Carrot and Stick? The Role of Financial Market Intermediaries in Corporate Social Performance

Abstract

This article examines the role of intermediaries in financial markets in fostering corporate sustainability. Responsible investment (RI) indices have been primarily identified as intermediaries that provide information regarding corporate social performance (CSP) for investors and other stakeholders. The authors argue that the role of these intermediaries is not confined solely to information provision, but they may also incentivize high levels of CSP through mechanisms such as exclusion threats, signaling, and engagement. The authors rely on unique access to the archives of the FTSE4Good Index to examine the effects of these mechanisms on CSP. The study shows that companies facing exclusion threats and signaling are more likely to comply with the intermediary’s criteria, and medium levels of engagement leads to higher levels of CSP. The authors contribute to the study of sustainability in financial markets by explicating the mechanisms that intermediaries and other financial actors could employ to foster greater corporate sustainability.

Keywords

engagement RI indices corporate social performance (CSP)intermediaries

Intermediaries play an increasingly important role for equity market participants interested in sustainability and business responsibility. Responsible investment (RI), defined here as the integration of investors’ financial objectives with an evaluation of corporate environmental, social, and governance concerns, relies on intermediaries to measure the responsible behavior of companies. One popular group of metrics are RI indices, which are essentially weighted listings of stocks that are typically constructed by filtering a broader universe of stocks according to a set of corporate social performance (CSP) criteria. RI indices, such as the Dow Jones Sustainability Index (DJSI), the Domini Social Index, and the FTSE4Good Index, are used by investors for the benchmarking of RI funds and the creation of related investment products. They may also serve to signal CSP reputation to investors and other stakeholders (Doh, Howton, Howton, & Siegel, 2010), as these indices allow them to make a judgment of the extra-financial “quality” of corporate stocks. Most of the studies regarding RI indices have focused on the effect of inclusion or exclusion from indices on corporate share prices (see Capelle-Blancard & Couderc, 2009; Collison, Cobb, Power, & Stevenson, 2008; Consolandi, Jaiswal-Dale, Poggiani, & Vercelli, 2009; Curran & Moran, 2007; Robinson, Kleffner, & Bertels, 2011; Cheung, 2011). However, the question remains whether these intermediaries may also have an effect on CSP, and if so, what the key mechanisms are to bring about these effects?

We argue that the role of intermediaries, such as RI indices, is not limited solely to information provision, but that they may also contribute to higher levels of CSP in a context where standards for CSP continue to evolve (Bertels & Peloza, 2008). We identify three mechanisms that may move companies toward high CSP over time: threatening exclusion from the index when CSP scores do not meet the benchmark, signaling CSP reputation through publication of index inclusion, and constructing effective dialogue on CSP through engagement. We show that these mechanisms can be effectively used to influence the CSP of large numbers of companies over time. We study the FTSE4Good Index, launched in 2001 by the U.K.-based index provider FTSE Group to evaluate the environmental and social performance of companies listed on the major stock exchanges around the world. Through unique access to the archives and data related to the FTSE4Good Index, we study the introduction of a new set of CSP criteria on anti-bribery and corruption practices. We highlight how exclusion threats, signaling, and engagement influence levels of CSP. In so doing, we contribute to reputation and legitimacy studies by showing that the role of intermediaries may involve more than pure information provision to investors and stakeholders. We also contribute to the theoretical development of studies on shareholder engagement and activism, by explicating the mechanisms that influence CSP. Very little is known about the effectiveness of engagement, especially on a larger scale (Gond & Piani, 2013). We provide what we believe is one of the first large-N studies of the effect of engagement on corporate behavior. Based on our data set, we show that engagement can be effective in influencing levels of CSP even when applied to large numbers of companies. Our results show how engagement interacts with corporate reputation and legitimacy in fostering compliance with intermediaries CSP criteria and may lead to higher levels of CSP. The article is structured below as follows. First, we review the role of intermediaries in conferring corporate reputation and legitimacy that has been previously identified in the literature. Next, we build hypotheses related to the three mechanisms for influencing CSP, and describe our methods for testing these hypotheses in the case of the FTSE4Good Index. We then present the results, and conclude with a discussion of the implications of our findings for the study of intermediaries, CSP, and engagement.

Intermediaries as Information Providers

Intermediaries play an important role in markets that are characterized by information asymmetries. Studies on a wide range of intermediaries such as certification bodies (Rao, 1994, 1998), media rankings (Rindova, Williamson, Petkova, & Sever, 2005), and financial analysts (Benner & Ranganathan, 2012; Zuckerman, 2000), show that where there is a lack of concrete information about aspects of firm performance, or performance is difficult for the general public to observe, evaluation by intermediaries provides legitimacy and reputation signals that might in turn influence financial performance. Fombrun and Shanley (1990) find evidence that CSP forms a subset of the signals that corporate reputation is based on, but due to information asymmetries, stakeholders are reliant on intermediaries to evaluate CSP. In a similar vein, Doh et al. (2010) argue that aspects of CSP are difficult to observe by an organization’s stakeholders, and that RI indices act as “institutional intermediaries,” which “by including (or excluding) firms from their indices . . . send clear and strong signals to investors about whether firms have met the credible CSR criteria established by these organizations” (Doh et al., 2010, p. 1466).

Research into the main RI indices in the RI market has focused mainly on the financial performance of these indices (Collison et al., 2008) or the effect of inclusion and exclusion from the indices on share price (Capelle-Blancard & Couderc, 2009; Doh et al., 2010; Robinson et al., 2011; Cheung, 2011). These studies have not examined whether inclusion in RI indices can have effects on CSP. There is increasing evidence that the measurement by intermediaries is not neutral, but may act to incentivize improvements in what is being measured, especially when reputational stakes are high (Sauder, 2006, 2008; Sauder & Espeland, 2009). Studies in the context of higher education show that obtaining a good evaluation by intermediaries in rankings and ratings becomes an important part of maintaining organizational reputation, and that many organizations will adjust their behavior to obtain a favorable evaluation (Sauder 2006, 2008; Sauder & Espeland, 2009). We argue that the signal for reputation and legitimacy provided by RI indices may similarly influence levels of CSP. Researching these effects will provide better understanding of the role of intermediaries in fostering sustainability, and responds to the call for research that examines CSP as the dependent variable (Waddock & Graves, 1997).

A potential explanation for the lack of understanding about intermediaries’ role in influencing CSP may derive from the fact that most studies focus on the informational role of intermediaries only. In essence, these studies take the outcome of the evaluation process undertaken by the intermediary as given and study the effect of the outcome on share price. In doing so, important processes and mechanisms that might spur higher levels of CSP are missed. For example, our study shows that intermediaries engage with corporations to gather CSP information and provide advice about the evaluation criteria. This engagement tries to convince companies to aim for higher levels of CSP, just like shareholder engagement (Clark & Hebb, 2004; Southwood, 2003) or shareholder dialogue (Logsdon & Van Buren, 2009) aims to do. Because shareholder engagement often takes place in private conversations between corporate management and the engaging party, it is difficult to measure its effects on a large scale. Based on the unique data set that we gathered for this study of the FTSE4Good Index, which includes correspondence between the two parties in engagement, we can show that engagement is an important mechanism that can be used by intermediaries to influence levels of CSP. In the next section, we build our theoretical framework explicating the mechanisms that may influence CSP through intermediaries in RI markets.

Three Mechanisms for Influencing CSP

Drawing from and extending previous literatures, we identify three mechanisms underlying RI indices that may influence levels of CSP: the threat of index exclusion (labeled exclusion threