model of five distinctive possible firm responses to normative pressures, which vary in terms of their substantive versus symbolic nature and whether those responses involve conformity or compliance. Notably, they argue that these different responses are shaped by decision–makers’ perception of issue salience (Bundy, Shropshire, & Buchholtz, 2013) and the cost–benefit analysis of resource mobilization.

Taken together, this body of literature offers important theoretical underpinnings of organizational responses to public criticism at a given point in time. However, despite this

8 important and longstanding inquiry on how media and public criticisms can elicit corporate sustainability–driven innovation, little is known about how such criticism might interact with conditions of global crisis and upheaval to reputationally constrain such innovation over time. As such, we complement these studies by beginning to theorize about the important ways in which organizations might be reputationally imprinted during periods of crisis, thus shaping divergent responses to ESG incidents over time and following the crisis.

How Public Criticism During Crisis Imprints on Corporate Reputations and Constrain

Sustainability–Driven Innovation

Prior work has largely assumed that firms operate with some discretion in managing their reputations—that they can act strategically as and when ESG incidents arise so as to avoid any associated costs (Barnett & Hoffman, 2008; Delmas & Toffel, 2008; McDonnell & King, 2013).

Drawing on the longstanding literature on organizational imprinting, however, we question these assumptions by exploring how firms’ exposure to extensive public criticism during periods of crisis imprints upon those organizations’ reputations thereby constraining firms’ discretion in responding to public criticism. Marquis and Tilcsik (2013, p. 199) define imprinting as “a process whereby, during a brief period of susceptibility, a focal entity develops characteristics that reflect prominent features of the environment, and these characteristics continue to persist despite significant environmental changes in subsequent periods.” These periods of susceptibility are thought to be characterized by organizational instability or upheaval, such as is often experienced amidst transition (Greenwood & Hinings, 1996; Tushman & Romanelli, 1985) or during instances of social unrest or economic crises (Pearson & Clair, 1998; Shrivastava, Mitroff, Miller, &

Miclani, 1988). As these periods of susceptibility unfold, features of the contemporary environment come to indelibly mark and define the organization. To date, much of this prior

9 research has focused on how particular organizational practices and structures come to reflect the organizations’ environments (Kimberly, 1975; Marquis & Huang, 2010). Here, we extend this line of argumentation to suggest that during these moments of crisis–induced instability and transition, firms’ reputations are additionally susceptible to such imprinting effects. Because crises raise audience awareness of firms who have been severely and publicly criticized for their ESG practices

(Coombs, 2007; Dean, 2004; Wei, Ouyang, & Chen 2017), such momentary scandals experienced during periods of crisis may form durable reputational effects even after the crisis has subsided.

How reputational imprints constrain access to innovation–related resources

As companies look to improve or supplement their sustainability practices or competencies in response to future ESG incidents, they need to innovate across different domains such as developing new toxic emission management and environmental management systems (e.g.,

Melnyk, Sroufe & Calantone, 2003; Russo & Harrison, 2005), new product designs (e.g., Waage,

2007), new health and safety procedures (e.g., Eccles & Serafeim, 2013) and new supply chains

(e.g., Linton, Klassen & Jayaraman, 2007). In pursuing such innovations, they often require substantial investments and indeed access to entirely different resource pools than the ones upon which they have relied historically. However, as the owners of those sustainability–oriented resources evaluate new potential partnerships, they do so in part by accounting for the new partner’s historical reputation. As such, they are more likely to withhold resources from companies which have been publicly and prominently criticized for their ESG practices during times of crisis, as those crisis-based criticisms imprint on the firms’ reputations, anchoring judgments about the risk of “greenwashing” and the risk that such negative reputations would spill over to tarnish their own (Barnett, 2014; Cheng et al., 2014). Anecdotal evidence by RepRisk research suggests that

ESG scandals increase not only a focal firm’s own reputational risk but also those of its suppliers

10 and customers, leading to falling stock prices of all the firms across the supply chains (RepRisk,

2020). This reputational spillover risk has intensified along with the dynamic growth in global supply chains and information technologies over the recent decades (Diestre & Rajagopolan,

2014). For these reasons, a negative sustainability reputation is likely to impede firms’ access to essential financial and sustainability–related resources, deterring sustainability–driven innovation

(Waldron, et al., 2013).

On the one hand, the preceding arguments may suggest that organizations which are severely criticized during periods of global crisis would merely look to uphold the status quo in response to future ESG incidents, retaining any existing sustainability practices while foregoing the pursuit of costly or inaccessible sustainability–driven innovations. Yet on the other hand, we suspect that some companies may counterintuitively shift resources away from any existing sustainability practices. Recent studies on stigma, for example, suggest that some organizations may actually embrace their potentially negative or deviant reputations as a means for signalling distinctiveness (Helms & Patterson, 2013; Hsu & Grodal, 2020; Zhao et al., 2017). In this way, organizations that have faced negative reputational imprinting effects during crisis may embrace the associated public criticism not as a reputational threat but rather as further affirmation of their ability to remain competitive despite and perhaps in light of their lacking sustainability practices. As such, organizations with negatively imprinted reputations may become incentivized by public criticism to further diminish their existing sustainability practices, which they view as neither central nor distinctive.

How reputational imprints constrain the ability to capitalize on innovation–related resources

Organizations with positively reputations likely benefit from increased “stakeholder influence capacity,” which refers to the ability of firms to identify, act on, and profitably exploit

11 sustainability–driven innovation by way of stakeholder relationships (Barnett, 2007; Shea &

Hawn, 2019). This is because such positive sustainability reputations enhance the mutual trust and close relationships with primary stakeholders, which are essential for the success of sustainability– driven innovation. Barnett & Salomon (2012) provide supportive evidence of an asymmetric U– shaped relationship between corporate social and financial performance, where firms with the strongest sustainability reputations realize the highest profits from their sustainability–related investments, whereas firms with the weakest reputations have higher profits than firms with moderate practices. In sum, not only is extensive public criticism directed at a firm during periods of crisis likely to determine the firm’s access to the resources needed to engage in sustainability– driven innovation, but the corresponding reputational imprints also hinder the ability of firms to capitalize on sustainability–driven innovation over time. Accordingly, we propose the following hypotheses:

Hypothesis 1: Firms exposed to relatively less extensive public criticism during periods of crisis are prone to learn from and increase sustainability–driven innovation in response to ESG incidents experienced after the crisis.

Hypothesis 2: Firms exposed to relatively more extensive public criticism during periods of crisis decrease sustainability–driven innovation in response to ESG incidents experienced after the crisis.

Accordingly and consistent with the above arguments, we further suggest that sustainability–driven innovation generates new competencies and competitive resources for the firms. We thus posit that ESG practices serve as legitimating actions which help reduce ESG incidents in the subsequent period:

Hypothesis 3: An increase in ESG practices leads to lower ESG incidents after the crisis for both types of firms.

The Moderating Effects of Product Market Competition

The prior arguments and hypotheses suggest two distinct patterns of corporate responses to public ESG incidents over time as illustrated in Figure 1. Specifically they suggest that firms which go into a crisis with similar levels of prior ESG practices, size and financial performance— and even similar prior responses to ESG incidents—will over time begin to diverge in those responses as a consequence of reputational imprints formed during periods of crisis. Firms that are exposed to a higher number of ESG incidents during these periods, despite initially engaging in similar ESG practices, draw the conclusion that further investment in sustainability–driven innovation is unjustified, thereby shifting their strategic innovation efforts elsewhere, triggering the “downward spiral.” Other firms, in contrast, which are exposed to positively imprinted reputations, perceive that they are likely to be rewarded for their sustainability–driven innovations with a relatively lower level of public criticism and thus draw the conclusion that these innovations are good strategic investments, thereby triggering a positive spiral. Thus, initially similar firms become more distinguishable over time as reputational imprints formed during periods of crisis constrain future strategic investment. To further illustrate these divergent reputational effects and explore an important contingency which might further reinforce the imprinting effects of public criticism during crisis in the next subsections we consider how product market competition may further condition these reputational imprinting effects.

***Insert Figure 1A and 1B about here***

Product market competition

In highly competitive markets, a firm faces intense competitive pressures to go beyond conforming with standards to differentiating itself from its rivals in order to sustain competitive advantage (Fisman, Heal & Nair, 2006). Mohliver, Crilly & Kaul (2019), for example, construct a mathematical model of organizational responses to corporate activism, suggesting that market competition is likely to accentuate any divergence in those responses by increasing firms’ perceived and actual need to stand out from their rivals.

How then might product market competition intersect with moments of global crisis to shape the effects of public criticism upon firms sustainability practices over time? Prior studies, for instance, emphasize the competitive advantages that might accrue to organizations as they enhance their sustainability reputations (Flammer, 2013; Hawn & Kang, 2018; Jones, 1995; Porter,

1980; Rindova & Fombrun, 1999). As such, we might then expect all organizations to pursue such enhancements. For instance, there are prominent anecdotal examples like that of Nike, in which the organization after receiving extensive public criticism regarding the ethics of its supply chain progressed to become a widely celebrated example of corporate governance and transparency. Yet, such anecdotes do not account for the effects of periods of crisis. As we have argued when prominent and extensive criticism is received during periods of crisis, it can constrain such progress and sustainable enhancements. Correspondingly then, it is equally likely that product market competition will accentuate both the benefits of learning from public ESG criticism but also the reputational constraints associated with trying to do so, when the firm has been extensively criticized during periods of crisis. Such arguments are consistent with prior empirical research, which has illustrated how competition amplifies the need for trust and strong stakeholder relationships (Fernández‐Kranz & Santaló, 2010; Flammer, 2015). In such competitive markets,

14 therefore, companies who are less extensively criticized and thus viewed as relatively superior in terms of ESG performance may actually benefit even more by way of cheaper sources of capital, increasing labor productivity, talent retention, and the ability to preempt costly regulatory actions

(Fernández‐Kranz & Santaló, 2010).

Conversely, competitive pressures are likely to amplify the deleterious effects of extensive public criticism received during periods of crisis, thereby further constraining respective firms’ ability to acquire sustainability–related resources and the ability to capitalize on sustainability– driven innovation. When faced with such accentuated competition, firms which were targets of more extensive public criticism will be even more likely to negatively respond to that criticism, further reducing their sustainability practices in an attempt to signal their distinctiveness. For these reasons and evidence, we posit that while competition may induce sustainability–driven innovation among those firms with positively imprinted sustainability reputations, it has perverse effects on those firms with negatively imprinted reputations by accelerating their vicious cycle of reducing

ESG practices.

Hypothesis 4: Operating within competitive industries will further increase sustainability–driven innovation in response to ESG incidents for those firms which received relatively less extensive public criticism during periods of crisis.

Hypothesis 5: Operating within competitive industries will further decrease sustainability–driven innovation in response to ESG incidents for those firms which received relatively more extensive public criticism during periods of crisis.

DATA

We construct two primary ESG outcome measures using data from two main sources: ESG incidents derived from the RepRisk database, and ESG practices derived from the Morgan Stanley

Capital International’s (MSCI) Intangible Value Assessment (IVA) database. The RepRisk database includes screened negative ESG news (incidents) from more than 80,000 traditional and social media, stakeholder, regulatory, NGOs and other third–party sources in 15 languages on a daily basis since 2007. All incidents are categorized into one of 28 predefined ESG issues under 5 themes: environmental footprint, community relations, employee relations, corporate governance, and cross–cutting issues. For our analysis, we construct the indices of ESG incidents from raw data on firm–level incident counts of 28 ESG categories across the RepRisk universe from 2007 to 2018.

Specifically, we construct severity–weighted news for sub–components of E, S, and G domains for each firm-year by summing and linearizing the three levels of severity—high severity incidents are counted with a value of three, medium with two, and low with one. This linearizing of RepRisk incidents is widely used in prior studies (e.g., Kölbel & Busch; 2013). RepRisk ranks the three levels of severity based on three equal–weighted subcategories: the extent of consequences (e.g., “health and safety” issue – minor injury, major injury, fatal), the number of people affected (one person, a group of people, a large number of people), and a firm’s intention

(caused by an accident, negligence, or systematic malpractice). We sum RepRisk’s monthly incident data to create annual indices for each firm in order to facilitate merging with the the MSCI data, which is reported annually. We then aggregate E, S and G categories into the “ESG incidents” measure. We also construct an alternative index for reach–weighted ESG incidents, which take into account the relative reach of different reporting information sources.1 Although the main analysis uses severity–weighted ESG incidents, our results (untabulated) are robust to this alternative measure.

1 High influence sources are international media with a strong global presence e.g., the New York Times, BBC, the Financial Times, among others. Medium reach includes national and regional media, international NGOs, state, national and international governmental bodies (circulation at least 150K). Low reach includes local media, local newspapers, small NGOs, blogs with a circulation less than 150K. 16

To examine how firm reputation constrains sustainability–driven innovation, we merge the

ESG incidents dataset from RepRisk with ESG practices data from MSCI IVA. MSCI ESG ratings are the most widely used indices by the investment community of over 1,000 institutional investors, including 46 of the world’s top 50 asset managers. The coverage includes the major equity indices around the globe, such as MSCI World Index, MSCI Emerging Market Index, and

MSCI country–specific Investible Market Indices, among others.

We use the weighted–average ESG management score before adjustment relative to a company’s industry peers’ standards and performance. ESG management is defined as the extent to which a company has developed policies, programs, and strategies to manage its ESG risks and opportunities of exposures to ESG issues. This is consistent with our previously stated definition of sustainability–driven innovation. ESG management data is compiled mainly from company reports, industry reports, and thematic reports, as well as specialized data sets from government and NGOs, and insights from analyst calls and webinars. The yearly absolute correlation between our measures of ESG practices and ESG incidents is very low ranging from the lowest -0.0027 in

2013 to the highest 0.11 in 2009. This low correlation suggests that our measures of ESG practices and ESG incidents are mutually exclusive—whereas our ESG practices measure captures a firm’s sustainability policies, programs, and strategies, our ESG incidents measure captures public criticism of a firm’s perceived ESG risks or scandals. The ESG ratings are classified into 37 key issues under 10 macro themes and 3 pillars: the E pillar includes climate change, natural resources, pollution and waste, and environmental opportunities; the S pillar includes human capital, product liability, stakeholder opposition, and social opportunities; the G pillar includes corporate governance and corporate behavior. Each key issue score is weighted based on MSCI’s material mapping framework and aggregated to an overall ESG score. All key issues and overall ESG scores

17 range from 0, being the worst, to 10, the best performance.

Finally, we merge our dataset with company financial and accounting variables from the

Thomson Reuters Worldscope to control for observable firm characteristics. We also obtain country-level variables from the World Bank. The final sample includes 34,231 firm–year observations for 4,738 unique firms in 76 countries over 12 years from the year 2007 to 2018.

METHODOLOGY

Firm Classification

To examine how public criticism received during the global financial crisis affect firms’ ongoing approaches to sustainability–driven innovation following the crisis, we classify firms based on the level of ESG incidents during the global financial crisis from 2007-2009 (often labelled as the Great Recession) and perform the analysis on how firms respond to incidents during

2010-2018, the remaining years after the global crisis. We employ the following industry–size matched procedures to classify firms into two distinct groups: low–incident (positively imprinted) and high–incident (negatively imprinted) firms. First, companies are classified into major industry groups based on a four–digit Thomson Reuters Industry Group Classification. Second, within an industry, we compute the average of market capitalization over the Great Recession and classify firms into two subgroups of big and small firms by comparing firm size with its industry median.

We then compute average ESG incidents over the three-year period for each firm and classify firms into two groups, low and high incidents, based on its group “median” ESG incidents.

The results are qualitatively similar when we classify firms by using the mean instead of median, regardless of matching by industry and size or only by industry, and regardless of using either two–digit Thomson Reuters major industry group or two–digit SIC code instead of the four– digit industry classification. The remaining samples after the first three years for the main analysis

18 in the PVAR model that have ESG incidents and ESG practices data, and all the financial variables are unbalanced longitudinal panel dataset, covering 8,703 low–incident and 7,032 high–incident

(negatively imprinted) firm–year observations from the year 2010 to 2018.

As the PVAR model allows us to control for only a limited set of exogenous variables in the regressions, we complement the above firm classification with a matched-sample design to precisely infer how public criticism during the crisis imprinted upon firms’ reputations, reinforcing positive and negative divergent spirals over time. We employ coarsened exact matching (CEM) to find matched sample firms to control for other important observable characteristics at t-1.

Specifically, treated firms are those firms that were severely criticized during the global crisis. We select matched control firms that are in exactly the same industry, were founded in the same periods and in countries with similar level of CO2 emissions per capita, have very similar size,

ESG practices, profitability, slack resources (i.e., current ratio), and firm risks (i.e., stock price volatility) with an exception that the control firms were exposed to low incidents during the global crisis. The matching method helps reduce imbalance in covariates between treated and control groups as well as possible unobserved heterogeneity as if a randomized experiment had been done, thus increasing the precision of the estimated causal inferences (Blackwell, Iacus, King, & Porro,

2009). Moreover, by including firms’ founding years and founding environmental conditions as matching attributes, we help control for possible earlier imprinting effects arising from formation periods. Finally, employing exact matching on firm size, profitability, slack resources, firm risks and initial ESG performance, we also help address endogeneity arising from a broader strategy of the firms; for instance, resources constraints, management decisions, and firms’ capabilities in addressing ESG issues.

The matched sample observations from 2010 onwards that have all variables for PVAR

19 analysis reduce to 7,446 observations for less severely criticized firms and 5,879 observations for more severely criticized firms. Multivariate L1 distance after matching is 0.79, which dropped significantly from unmatched data. The t-test and Kolmogorov-Smirnov test, which indicate differences between treatment and control groups for all variables after matching are also insignificant at 10% level. Both of these statistics justify the reliability of our matching samples.

We report the feedback loops from the PVAR analyses of both the full sample (Table 3) and the matched sample (Table 4), which will be discussed in the Results section.

Empirical Model

To test our conceptual model, we use Granger causality test (Granger, 1969) in the panel vector autoregression (PVAR) controlling for exogenous variables. This method has been previously adopted by Lev, Petrovits & Radhakrishnan (2010) and Dowell, Hart & Yeung (2000), both of which employ PVAR analysis in the sustainability context. PVAR is particularly suitable for our study given that the method including its associated impulse response functions allows us to examine the dynamic recursive feedback loops between firms’ sustainability reputations and their sustainability–driven innovation as well as the dynamic evolution of the systems. Thus,

PVAR enables us to test our conceptual model, exploring how public criticism received during a period of crisis can imprint on firms, thus compelling divergent reputational spirals over time.

We use the generalized method of moments (GMM) estimation to estimate the following equations:

퐸푆퐺 푝푟푎푐푡푖푐푒푠푖푡

= 훼 + 훽1퐸푆퐺 푝푟푎푐푡푖푐푒푠푖푡−1 + 훽2ESG practices푖푡−2

+ 훽3Ln (1 + 퐸푆퐺 푖푛푐푖푑푒푛푡푠)푖푡−1 + 훽4Ln (1 + 퐸푆퐺 푖푛푐푖푑푒푛푡푠)푖푡−2

+ 훽5Ln (푚푎푟푘푒푡 푐푎푝푖푡푎푙푖푧푎푡푖표푛)푖푡−1 + 훽6푅푂퐴푖푡−1 + 훽7퐿푛(푎푔푒)푖푡−1

+ 휀푖푡 (1)

퐿푛(1 + 퐸푆퐺 푖푛푐푖푑푒푛푡푠)푖푡

= 훼 + 훽1Ln (1 + 퐸푆퐺 푖푛푐푖푑푒푛푡푠)푖푡−1 + 훽2Ln (1 + 퐸푆퐺 푖푛푐푖푑푒푛푡푠)푖푡−2

+ 훽3퐸푆퐺 푝푟푎푐푡푖푐푒푠푖푡−1 + 훽4퐸푆퐺 푝푟푎푐푡푖푐푒푠푖푡−2

+ 훽5Ln (푚푎푟푘푒푡 푐푎푝푖푡푎푙푖푧푎푡푖표푛)푖푡−1 + 훽6푅푂퐴푖푡−1 + 훽7퐿푛(푎푔푒)푖푡−1

+ 휀푖푡 (2) where 퐸푆퐺 푝푟푎푐푡푖푐푒푠푖푡 is MSCI IVA ESG scores of firm i in year t. Ln(1 + 퐸푆퐺 푖푛푐푖푑푒푛푡푠)푖푡 is the natural logarithm of a number of ESG incidents of firm i in year t. A logarithmic function is used only for ESG incidents because the variable is highly right skewed and then plus one to retain all observations that have zero ESG incidents. In comparison the ESG practices variable requires no transformation as it has less variation and is very close to normally distributed. We include two lags of the explanatory variables as suggested by lag order selection criteria from a regression selection model.

Despite the robustness of coarsened exact matching (CEM) on several observational characteristics, we also control for the remaining imbalance by including a set of exogenous firm– level control variables. As PVAR models allow only few exogenous variables, we select the three control variables that potentially influence the amount of public criticism and firms’ responses to such criticism in the PVAR analysis. First, we include the log of market capitalization as a proxy for firm size, since firm size may potentially influence both ESG incidents and ESG practices; for

21 instance, larger firms might receive more media attention and criticism, and yet at the same time they have more power and resources to address such criticism. Second, we control for profitability as measured by the return on assets (ROA) to account for differences in firms’ capacity to invest in “discretionary” sustainability investments following the crisis. Finally, we include firm age (in log) calculated from the difference between the current year and the founding year of each firm- year observation to help control for the possibility of imprinting effects during formative periods as suggested by a large body of work (Marquis & Tilcsik, 2013; Marquis & Qiao, 2020; Tilcsik &

Marquis, 2014); 휀푖푡 is an error term.

RESULTS

Summary Statistics

Table 1 presents the average of the main observational characteristics of positively imprinted and negatively imprinted firms during the three-year classification period. During the

2007–2009 global financial crisis, firms which were less extensively criticized were exposed to an average of 3 ESG incidents per year, whereas firms which were more extensively criticized were exposed to an average of 15 and a maximum of 650 ESG incidents per year. The majority of our sample companies (66.12%) are classified into the less extensively criticized group, and the remaining 33.88% are classified into the more extensively criticized group. The presence of unequal distribution is because there are several industries in which only few companies are very highly criticized, while a large number of firms have an equally low level of ESG incidents. For instance, 81% of companies in the Medical Services industry were exposed to 3 incidents during

2007-2009, whereas the remaining 19% were exposed to greater number of incidents from 5 up to

27 incidents on average over the 3-year crisis. Therefore, we classify all 81% into the less extensively criticized group.

Interestingly, however, those firms which were highly criticized during the crisis actually demonstrated a slightly higher (i.e., more sustainable) level of ESG practices at that period (mean

ESG practices equal 5.05 for positively imprinted firms, 5.27 for negatively imprinted firms) and also are slightly more profitable (average ROA equals 5.49 for positively imprinted firms and 5.78 for negatively imprinted firms). This initial finding suggests that the firms that were more highly criticized neither have more constrained resources nor less capable management, which might otherwise be reflected in lower profits, comparing to average firms (t-stat ROA -1.24; p-value

0.215, which is insignificant at 10% level). In other words, resource constraints during the crisis, different strategies, initial capabilities or other organizationally determined inertial forces are unlikely contingencies that drive divergent responses over time after the crisis. Table 1 presents more detailed distribution of ESG practices, size, profits and other financial variables over three years. Apart from ESG incidents, all other observational characteristics including financial performance are not significantly different between our two classified groups of firms during the crisis. Perhaps most interestingly, firms’ responses to ESG incidents initially during the crisis are also not different. This finding suggests that the public criticism of firms which occurs during the financial crisis is partially “random”, and yet because firms are more highly susceptible to environmental imprinting effects during periods of the global crisis, such accrued criticism is an important contingency that subsequently creates lasting reputational imprints on organizational responses. We investigate this possibility in the following subsection.

***Insert Table 1 about here***

As noted, we classify firms based on the extensiveness of public ESG incidents experienced during the 2007–2009 global financial crisis, and we base the subsequent analysis on the sample years from 2010 onwards. Table 2 presents the summary statistics and correlation table

23 of key variables over the analysis period of 2010-2018. Firm-level variables include ESG incidents, ESG practices, firm size, ROA, firm age, slack resources as measured by current ratio

(current assets to current liabilities), leverage (total debt to total equity), percentage of total debt to total assets, and firm risk as measured by annual price volatility. Industry-level variable includes market concentration as measured by the Herfindahl-Hirschman Index (HHI). Country-level variables include GDP per capita (in logarithm), CO2 emissions per capita, and life expectancy.

The pairwise correlation between ESG incidents and ESG practices is positive and statistically significant at 1% level. Larger and older firms are likely exposed to higher ESG incidents, which result in greater learning or higher ESG practices. Firms in wealthy countries are also likely to have both higher ESG incidents and ESG practices. Once again, as the PVAR model only allows for a limited set of exogenous variables, in this main PVAR analysis we select only firm-level factors that most likely affect both ESG incidents and ESG practices as previously discussed in the Empirical Model section. Other industry and country factors are further controlled for and investigated in the subsequent analyses of Robustness Checks and Further Analyses section.

***Insert Table 2 about here***

Hypothesis Tests

Table 3 presents the main results from the PVAR analysis, which tests the conceptual model and Hypotheses 1, 2 and 3.

Column 1 of Table 3 shows that those firms which during the crisis experienced less public criticism continue to engage in sustainability–driven innovation, significantly increasing ESG practices in response to ESG incidents that arise after the crisis. In turn, these increased ESG practices predict lower ESG incidents (Column 2), leading to a positive feedback loop consistent

24 with Figure 1A. Firms which during the crisis experienced more extensive public criticism by contrast continue to reduce investments in ESG practices in response to ESG incidents that arise after the crisis (Column 3). After the crisis as these firms lower ESG practices, this increases future

ESG incidents (Column 4), thereby further embedding these organizations into a vicious cycle of accelerating ESG incidents as in Figure 1B. All these coefficients are statistically significant and have the correct signs as expected in the first three hypotheses. Overall, the results in Table 3 support the first three hypotheses.

***Insert Table 3 about here***

The findings from matched sample analysis in Table 4 are very similar to those from the full sample analysis. Despite a smaller sample size, the coefficients of public criticism during the crisis producing divergent responses over time are highly significant for both groups of firms. The key interpretation of these findings is thus: during a period of financial or social crisis, firms’ reputations are particularly susceptible to the imprinting effects of public criticism regarding ESG practices, and such reputational imprints impose lasting constraints on whether and how those firms engage in sustainability–driven innovation. Among firms that do not differ in their ESG practices, those that are criticized most during the crisis “learn” to treat a negative reputation as a cost of doing business and reduce ESG investments, while firms on average that are weakly criticized during the crisis invest more and innovate in ESG practices. This dynamic evolution of sustainability–driven innovation and ESG incidents can also be illustrated with impulse response functions associated with the PVAR model in Figure 2A and 2B.

***Insert Table 4 about here***

Figure 2A represents impulse functions of those firms with less extensive criticism as per the matched sample analysis, which controls for the many important observational characteristics

25 in Table 4. Table 5 compares forecasts from the impulse response functions between less and more extensively criticized firms. The top–right diagram in Figure 2A shows that for these firms a one– standard–deviation shock to ESG incidents following the crisis increases future ESG practices. It is important to note that this positive impact is not contemporaneous as it requires significant time and efforts for firms to make substantive investments to innovate or implement new sustainability strategies and policies. It takes at least one year to observe a sharp increase in ESG practices (t+1) in response to a shock in period t0 and then a spike in t+2 and peak at 0.03 before going back to the steady state after 10 years. Likewise, as indicated in Table 5 when firms implement ESG practices after the crisis, this shock in ESG practices has the strongest impacts in reducing ESG incidents in the next 2 years when new systems and innovation can be executed and operated at their full capacity. It is clear from Figure 2A that firms with positively imprinted sustainability reputations actively address incidents by investing in sustainability practices (top–right), and these practices significantly help resolve ESG incidents—ESG incidents have declined very rapidly by more than 90% within 3 years (top–left in Figure 2A and Table 5).

In contrast, as per Figure 2B, firms that are more extensively criticized during the global crisis reduce ESG innovation following the crisis in response to a shock to incidents (top–right).

As such, they face increasing ESG incidents or consistent public criticism of sustainability issues for very long–time periods (top–left). This effect again is not contemporaneous as organizations take time to mobilize resources to pursue competitive advantage elsewhere. As compared to Figure

2A, Figure 2B presents a much higher degree of inertia as firms’ negatively imprinted reputations constrain their sustainability investments and innovation, which then establishes a new negative equilibrium of persistently high incidents before the system adjusts to the long–run steady state in a more distant future. As illustrated in Table 5, while a shock to ESG incidents for positively

26 imprinted firms vanishes by 3 years, it takes 40 years for negatively imprinted firms to reduce a shock by 90% (from 1 to lower than 0.1). Finally, the bottom–left diagram is consistent with our hypothesis 3—a positive shock to ESG practices significantly lowers ESG incidents. In turn, a negative shock will have the reverse effect of accelerating ESG incidents. Only a single shock to

ESG incidents can create persistent effects of decreasing innovations, which further increase incidents, starting the next recursive vicious cycle. This suggests that when firms’ negatively imprinted reputation triggers them into a vicious cycle with multiple shocks to ESG incidents, the deleterious effects on innovation will be even more persistent and severe than the impact from a single shock illustrated in Figure 2B. These impulse response functions enable us to confirm the lasting constraints posed by public criticism during periods of global crisis on firms’ sustainability–driven innovation.

***Insert Figure 2A and Figure 2B about here***

***Insert Table 5 about here***

Product Market Competition

Table 6 reports the regression results in support of our Hypotheses 4 and 5. We measure industry concentration within a three–digit SIC by calculating the Herfindahl–Hirschman index

(HHI), which is the sum of squared market shares of all firms competing in an industry. We divide our two groups of firms further into two subgroups: firms in competitive and concentrated industries. Table 6 shows that intense product market competition intensifies the observed reputational spirals, widening the discrepancy of sustainability–driven innovation between firms that are less and more extensively criticized during periods of global crisis. The effects of imprinted reputations on sustainability–driven innovation have a higher magnitude and are more statistically significant for firms in competitive markets despite the smaller sample size. Taken together, these

27 results offer support for both Hypotheses 4 and 5, suggesting that competitive pressures amplify both virtuous and vicious reputational cycles as depicted in Figure 1A and 1B.

***Insert Table 6 about here***

ROBUSTNESS CHECKS AND FURTHER ANALYSES

As a robustness check, we further control for and examine alternative explanations of potential reputational imprints since organizational founding before the global financial crisis. As the formative periods of organizations is analogous to a biologically driven sensitive period

(Bamford, Dean, & McDougall, 2000; Johnson, 2007), we examine this possibility of imprinting effects during formative periods by controlling for different organizations’ founding years and founding environments. Finally, we complement the key PVAR analysis with the panel regressions and matching design by using cem weights generated by cem command for each observation to estimate the causal effect. The nonparametric CEM approach sorts all the observations into strata.

Within each stratum, the approach computes the weights for each control group observation based on its similarity to the members of the treatment group observations and discards (i.e., assigning weight 0) to those within any stratum that are dissimilar to members of the treatment group. We use the cem matching weights for panel regression analysis to confirm the robustness of the key results.

Organizations’ Founding Periods

We subcategorize both groups of firms which were exposed to high and low incidents during the crisis into 3 subgroups by their founding years. The first group is firms that were founded before the Friedman’s (1970) doctrine of shareholder primacy. The influential shareholder theory was introduced in September 1970 by Milton Friedman in his New York Times essay entitled: “A Friedman Doctrine: The Social Responsibility of Business Is to Increase Its Profits”.

Since then, shareholder and agency theories had a significant impact on the corporate world and led to a dramatic increase in executive equity-based compensation as firms attempt to align managers’ interests with those of shareholders. Thus, the second period is from 1971 to 1990, starting with Friedman’s (1970) shareholder theory and subsequently agency theory (Eisenhardt,

1989; Jensen 1986; Jensen & Meckling, 1976). The third period is after 1990 when corporations increasingly began broadening their focus on maximizing shareholder values toward embracing and acting on a larger set of stakeholder interests. In 1990, for instance, MSCI launched the world’s first socially responsible index, the MSCI KLD 400 Index to help socially conscious investors factor social and environmental issues into their investment decisions (KLD, 2008). The introduction of ESG data in 1990 led to an emergence of corporate adoption of ESG policies and a cultural commitment to sustainability, which has evolved and significantly increased over 30 years in response to pressing global challenges.

As shown in Table 7, although firms were founded in different time periods with different doctrines of how to run a successful private enterprise, average firms that were weakly criticized during the global financial crisis learn from public criticism by increasing ESG practices in response to ESG incidents. In contrast, the firms that were strongly criticized during 2007-2009 global crisis reduce ESG practices in response to ESG incidents and fall into a vicious cycle. These divergent responses between two groups of firms are statistically and economically significant across 3 different founding periods. The robust finding confirms that the global crisis is a sensitive period when organizations’ prior reputations are susceptible to public criticism, resulting in the enduring influence of reputational imprints that constrain future sustainability-driven innovation.

***Insert Table 7 about here***

Founding Environments

As our analysis focuses on global companies across 70 countries with diverse founding environments, these external founding conditions might institutionalize organizational structures, culture, routines and processes, which contribute to lasting impacts on firms’ strategies (Marquis

& Huang, 2010). In Table 8, we control for one key founding environmental characteristic, which is country environmental sustainability as measured by CO2 emissions per capita during the periods when the firms were founded. As ESG data has not yet emerged before the 1990s, founding country CO2 emissions per capita is the variable that has the greatest coverage across founding periods of firms in the sample. Thus, we use this CO2 emissions variable as a proxy for founding country ESG performance.

In Table 8, we subcategorize our two groups of firms into subgroups by the median of country-year level of CO2 emissions per capita. The first subgroup consists of firms that were founded in country-year in which CO2 emissions less than the median of 12 metric tons per capita, and the second subgroup include firms founded in country-year with worsen air pollution, CO2 emissions at least 12 metric tons per capita. Table 8 provides strong evidence of firms’ divergent approaches to sustainability-driven innovation after the global financial crisis despite different founding environmental conditions. As shown in Columns 1 and 3, firms which are less extensively criticized during the crisis significantly increase ESG practices in response to ESG incidents after the crisis, whereas Columns 5 and 7 show significant negative impacts of ESG incidents on future ESG practices for negatively imprinted firms.

Thus, our findings provide strong evidence that on average firms learn from public criticism, yet this learning effect hides a deeper contingency of crisis and the enduring influence of reputational imprints from the crisis that contribute to firms’ lasting divergent responses to

30 public criticism regardless of the difference in their founding environments.

***Insert Table 8 about here***

Panel regressions

Table 9 presents panel regressions estimated from the matching weights from the CEM matching approach. Column 1 provides strong evidence that firms on average and across periods after the crisis (2010-2018) learn from public criticism – the impact of ESG incidents on future

ESG practices is positive and significant. Our variable of interest, however, is the interaction term between ESG incidents and the dummy for treatment groups of negatively imprinted firms (high

ESG incidents) from the global crisis. The coefficient is negative and statistically significant at the

1% level with high economic magnitude of -0.055. These negative responses offset the positive and significant impacts of ESG incidents on ESG practices (0.018) of average firms. The panel regression results in Table 9 confirm the robustness of the divergent reputational spirals between the two groups of firms and the feedback loops from the main analysis of PVAR model. Despite the fact that learning effects hold on average, it hides a deeper contingency of the crisis such that a smaller number of firms which were strongly criticized during the global financial crisis become stigmatized, spiraling into a vicious cycle over time.

It is also interesting to note that the coefficient of the interaction between founding CO2 emissions per capita and the treatment dummy is very small and not statistically significant. This supports our conjecture that lasting difference in future sustainability innovation is not shaped by differences in founding environmental conditions.

***Insert Table 9 about here***

DISCUSSION

Sustainability issues have become increasingly essential to firm reputations and their successful interactions with stakeholders. Some scholars suggest that this trend has, in part, arisen from a new wave of economic globalization and digital interconnectedness, which has expanded the scope and long–run costs of negative externalities across global supply chains. These negative externalities together with the repeated failures of laws and regulations urge stakeholders to protect their own interests by pressuring firms to act in socially responsible ways (Bénabou & Tirole,

2010; de Bettignies & Robinson, 2018). In addition to pressures from key stakeholders, the increasing influence of activist groups and scrutiny by the media inevitably elevates sustainability issues as a core feature of corporate reputations. According to the recent Accenture and United

Nations Global Compact (UNGC) CEO survey, 99 percent of the CEOs of the world largest companies report that sustainability issues are critical to corporate reputation and the future success of their businesses (Accenture & UNGC, 2019). Given such extensive recognition of the criticality of sustainability–related issues, we might expect for all firms to actively pursue sustainability– driven innovation as a source of reputational advantage, particularly when prompted to do so by way of public criticism.

Our findings, however, demonstrate that periods of global crisis serve as an important contingency, shaping the effects of public criticism on firms’ responses over time. When firms are exposed to public criticism of their ESG practices during periods of crisis—in our case the global financial crisis of 2007–2008, this criticism or lack thereof imprints on those firms’ reputations, ultimately constraining their capacity to engage in sustainability–driven innovation in the future.

As such, although firms that were less extensively criticized during the crisis are able to learn from that criticism in ways that prompt ongoing sustainability–driven innovation, those firms that are

32 more extensively criticized during the crisis are reputationally constrained such that they consistently compromise their sustainability practices when faced with ESG incidents following the crisis. And moreover, our findings demonstrate that competition further amplifies these divergent reputational spirals. These findings offer several important contributions to the growing literature on organizational imprinting and that which is focused on the intersection of corporate reputation and sustainability.

The Effects of Global Crisis on Organizational Imprinting

Recent reviews of the imprinting literature have reaffirmed the theoretical importance of crises as periods of upheaval during which organizations and other actors are increasingly susceptible to the influence of the external environment (Marquis & Tilcsik, 2013; Simsek, Fox,

& Heavey, 2015). Such reaffirmations helpfully move the imprinting literature beyond its original focus on founding periods, and yet the aforementioned reviews also point to the paucity of existing scholarship which has empirically examined crisis-related imprints. As such, this literature has left a number of critical open questions as to how crisis periods may correspond with or differ from other susceptible periods during an organization’s life cycle. Our findings, which focus specifically on the global financial crisis of 2007-2009 thus offer important extensions to our understanding of organizational imprinting by clarifying both the mechanism by which the environment exerts influence during this period (i.e., public criticism) as well as the features of the organization that are most susceptible to such influence (i.e., firm reputations).

Each of these extensions offer important contributions to the literature. First, prior research on organizational imprinting has argued that during more susceptible periods organizations are shaped by prominent or more legitimate features of the current environment (Baron, Hannan, &

Burton, 1999). The mechanism by which such an environmental stamp is imposed is oftentimes

33 assumed to be that of environmental selection, wherein only those organizations that align with legitimate values or logics from the environment are selected as appropriate and thus resourced

(Carroll & Hannan, 2004). Our study extends this reasoning to empirically illustrate how public criticism of firms often serves as a prominent feature of the environment during global crises.

Moreover, we highlight how such criticism can then be directed to a greater or lesser extent and in partially random fashion, thereby shaping how and to what extent firms are imprinted by such criticism. This finer-grained theoretical mechanism of public criticism during crisis thus pushes the literature to move beyond presumptions that all organizations are equally imprinted by the environment during susceptible periods, while revealing important variation in both the extent and even the direction of such imprints.

Second, much of the literature on organizational imprinting has focused on how internal features of an organization such as the organization’s identity, business model, or its form come to reflect the external environment (Baron, Hannan, & Burton, 1999; Ferriani, Garnsey, Lorenzoni,

2012). Equally important to the organization’s functioning and its success, however, are its external features such as its reputation. In our study we theorized and tested a model of how public criticism specifically imprints on organizations’ reputations during periods of global crisis, thus affecting its ability to access innovation-related resources and to capitalize on those innovations.

Such findings are again theoretically important, as they offer a useful bridge between longstanding literatures on corporate reputation and organizational imprinting, which have to date presented as different albeit not inconsistent theories for understanding the outside-in related influences imposed upon organizations. In the next two subsections, we build on these insights to reveal how this theoretical bridge also helpfully advances existing scholarship on corporate reputation.

How Reputational Imprinting Extends Scholarship on Reputational Inertia

Although recent studies offer important insight into how organizations’ reputations might change over time, either due to inconsistent actions by the firm or collective action from external actors (Briscoe & Gupta, 2016; McDonnell & King, 2013; Pfarrer, Decelles, Smith, & Taylor

2008), research on corporate reputation has more often emphasized the inertial tendencies of reputation (Ravasi, et al., 2018). To date this research on the durability of firms’ reputations has mostly focused on how early reputations cognitively anchor outside audiences’ future judgments of those firms (Barnett, 2014). Such audience–based views of reputational inertia highlight how early events get lodged into collective memory (Schultz, et al., 2001) and shape the extent to which future events are noticed and how they are interpreted (Bitektine, 2011). Other work, however, suggests that reputational inertia is rather best understood as a by–product of organizational path dependence more generally (Sydow, Schreyögg, & Koch, 2009). In other words, as organizations face pressures to reproduce themselves, either because of identity–related commitments (Hannan

& Freeman, 1984) or because of ongoing investments in familiar organizational capabilities

(Repenning & Sterman, 2002), such organizational reproduction also reinforces external audiences’ judgements regarding the firms’ reputations.

Our findings extend this research on the reputational inertia of firms by contributing the concept of reputational imprinting. Specifically, the concept of reputational imprinting and our related findings confirm the importance of the anchoring effects of prior firm reputations.

However, it is not merely that these reputations cognitively replicate over time across external audiences, but rather that during moments of profound crisis or transition firm reputations are particularly susceptible to public criticism, and that these effects persist even as the moments of crisis or transition wane. In this way, our findings reveal that reputational imprinting during

35 moments of crisis ultimately encourages divergent path dependent spirals by narrowing firms’ potential strategic options for innovation.

While acknowledging the potential for such path dependence, the concept of reputational imprinting and our associated findings also challenge the presumed mechanisms that scholars frequently attribute to organizational path dependency. Prior work suggests that such path dependency is a product of organizations’ reinforcing values, identities, and capabilities (e.g.,

Sydow et al., 2009). However, our findings illustrate that the firms which radically diverged in their approaches to sustainability–driven innovation over time, in fact, initially embraced comparable levels of sustainable practices. This suggests that while the firms may have initially operated with similar sustainability practices, competencies, and perhaps even identities, the outside–in force of public criticism experienced during a moment of global crisis ultimately imprinted on these firms’ reputations, constraining what was perceived as strategically possible.

Future work might further investigate this concept of reputational imprinting and its associated implications. For example, at what moments other than economic crises are firms’ reputations most susceptible to the imprinting effect of public criticism? How strong or frequent must the criticism be to imprint upon the organization? How do such reputational imprints inform the organizations’ identities?

Challenging Strategic Views of Reputation Management

Our findings also challenge prior research which has viewed corporate reputation as a resource that firms strategically develop and manage in order to bolster their legitimacy and competitive standing (Fombrun & Shanley, 1990; Scherer, Palazzo & Seidl, 2013; Ravasi,

Rindova et al., 2018). This research is grounded in Oliver’s seminal study (1991), which argued that public criticism and other various stakeholder demands serve as a form of feedback from the

36 institutional environment, prompting corrective strategic responses. While such corrective responses are thought to range from more accommodative strategies involving innovation and learning to more defensive strategies involving decoupling and impression management, each of these responses is meant to minimize the negative effects of those firms’ presumed misconduct while preventing future criticism (Grimes et al., 2019; Hersel, Helmuth, Zorn, Shrophire & Ridge,

2019; Zavyalova, et al., 2016). However, our findings illustrate that while firms may indeed learn from and strategically respond to public criticism regarding ESG issues, reputational imprinting constrains the possible innovation pathways available to different firms. Specifically, these findings suggest that even if firms and their leaders recognize the need for and are motivated to pursue sustainability–driven innovation, such reputational imprints create divergence in both the accessibility of sustainability–related resources as well as the capacity to effectively capitalize on related investments.

In this way, our findings also extend research that has suggested and illustrated various factors that determine firms’ responses to normative pressures. This work has ranged widely in its level of analysis, emphasizing various field or structural conditions, organizational governance– related differences, and also socio–cognitive factors which work collectively to inform how organizations respond to normative pressures often related to social or environmental issues

(Bansal, Kim, & Wood, 2018; Durand et al., 2019; Greening & Gray, 1994; Greenwood, Raynard,

Kodeih, Micelotta & Lounsbury, 2011). In particular, Durand and his co–authors (2019) argue that organizations will take substantive actions only if they perceive that particular issues are salient and that there are likely net benefits associated with taking such action. Our findings and associated conceptual model offer a useful theoretical bridge between these socio–cognitive or strategic arguments and the aforementioned structural arguments by emphasizing the important role that

37 reputational imprinting and outside audiences play in constraining the perceived salience and net benefits associated with substantive reform. Although outside audiences and stakeholders may exert normative pressures upon organizations to respond to social and environmental issues, the potential for reputational spillovers may incline those same audiences and stakeholders to withhold the very resources organizations need in order to respond. As such we demonstrate how external factors—in this case, reputational imprints formed during crisis—may constrain internal strategies, as organizations are prompted to respond to future ESG incidents.

Finally, we extend research on firms’ responses to normative issues by illustrating how reputational imprinting effects which occur amid crisis are intensified under competitive conditions. While prior work suggests that competition may compel firms to increasingly consider sustainability–driven innovation as a source of differentiation (Fernández‐Kranz & Santaló, 2010;

Fisman, Heal & Nair, 2006; Flammer, 2015), our study, however, reveals how competition might contribute to increased organizational polarization. Less extensive criticism during crisis may indeed bolster motivations for pursuing sustainability–focused differentiation and thus out- competing peers. However, our findings demonstrate that the opposite is equally possible—that more extensive criticism during crisis may constrain the means by which firms might outcompete their peers, causing these firms to actively move away from sustainability and pursue alternative sources of differentiation.

These novel insights regarding reputational constraints and spirals pose several interesting opportunities for future studies. Although we have specifically focused on how public criticism received amid crisis might cyclically constrain organizations’ responses to ESG incidents, future research might wish to consider whether such criticism might constrain other strategic or innovation–related pursuits. Additionally, while our study focused on the moderating conditions

38 of competition, future research might consider other industry–level factors which might additionally intensify or diminish the reputational effects of public criticism during crisis such as the degree of accountability or scrutiny placed on the firms’ supply chains. Finally, while our study discussed the possibilities of reputational spillover effects, future studies might consider the extent to which particular industries or geographies are exposed to such spillover effects, and whether reputational spillovers might similarly constrain the strategies of affected organizations.

CONCLUSION

Public criticism of ESG–related incidents and associated third–party audits of such incidents are currently thought to not only offer important sustainability–related information to investors but also important sources of accountability, which might then compel a virtuous cycle of continuous organizational improvement on sustainability issues. This is particularly true during periods of social or economic distress, when organizations are often called upon to improve their social and environmental impact. However, our study suggests that extensive public criticism of firms during these periods of time can be counterproductive, imprinting on those organizations’ reputations in ways that ultimately constrain perceived strategic options, thereby contributing to a

“race to the bottom” as the organizations continue to further reduce their ESG practices. And although we might assume that the potential competitive differentiation afforded by sustainability practices might compel all firms to pursue sustainability–driven innovation, our findings suggest that competition merely serves to reinforce the potential for divergent reputational spirals. In this way, it appears that policy makers should not solely rely on public criticism to drive social changes.

Instead, if such criticism can be supplemented with additional supply–chain based scrutiny, positive reputational spirals may be enhanced while simultaneously eliminating the possibility of negative reputational spirals. Ultimately, we hope this study will compel further research into

39 reputational imprinting, how such imprints constrain organizational actions, and how such constraints might be overcome. We particularly hope that our findings will help to inform enhancements in how we seek to hold organizations accountable, thus supporting a broader–scale transition toward sustainability.

REFERENCES

Accenture & UNGC. (2019). The New Decade to Deliver: A call to Business Action, UN Global Compact–Accenture Strategy CEO study on Sustainability 2019, New York: United Nations Global Compact and Accenture. Accenture & UNGC. (2020). SDG Ambition: Scaling Business Impact for the Decade of Action, New York: United Nations Global Compact and Accenture. Almandoz, J., Lee, M., & Marquis, C. (2017). Different Shades of Green: Environment Uncertainty and the Strategies of Hybrid Organizations. In Emergence, 50, 31–67. Emerald Publishing Limited. Ariely, D., Bracha, A., & Meier, S. (2009). Doing Good or Doing Well? Image Motivation and Monetary Incentives in Behaving Prosocially. American Economic Review, 99(1), 544– 555. Bamford, C. E., Dean, T. J., & McDougall, P. P. (2000). An examination of the impact of initial founding conditions and decisions upon the performance of new bank start-ups. Journal of Business Venturing, 15(3), 253–277. Bagnoli, M., & Watts, S. G. (2003). Selling to Socially Responsible Consumers: Competition and The Private Provision of Public Goods. Journal of Economics & Management Strategy, 12(3), 419–445. Bansal, P., Kim, A., & Wood, M. O. (2018). Hidden in Plain Sight: The Importance of Scale in Organizations’ Attention to Issues. Academy of Management Review, 43(2), 217–241. Barnett, M. L. (2007). Stakeholder Influence Capacity and the Variability of Financial Returns to Corporate Social Responsibility. Academy of Management Review, 32(3), 794–816. Barnett, M. L. (2014). Why Stakeholders Ignore Firm Misconduct: A Cognitive View. Journal of Management, 40(3), 676–702. Barnett, M. L., Jermier, J. M., & Lafferty, B. A. (2006). Corporate Reputation: The Definitional Landscape. Corporate Reputation Review, 9(1), 26–38. Barnett, M. L., & Hoffman, A. J. (2008). Beyond Corporate Reputation: Managing Reputational Interdependence. Corporate Reputation Review, 11(1), 1–9. Barnett, M. L., & King, A. A. (2008). Good Fences Make Good Neighbors: A Longitudinal Analysis of an Industry Self–Regulatory Institution. Academy of Management Journal, 51(6), 1150–1170. Barnett, M. L., & Salomon, R. M. (2012). Does it pay to be really good? Addressing the shape of the relationship between social and financial performance. Strategic Management Journal, 33(11), 1304–1320. Barney, J. B. (1986). Strategic Factor Markets: Expectations, Luck, and Business Strategy. Management Science, 32(10), 1231–1241. Barney, J. (1991). Firm Resources and Sustained Competitive Advantage. Journal of Management, 17(1), 99–120. Baron, J. N., Hannan, M. T., & Burton, M. D. (1999). Building the Iron Cage: Determinants of Managerial Intensity in the Early Years of Organizations. American Sociological Review, 64(4), 527–547. Barrot, J., Sauvagnat, J., (2016). Input specificity and the propagation of idiosyncratic shocks in production networks. Quarterly. Journal of. Economics. 131, 1543–1592 . Bednar, M. K. (2012). Watchdog or Lapdog? A Behavioral View of the Media as a Corporate Governance Mechanism. Academy of Management Journal, 55(1), 131–150.

Bednar, M. K., Boivie, S., & Prince, N. R. (2012). Burr Under the Saddle: How Media Coverage Influences Strategic Change. Organization Science, 24(3), 910–925. Bénabou, R., & Tirole, J. (2010). Individual and Corporate Social Responsibility. Economica, 77(305), 1–19. Berrone, P., Cruz, C., Gomez–Mejia, L. R., & Larraza–Kintana, M. (2010). Socioemotional Wealth and Corporate Responses to Institutional Pressures: Do Family–Controlled Firms Pollute Less? Administrative Science Quarterly, 55(1), 82–113. Besley, T., & Ghatak, M. (2005). Competition and Incentives with Motivated Agents. American Economic Review, 95(3), 616–636. Bhattacharya, C. B., & Sen, S. (2003). Consumer–Company Identification: A Framework for Understanding Consumers’ Relationships with Companies. Journal of Marketing, 67(2), 76–88. Bitektine, A. (2011). Toward a theory of social judgments of organizations: The case of legitimacy, reputation, and status. Academy of Management Review, 36(1), 151–179. Blackwell, M., Iacus, S., King, G., & Porro, G. 2009. cem: Coarsened exact matching in Stata. Stata Journal, 9(4): 524-546. Bowen, F. (2014). After Greenwashing: Symbolic Corporate Environmentalism and Society. Cambridge University Press. Brammer, S. J., & Pavelin, S. (2006). Corporate Reputation and Social Performance: The Importance of Fit. Journal of Management Studies, 43(3), 435–455. Briscoe, F., & Gupta, A. (2016). Social Activism in and Around Organizations. Academy of Management Annals, 10(1), 671–727. Bromley, P., & Powell, W. W. (2012). From Smoke and Mirrors to Walking the Talk: Decoupling in the Contemporary World. Academy of Management Annals, 6(1), 483–530. Bundy, J., Shropshire, C., & Buchholtz, A. K. 2013. Strategic cognition and issue salience: Toward an explanation of firm responsiveness to stakeholder concerns. Academy of Management Review, 38(3): 352–376. Carroll, G. R., & Hannan, M. T. (2004). The Demography of Corporations and Industries. Princeton University Press. Cheng, B., Ioannou, I., & Serafeim, G. (2014). Corporate social responsibility and access to finance. Strategic Management Journal, 35(1), 1–23. Coombs, W. T. (2007). Protecting Organization Reputations During a Crisis: The Development and Application of Situational Crisis Communication Theory. Corporate Reputation Review, 10(3), 163–176. de Bettignies, J. E., & Robinson, D. T. (2018). When Is Social Responsibility Socially Desirable? Journal of Labor Economics, 36(4), 1023–1072. Dean, D. H. (2004). Consumer Reaction to Negative Publicity: Effects of Corporate Reputation, Response, and Responsibility for a Crisis Event. Journal of Business Communication (1973), 41(2), 192–211. Delmas, M. A., & Toffel, M. W. (2008). Organizational responses to environmental demands: Opening the black box. Strategic Management Journal, 29(10), 1027–1055. den Hond, F., & de Bakker, F. G. A. (2007). Ideologically motivated activism: How activist groups influence corporate social change activities. Academy of Management Review, 32(3), 901–924.

Dhaliwal, D. S., Li, O. Z., Tsang, A., & Yang, Y. G. (2011). Voluntary Nonfinancial Disclosure and the Cost of Equity Capital: The Initiation of Corporate Social Responsibility Reporting. Accounting Review, 86(1), 59–100. Dieleman, M. (2010). Shock-imprinting: External shocks and ethnic Chinese business groups in Indonesia. Asia Pacific Journal of Management, 3(27), 481–502. Diestre, L., & Rajagopalan, N. (2014). Toward an Input–Based Perspective on Categorization: Investor Reactions to Chemical Accidents. Academy of Management Journal, 57(4), 1130–1153. Dixon, S. E. A., Meyer, K. E., & Day, M. (2007). Exploitation and exploration learning and the development of organizational capabilities: A cross-case analysis of the Russian oil industry. Human Relations, 60(10), 1493–1523. Dorobantu, S., Henisz, W. J., & Nartey, L. (2017). Not All Sparks Light a Fire: Stakeholder and Shareholder Reactions to Critical Events in Contested Markets. Administrative Science Quarterly, 62(3), 561–597. Dowell, G. W. S., & Muthulingam, S. (2017). Will firms go green if it pays? The impact of disruption, cost, and external factors on the adoption of environmental initiatives. Strategic Management Journal, 38(6), 1287–1304. Dowell, G., Hart, S., & Yeung, B. (2000). Do Corporate Global Environmental Standards Create or Destroy Market Value? Management Science, 46(8), 1059–1074. Dowell, G., & Swaminathan, A. (2006). Entry timing, exploration, and firm survival in the early U.S. bicycle industry. Strategic Management Journal, 27(12), 1159–1182. Durand, R., Hawn, O., & Ioannou, I. (2019). Willing and able: A general model of organizational responses to normative pressures. Academy of Management Review, 44(2), 299–320. Eccles, R. & Serafeim, G. (2013). The Performance Frontier: Innovating for a Sustainable Strategy, Harvard Business Review. Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. Academy of Management Review, 14(1), 57–74. Eccles, R. G., Ioannou, I., & Serafeim, G. (2014). The Impact of Corporate Sustainability on Organizational Processes and Performance. Management Science, 60(11), 2835–2857. Etter, M., Ravasi, D., & Colleoni, E. (2019). Social Media and the Formation of Organizational Reputation. Academy of Management Review, 44(1), 28–52. Fernández‐Kranz, D., & Santaló, J. (2010). When Necessity Becomes a Virtue: The Effect of Product Market Competition on Corporate Social Responsibility. Journal of Economics & Management Strategy, 19(2), 453–487. Ferriani, S., Garnsey, E., & Lorenzoni, G. (2012). Continuity and change in a spin-off venture: The process of reimprinting. Industrial and Corporate Change, 21(4), 1011–1048. Flammer, C. (2013). Corporate Social Responsibility and Shareholder Reaction: The Environmental Awareness of Investors. Academy of Management Journal, 56(3), 758– 781. Flammer, C. (2015). Does product market competition foster corporate social responsibility? Evidence from trade liberalization. Strategic Management Journal, 36(10), 1469–1485. Flammer, C., & Kacperczyk, A. (2019). Corporate social responsibility as a defense against knowledge spillovers: Evidence from the inevitable disclosure doctrine. Strategic Management Journal, 40(8), 1243–1267. Fombrun, C., & Shanley, M. (1990). What’s in a Name? Reputation Building and Corporate Strategy. Academy of Management Journal, 33(2), 233–258.

Friedman, M. 1970. The social responsibility of business is to increase its profits. New York Times Magazine, September 13. George, G., Dahlander, L., Graffin, S. D., & Sim, S. (2016). Reputation and Status: Expanding the Role of Social Evaluations in Management Research. Academy of Management Journal, 59(1), 1–13. Gilbert, C. G. (2005). Unbundling the Structure of Inertia: Resource versus Routine Rigidity. The Academy of Management Journal, 48(5), 741–763. JSTOR. Granger, C. (1969). Investigating Causal Relations by Econometric Models and Cross–Spectral Methods. Econometrica, 37(3), 424–438. Greening, D. W., & Gray, B. (1994). Testing a Model of Organizational Response to Social and Political Issues. Academy of Management Journal, 37(3), 467–498. Greenwood, R., & Hinings, C. R. (1996). Understanding Radical Organizational Change: Bringing Together the Old and the New Institutionalism. Academy of Management Review, 21(4), 1022–1054. Greenwood, R., Oliver, C., Sahlin, K., & Suddaby, R. 2012. Institutional Theory in Organization Studies. London: Sage. Greenwood, R., Raynard, M., Kodeih, F., Micelotta, E. R., & Lounsbury, M. (2011). Institutional Complexity and Organizational Responses. Academy of Management Annals, 5(1), 317– 371. Grimes, M. G., Williams, T. A., & Zhao, E. Y. (2018). Anchors Aweigh: The Sources, Variety, and Challenges of Mission Drift. Academy of Management Review, 44(4), 819–845. Gupta, A., & Briscoe, F. (2020). Organizational Political Ideology and Corporate Openness to Social Activism. Administrative Science Quarterly, 65(2), 524–563. Hawn, O., & Kang, H. (2018). The Effect of Market and Nonmarket Competition on Firm and Industry Corporate Social Responsibility. In S. Dorobantu, R.V. Aguilera, J. Luo, & F. J. Milliken (Eds.), Sustainability, Stakeholder Governance, and Corporate Social Responsibility (vol. 38, pp. 313–337). Emerald Publishing Limited. Hannan, M. T., & Freeman, J. (1984). Structural Inertia and Organizational Change. American Sociological Review, 49(2), 149–164. Helms, W. S., & Patterson, K. D. W. (2013). Eliciting Acceptance For “Illicit” Organizations: The Positive Implications of Stigma for MMA Organizations. Academy of Management Journal, 57(5), 1453–1484. Hersel, M., Helmuth, C., Zorn, M., Shropshire, C., & Ridge, J. (2019). The corrective actions organizations pursue following misconduct: A review and research agenda. Academy of Management Annals,13(2),547–585 Hrebiniak, L. G., & Joyce, W. F. (1985). Organizational Adaptation: Strategic Choice and Environmental Determinism. Administrative Science Quarterly, 30(3), 336–349. Hsu, G., & Grodal, S. (2020). The Double–edged Sword of Oppositional Category Positioning: A Study of the U.S. E–cigarette Category, 2007–2017. Administrative Science Quarterly, 1– 47. Jensen, M. C. (1986). Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American Economic Review, 76(2), 323–329. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. Johnson, V. (2007). What Is Organizational Imprinting? Cultural Entrepreneurship in the Founding of the Paris Opera. American Journal of Sociology, 113(1), 97–127.

Jones, T. M. (1995). Instrumental stakeholder theory: A synthesis of ethics and economics. Academy of Management Review, 20(2), 404–437. Joseph, J., Ocasio, W., & McDonnell, M.–H. (2014). The Structural Elaboration of Board Independence: Executive Power, Institutional Logics, and the Adoption of CEO–Only Board Structures in U.S. Corporate Governance. Academy of Management Journal, 57(6), 1834–1858. Kimberly, J. R. (1975). Environmental Constraints and Organizational Structure: A Comparative Analysis of Rehabilitation Organizations. Administrative Science Quarterly, 20(1), 1–9. King, B. G. (2008). A Political Mediation Model of Corporate Response to Social Movement Activism. Administrative Science Quarterly, 53(3), 395–421. King, B. G., & Soule, S. A. (2007). Social Movements as Extra–Institutional Entrepreneurs: The Effect of Protests on Stock Price Returns. Administrative Science Quarterly, 52(3), 413– 442. Kiron, D., Kruschwitz, N., Reeves, M., Haanaes, K., & Goh, E. (2015). The Benefits of Sustainability–Driven Innovation. In M. Deimler, R. Lesser, D. Rhodes, & J. Sinha (Eds.), Own the Future (pp. 119–123). John Wiley & Sons, Inc. KLD (2008). KLD Celebrates 20 Years of Socially Responsible Investing (SRI) Research Leadership. Klewitz, J., & Hansen, E. G. (2014). Sustainability–oriented innovation of SMEs: A systematic review. Journal of Cleaner Production, 65, 57–75. Kölbel, Julian F., Busch, T., & Jancso, L. M. (2017). How Media Coverage of Corporate Social Irresponsibility Increases Financial Risk. Strategic Management Journal, 38(11), 2266– 2284. Lev, B., Petrovits, C., & Radhakrishnan, S. (2010). Is doing good good for you? How corporate charitable contributions enhance revenue growth. Strategic Management Journal, 31(2), 182–200. Lim, A., & Tsutsui, K. (2012). Globalization and Commitment in Corporate Social Responsibility: Cross–National Analyses of Institutional and Political–Economy Effects. American Sociological Review, 77(1), 69–98. Linton, J. D., Klassen, R., & Jayaraman, V. (2007). Sustainable supply chains: An introduction. Journal of Operations Management, 25(6), 1075–1082. Luo, X. R., Zhang, J., & Marquis, C. (2016). Mobilization in the Internet Age: Internet Activism and Corporate Response. Academy of Management Journal, 59(6), 2045–2068. Lyon, T. P., & Montgomery, A. W. (2015). The Means and End of Greenwash. Organization & Environment, 28(2), 223–249. Marquis, C., & Huang, Z. (2010). Acquisitions As Exaptation: The Legacy of Founding Institutions in the U.S. Commercial Banking Industry. Academy of Management Journal, 53(6), 1441–1473. Marquis, C., & Tilcsik, A. (2013). Imprinting: Toward a Multilevel Theory. Academy of Management Annals, 7(1), 195–245. Marquis, C., & Qiao, K. (2020). Waking from Mao’s Dream: Communist Ideological Imprinting and the Internationalization of Entrepreneurial Ventures in China. Administrative Science Quarterly, 65(3), 795–830. McDonnell, M.–H., & King, B. (2013). Keeping up Appearances: Reputational Threat and Impression Management after Social Movement Boycotts. Administrative Science Quarterly, 58(3), 387–419.

McWilliams, A., & Siegel, D. (2001). Corporate Social Responsibility: A Theory of the Firm Perspective. Academy of Management Review, 26(1), 117–127. Melnyk, S. A., Sroufe, R. P., & Calantone, R. (2003). Assessing the impact of environmental management systems on corporate and environmental performance. Journal of Operations Management, 21(3), 329–351. Mena, S., Rintamäki, J., Fleming, P., & Spicer, A. (2015). On the Forgetting of Corporate Irresponsibility. Academy of Management Review, 41(4), 720–738. Metz, P., Burek, S., Hultgren, T. R., Kogan, S., & Schwartz, L. (2016). The Path to Sustainability– Driven Innovation. Research–Technology Management, 59(3), 50–61. Noda, T., & Collis, D. J. (2001). The Evolution of Intraindustry Firm Heterogeneity: Insights From a Process Study. Academy of Management Journal, 44(4), 897–925. Norton, L. (2020). Barron's Here's What ESG Investors Want to See From Companies During a Crisis. April 10, 2020. Oliver, C. (1991). Strategic Responses to Institutional Processes. Academy of Management Review, 16(1), 145–179. Pearson, C. M., & Clair, J. A. (1998). Reframing Crisis Management. Academy of Management Review, 23(1), 59–76. Peteraf, M. A. (1993). The cornerstones of competitive advantage: A resource–based view. Strategic Management Journal, 14(3), 179–191. Pfarrer, M. D., Decelles, K. A., Smith, K. G., & Taylor, M. S. 2008. After the fall: Reintegrating the corrupt organization. Academy of Management Review, 33(3), 730–749. Pfeffer, J., & Salancik, G. R. (1978). The External Control of Organizations: A Resource Dependence Perspective, Stanford Business Classics. Pollock, T. G., & Rindova, V. P. (2003). Media Legitimation Effects in the Market for Initial Public Offerings. Academy of Management Journal, 46(5), 631–642. Porter, M. E., 1980. Competitive Strategy. New York, NY: Free Press. Porter, M. E., & van der Linde, C. (1995). Toward a New Conception of the Environment– Competitiveness Relationship. Journal of Economic Perspectives, 9(4), 97–118. Ravasi, D., Rindova, V., Etter, M., & Cornelissen, J. (2018). The Formation of Organizational Reputation. Academy of Management Annals, 12(2), 574–599. Repenning, N. P., & Sterman, J. D. (2002). Capability Traps and Self–Confirming Attribution Errors in the Dynamics of Process Improvement. Administrative Science Quarterly, 47(2), 265–295. Rindova, V. P., & Fombrun, C. J. (1999). Constructing competitive advantage: The role of firm– constituent interactions. Strategic Management Journal, 20(8), 691–710. Russo, M. V., & Harrison, N. S. (2005). Organizational Design and Environmental Performance: Clues From the Electronics Industry. Academy of Management Journal, 48(4), 582–593. Scherer, A. G., Palazzo, G., & Seidl, D. (2013). Managing Legitimacy in Complex and Heterogeneous Environments: Sustainable Development in a Globalized World. Journal of Management Studies, 50(2), 259–284. Schultz, M., Mouritsen, J., & Gabrielsen, G. (2001). Sticky Reputation: Analyzing a Ranking System. Corporate Reputation Review, 4(1), 24–41. Shea, C. T., & Hawn, O. V. (2019). Microfoundations of Corporate Social Responsibility and Irresponsibility. Academy of Management Journal, 62(5), 1609–1642. Shrivastava, P., Mitroff, I. I., Miller, D., & Miclani, A. (1988). Understanding Industrial Crises. Journal of Management Studies, 25(4), 285–303.

Simsek, Z., Fox, B. C., & Heavey, C. (2015). “What’s Past Is Prologue”: A Framework, Review, and Future Directions for Organizational Research on Imprinting. Journal of Management, 41(1), 288–317. Sydow, J., Schreyögg, G., & Koch, J. (2009). Organizational Path Dependence: Opening the Black Box. Academy of Management Review, 34(4), 689–709. Tilcsik, A., & Marquis, C. (2013). Punctuated Generosity: How Mega-events and Natural Disasters Affect Corporate Philanthropy in U.S. Communities. Administrative Science Quarterly, 58(1), 111–148. Tilcsik, A., & Marquis, C. (2014). Imprinting. In Oxford Bibliographies in Management. Ed. Ricky Griffin. New York: Oxford University Press. Turban, D. B., & Greening, D. W. (1997). Corporate Social Performance and Organizational Attractiveness to Prospective Employees. Academy of Management Journal, 40(3), 658– 672. Tushman, M. L., & Romanelli, E. (1985). Organizational evolution: A metamorphosis model of convergence and reorientation. Research in Organizational Behavior, 7, 171–222. Greenwich, CT: JAI Press. Waage, S. A. (2007). Re–considering product design: A practical “road–map” for integration of sustainability issues. Journal of Cleaner Production, 15(7), 638–649. Waldron, T. L., Navis, C., Aronson, O., York, J. G., & Pacheco, D. F. (2018). Values-Based Rivalry: A Theoretical Framework of Rivalry Between Activists and Firms. Academy of Management Review, 44(4), 800–818. Waldron, T. L., Navis, C., & Fisher, G. (2013). Explaining Differences in Firms’ Responses to Activism. Academy of Management Review, 38(3), 397–417. Wei, J., Ouyang, Z., & Chen, H. (Allan). (2017). Well Known or Well Liked? The Effects of Corporate Reputation on Firm Value at the Onset of a Corporate Crisis. Strategic Management Journal, 38(10), 2103–2120. Zavyalova, A., Pfarrer, M. D., Reger, R. K., & Shapiro, D. L. (2012). Managing the Message: The Effects of Firm Actions and Industry Spillovers on Media Coverage Following Wrongdoing. Academy of Management Journal, 55(5), 1079–1101. Zhao, E. Y., Fisher, G., Lounsbury, M., & Miller, D. (2017). Optimal distinctiveness: Broadening the interface between institutional theory and strategic management. Strategic Management Journal, 38(1), 93–113.

TABLE 1: SUMARY STATISTICS OF 3-YEAR AVERAGE FIRM CHARACTERISTICS BETWEEN POSITIVELY-

IMPRINTED AND NEGATIVELY-IMPRINTED REPUTATIONS DURING THE GREAT RECESSION (2007-2009)

Reputational 25th 50th 75th Variable imprints Mean percentile percentile percentile Std. Dev. Min Max ESG incidents Positive 2.116 3 3 3 0.036 3 5 Negative 15.051 5 7 14 32.104 3.33 650 ESG practices Positive 5.050 4.114 5.085 5.964 1.360 1.239 9.122 Negative 5.266 4.237 5.255 6.269 1.455 0.889 9.609 Log(market capitalization) Positive 21.105 20.190 21.174 22.068 1.431 14.051 26.593 Negative 22.409 21.395 22.422 23.509 1.593 14.465 26.946 Return on Asset (percent) Positive 5.493 1.901 5.484 9.854 12.963 –36.704 36.738 Negative 5.783 1.874 5.258 9.501 9.910 –25.702 34.843 Total debts to total assets (percent) Positive 25.022 9.647 23.338 36.695 0.192 0 76.30 Negative 25.881 13.359 24.236 35.736 0.173 0 72.63 Capital Expenditure to Sales (percent) Positive 13.053 2.126 4.392 10.516 32.062 0 816.443 Negative 12.219 2.734 5.327 11.903 27.263 0 623.228 Leverage (percent) Positive 25.045 10.053 23.371 36.467 0.193 0 75.353 Negative 25.375 13.739 24.254 35.111 0.160 0 68.658

TABLE 2: SUMMARY STATISTICS AND CORRELATIONS

ESG incidents refer to log(1+total number of severity–weighted ESG incidents of firm I in year t) from RepRisk. Firm size is measured by Ln(market capitalization), which is the natural log of the market value of a company’s outstanding shares. Ln(age) is the natural log of number of years since firm inception. ROA is EBITDA divided by total assets. Leverage is the ratio of total debt to shareholders’ equity. Firm risk is the standard deviation of daily stock prices for each firm and each year. Industry competition is measured by the Herfindahl-Hirschman index (HHI), which is the sum of squared market share of firms competing in the industry. Country-level variables include country GDP per capita (logged), country wellbeing as measured by years of life expectancy, and country CO2 emissions per capita (metric tons per capita).

Standard Variable Mean deviation 1 2 3 4 5 6 7 8 1 ESG incidents 1.530 1.294 1.000 2 ESG practices 4.544 1.273 0.031 1.000 3 Ln(market capitalization) 21.879 1.541 0.484 0.171 1.000 4 ROA 5.462 11.679 -0.001 0.028 0.236 1.000 5 Ln(age) 3.223 1.085 0.065 0.115 0.097 -0.019 1.000 6 Slack resource 0.470 0.499 -0.079 -0.096 -0.075 0.062 -0.060 1.000 7 Total debts to total assets 0.253 0.262 0.049 -0.039 0.006 -0.171 -0.019 -0.184 1.000 8 Firm risk 28.680 9.738 -0.087 -0.177 -0.399 -0.160 -0.221 0.059 -0.028 1.000 9 HHI 0.007 0.009 -0.003 -0.020 -0.002 0.034 0.060 0.010 0.038 0.006 Country-level variables 10 GDP per capita (in log) 10.419 0.899 0.002 0.156 0.000 -0.088 -0.012 0.042 0.004 -0.080 11 Life expectancy 78.779 4.543 0.001 0.187 0.037 -0.125 0.052 -0.061 0.008 -0.096 12 CO2 emissions 11.622 5.494 0.004 -0.113 -0.043 -0.031 -0.068 0.144 0.005 0.047 (metric tons per capita) -0.063 -0.103 -0.024 -0.796 0.082 0.037 0.152 Founding GDP per capita 13 8.490 1.746 -0.064 -0.050 -0.084 0.004 -0.722 0.108 0.019 0.087 (in log) 14 Founding life expectancy 71.270 7.541 -0.051 -0.125 -0.065 0.009 -0.282 0.167 0.008 0.060 15 Founding CO2 emissions 11.165 7.227 -0.026 0.156 0.000 -0.088 -0.012 0.042 0.004 -0.080 (metric tons per capita)

Variables 9 10 11 12 13 14 15 9 HHI 1.000 10 GDP per capita (in log) -0.004 1 49

11 Life expectancy 0.028 0.713 1 CO2 emissions -0.019 0.574 0.098 1 12 (metric tons per capita) Founding GDP per capita 13 (in log) -0.057 0.370 0.132 0.286 1 14 Founding life expectancy -0.054 0.694 0.588 0.356 0.833 1 15 Founding CO2 emissions -0.053 0.543 -0.008 0.843 0.713 0.509 1

TABLE 3: POSITIVE and NEGATIVE FEEDBACK LOOPS (FULL SAMPLE)

This table reports panel vector autoregressions (2 lags) with GMM estimation over the years 2010 to 2018. ESG practices and ESG incidents are endogenous variables; ln(market capitalization), ROA and ln(age) are exogenous variables. ESG practices refer to the overall ESG rating from the MSCI IVA. ESG incidents refer to log(1+total number of severity–weighted ESG incidents) from RepRisk. The robust p-value in parentheses.

Positively Imprinted Firms Negatively Imprinted Firms (1) (2) (3) (4) ESG practices ESG incidents ESG practices ESG incidents ESG practicest–1 0.687*** -0.199*** 0.694*** -0.240*** (0) (0.002) (0) (0) ESG practicest–2 -0.039*** -0.0766** -0.0364** -0.0772 (0.003) (0.039) (0.038) (0.101) ESG incidentst–1 0.028** 0.155*** -0.295*** 1.058*** (0.027) (0.000) (0) (0) ESG incidentst–2 0.028*** 0.0758*** -0.154*** 0.618*** (0.001) (0.009) (0.001) (0) Ln(market capitalization) 0.742*** -3.540*** 1.252*** -4.417*** (0.002) (0) (0.004) (0) ROA -0.0004 0.005 -0.00859* 0.0337** (0.821) (0.395) (0.087) (0.017) Ln(age) 0.076 -0.331 0.433*** -1.427*** (0.342) (0.142) (0) (0) Observations 8,703 8,703 7,032 7,032 p-value in parentheses; *** p<0.01, ** p<0.05, * p<0.1

TABLE 4: POSITVE AND NEGATIVE FEEDBACK LOOPS (MATCHED SAMPLE)

This table reports panel vector autoregressions (2 lags) with GMM estimation over the years 2010 to 2018. Negatively imprinted firms are treated firms, which were highly criticized during 2007-2009 global crisis. We match treated with control firms by using coarsened exact matching (CEM) on previous ESG practices, profitability, slack resources, size, risks or annual stock price volatility, age (all at t-1), country and industry. In PVAR model, ESG practices and ESG incidents are endogenous variables; ln(market capitalization), ROA and ln(age) are exogenous variables. ESG practices refer to the overall ESG rating from the MSCI IVA. ESG incidents refer to log(1+total number of severity–weighted ESG incidents) from RepRisk. The robust p-value in parentheses.

Positively imprinted firms Negatively imprinted firms (matched sample) (matched sample) (1) (2) (3) (4) ESG practices ESG incidents ESG practices ESG incidents ESG practicest–1 0.678*** -0.055 0.661*** -0.125** (0) (0.179) (0) (0.012) ESG practicest–2 -0.038*** -0.046* -0.038*** -0.037 (0.001) (0.060) (0.004) (0.173) ESG incidentst–1 0.016 0.194*** -0.124*** 0.575*** (0.159) (0) (0) (0) ESG incidentst–2 0.026*** 0.066*** -0.046** 0.283*** (0.001) (0.004) (0.027) (0) Ln(market 0.303*** -1.306*** 0.310** -1.426*** capitalization)t–1 (0.001) (0) (0.010) (0)

ROAt–1 -0.002 0.004 -0.005 0.023*** (0.463) (0.517) (0.134) (0.003) Ln(age)t–1 -0.031 -0.223 0.223** -0.464** (0.740) (0.270) (0.019) (0.017) Observations 7,446 7,446 5,879 5,879 pval in parentheses; *** p<0.01, ** p<0.05, * p<0.1

TABLE 5: SPIRAL EFFECTS BETWEEN MATACHED POSITIVELY IMPRINTED AND NEGATIVELY IMPRINTED FIRMS ESTIMATED FROM THE IMPULSE RESPONSE FUNCTIONS Positively imprinted firms Negatively imprinted firms Impulse variables Impulse variables Response Forecast ESG ESG Response Forecast ESG ESG variable horizon incidents practices variable horizon incidents practices ESG 0 1 0 ESG 0 1 0 incidents 1 0.194 -0.055 incidents 1 0.575 -0.125 2 0.103 -0.094 2 0.629 -0.191 3 0.030 -0.077 3 0.553 -0.221 4 0.009 -0.055 4 0.532 -0.231 5 0.001 -0.036 5 0.501 -0.231 6 -0.001 -0.023 6 0.478 -0.226 7 -0.001 -0.014 7 0.455 -0.219 8 -0.001 -0.009 8 0.434 -0.210 9 -0.001 -0.005 9 0.414 -0.202 10 -0.001 -0.003 10 0.396 -0.193 … ….... ….... 35 0.125 -0.061 36 0.119 -0.058 37 0.114 -0.056 38 0.109 -0.053 39 0.104 -0.051 40 0.099 -0.049 Response Forecast ESG ESG Response Forecast ESG ESG variable horizon incidents practices variable horizon incidents practices ESG 0 0 1 ESG 0 0 1 practices 1 0.016 0.678 practices 1 -0.124 0.661 2 0.039 0.421 2 -0.199 0.414 3 0.033 0.257 3 -0.231 0.278 4 0.024 0.155 4 -0.242 0.204 5 0.016 0.092 5 -0.243 0.163 6 0.010 0.055 6 -0.238 0.139 7 0.006 0.032 7 -0.230 0.124 8 0.004 0.019 8 -0.221 0.114 9 0.002 0.011 9 -0.212 0.107 10 0.001 0.007 10 -0.203 0.101 … ….... ….... 35 -0.064 0.031 36 -0.061 0.030 37 -0.059 0.029 38 -0.056 0.027 39 -0.054 0.026 40 -0.051 0.025

TABLE 6: COMPETITIVENESS AND FIRMS’ DIVERGENT RESPONSES

This table reports panel vector autoregressions (2 lags) with GMM estimation over the years 2010 to 2018. ESG practices and ESG incidents are endogenous variables; firm size as measured by log(market capitalization), ROA and firm age are exogenous variables. ESG practices refer to the overall ESG rating from the MSCI IVA. ESG incidents refer to log(1+total number of severity–weighted ESG incidents) from RepRisk. Competitive industry are firms in the industries that have HHI index below the median, and concentrated industry are firms in the industries with HHI index above the sample median. The robust p–values are in parentheses. Positively imprinted firms Negatively imprinted firms Competitive Concentrated Competitive Concentrated industry industry industry industry (1) (2) (3) (4) (5) (6) (7) (8) ESG ESG ESG ESG ESG ESG ESG ESG practices incidents practices incidents practices incidents practices incidents ESG practicest–1 0.636*** -0.0738 0.669*** -0.006 0.667*** -0.106* 0.654*** -0.087 (0) (0.158) (0) (0.886) (0) (0.086) (0) (0.170) ESG practicest–2 -0.042*** -0.044 -0.058*** -0.049 -0.064*** -0.020 -0.040** -0.031 (0.008) (0.182) (0) (0.107) (0) (0.579) (0.024) (0.444) ESG incidentst–1 0.012 0.186*** 0.016 0.214*** -0.223*** 0.649*** -0.180** 0.924*** (0.406) (0) (0.275) (0) (0) (0) (0.013) (0) ESG incidentst–2 0.036*** 0.076*** 0.018* 0.071** -0.076** 0.332*** -0.094** 0.500*** (0) (0.009) (0.086) (0.015) (0.011) (0) (0.038) (0) Ln(market 0.218** -1.277*** 0.338*** -1.296*** 0.452** -1.764*** 0.545** -2.029*** capitalization)t–1 (0.036) (0) (0.002) (0) (0.014) (0) (0.011) (0) ROAt–1 -0.001 0.002 -0.004 0.004 0.001 0.0133 -0.014*** 0.032*** (0.850) (0.753) (0.126) (0.382) (0.909) (0.104) (0.005) (0.005) Ln(age)t–1 0.0727 -0.269 -0.0140 0.0473 0.173 -0.366 0.229* -0.994*** (0.393) (0.177) (0.890) (0.818) (0.149) (0.148) (0.063) (0) Observations 4,475 4,475 4,485 4,485 3,563 3,563 3,578 3,578 p-value in parentheses; *** p<0.01, ** p<0.05, * p<0.1

TABLE 7 FOUNDING PERIODS

Panel A: Positively imprinted firms Founding years 1516-1970 1971-1990 1991-2007 (1) (2) (3) (4) (5) (6) ESG practices ESG incidents ESG practices ESG incidents ESG practices ESG incidents ESG practicest–1 0.681*** -0.052 0.709*** -0.026 0.617*** 0.023 (0) (0.306) (0) (0.721) (0) (0.684) ESG practicest–2 -0.043** -0.107*** -0.015 -0.015 -0.066*** -0.008 (0.021) (0.002) (0.645) (0.718) (0) (0.817) ESG incidentst–1 0.029 0.217*** 0.032 0.236*** -0.005 0.182*** (0.133) (0) (0.289) (0) (0.748) (0) ESG incidentst–2 0.024* 0.049 0.0419** 0.096** 0.0192* 0.086*** (0.075) (0.141) (0.045) (0.022) (0.099) (0.007) Ln(market 0.526 -1.050 1.063** -0.092 0.486*** -1.436*** capitalization)t–1 (0.110) (0.108) (0.013) (0.891) (0.002) (0) ROAt–1 -0.018*** 0.008 -0.006 -0.009 -0.001 0.003 (0.002) (0.439) (0.234) (0.190) (0.695) (0.610) Ln(age)t–1 -0.832 -1.857 -1.718* -1.967 -0.024 -0.065 (0.452) (0.384) (0.057) (0.125) (0.726) (0.683) Observations 3,179 3,179 2,162 2,162 3,619 3,619 Panel B: Negatively imprinted firms Founding years 1516-1970 1971-1990 1991-2007 (1) (2) (3) (4) (5) (6) ESG practices ESG incidents ESG practices ESG incidents ESG practices ESG incidents ESG practicest–1 0.609*** -0.076 0.687*** -0.105 0.734*** -0.209** (0) (0.111) (0) (0.172) (0) (0.024) ESG practicest–2 -0.009 0.032 -0.080*** -0.151*** -0.071*** -0.082 (0.613) (0.280) (0.001) (0.001) (0.007) (0.112) ESG incidentst–1 -0.135*** 0.521*** -0.125** 0.501*** -0.209*** 0.694*** (0.002) (0) (0.043) (0) (0.002) (0) ESG incidentst–2 -0.067** 0.313*** -0.017 0.215*** -0.090** 0.359*** (0.023) (0) (0.634) (0.001) (0.031) (0) Ln(market -0.212 0.312 -0.132 -0.204 1.238*** -2.301*** capitalization)t–1 (0.345) (0.498) (0.799) (0.824) (0) (0.001) ROAt–1 -0.003 0.002 -0.002 0.012 -0.009 0.0196 (0.446) (0.841) (0.825) (0.400) (0.192) (0.136) Ln(age)t–1 1.850*** -5.428*** 0.866 -2.051* -0.037 -0.403** (0.006) (0) (0.252) (0.078) (0.696) (0.027) Observations 3,259 3,259 1,556 1,556 2,326 2,326 pval in parentheses; *** p<0.01, ** p<0.05, * p<0.1

TABLE 8 FOUNDING COUNTRY CO2 EMISSIONS PER CAPITA

Positively imprinted firms Negatively imprinted firms CO2 emissions less CO2 emissions more CO2 emissions less CO2 emissions more than 12 metric tons than or equal to12 than 12 metric tons than or equal to12 per capita metric tons per capita per capita metric tons per capita (1) (2) (3) (4) (5) (6) (7) (8) ESG ESG ESG ESG ESG ESG ESG ESG practices incidents practices incidents practices incidents practices incidents ESG practicest–1 0.701*** -0.018 0.595*** -0.064 0.682*** -0.250* 0.644*** -0.073 (0) (0.719) (0) (0.179) (0) (0.076) (0) (0.144) ESG practicest–2 -0.074*** -0.061* -0.025 -0.035 -0.031 -0.073 -0.058*** -0.073** (0) (0.057) (0.142) (0.254) (0.132) (0.386) (0.003) (0.025) ESG incidentst–1 0.0187 0.165*** 0.008 0.242*** -0.140* 1.156*** -0.281*** 0.638*** (0.189) (0) (0.591) (0) (0.096) (0.001) (0) (0) ESG incidentst–2 0.024** 0.043 0.026** 0.105*** -0.076 0.665*** -0.105*** 0.300*** (0.017) (0.142) (0.010) (0.001) (0.124) (0.002) (0.004) (0) Ln(market 0.227* -1.654*** 0.412*** -0.920*** 0.817 -6.192** 0.591*** -1.024*** capitalization)t–1 (0.056) (0) (0) (0) (0.168) (0.015) (0) (0) ROAt–1 -0.008** 0.021*** 0.001 -0.005 -0.008 0.058* -0.007* 0.014** (0.024) (0.005) (0.764) (0.377) (0.293) (0.057) (0.054) (0.032) Ln(age)t–1 -0.011 0.121 -0.017 -0.326* 0.327** -2.191*** 0.143 -0.569** (0.915) (0.586) (0.845) (0.072) (0.049) (0.002) (0.308) (0.019) Observations 4,525 4,525 4,435 4,435 3,831 3,831 3,269 3,269 pval in parentheses; *** p<0.01, ** p<0.05, * p<0.1

TABLE 9

CEM RESULTS OF DIVERGENT RESPONSES TO ESG CRITICISM DURTING 2010- 2018 BETWEEN FIRMS WITH POSITVE AND NEGATIVE REPUTATIONS IMPRINTED BY THE GLOBAL FINANCIAL CRISIS This table reports CEM matching analysis in panel regressions with year, industry and country fixed effects over the years 2010 to 2018. ESG practices refer to the overall ESG rating from the MSCI IVA. ESG incidents refer to log(1+total number of severity–weighted ESG incidents) from RepRisk. Control variables include ROA, firm age, slack resources, total debts to total assets and firm risk. All controls are lagged by one year. The robust p– values are in parentheses

(1) ESG practices ESG incidents 0.018** (0.014) Negative reputation imprinted by financial crisis 0.058* (0.098) ESG incidents x Negative reputation imprinted by -0.055*** financial crisis (0)

Founding CO2 emissions per capita -0.002 (0.629) Founding CO2 emissions per capita x Negative 0.002 reputation imprinted by financial crisis (0.330)

Control variables Yes Constant 5.425*** (0) Year fixed effects Yes Industry fixed effects Yes Country fixed effects Yes Observations 23,467 R-squared 0.337 pval in parentheses; *** p<0.01, ** p<0.05, * p<0.1

FIGURE 1A: VIRTUOUS CYCLE Less extensively criticized firms “race to the top”

Higher ESG practices ESG incidents

+ Competitive pressures

FIGURE 1B: VICIOUS CYCLE More extensively criticized firms “race to the bottom”

Lower ESG practices ESG incidents

Competitive pressures

FIGURE 2A: IMPULSE RESPONSE FUNCTION OF FIRMS WITH POSITIVE REPUTATIONAL IMPRINTS

ESG_incidents : ESG_incidents ESG_incidents : ESG_practices 1 .04

.03

.5 .02

.01

0 0

ESG_practices : ESG_incidents ESG_practices : ESG_practices IRF 0 1

-.05 .5

-.1 0 0 5 10 0 5 10 step impulse : response

FIGURE 2B: IMPULSE RESPONSE FUNCTION OF FIRMS WITH NEGATIVE REPUTATIONAL IMPRINTS

ESG_incidents : ESG_incidents ESG_incidents : ESG_practices 1 0

-.1

-.2

0 -.3

ESG_practices : ESG_incidents ESG_practices : ESG_practices IRF 0 1

-.1

-.2

-.3 0 0 10 20 30 40 0 10 20 30 40 step impulse : response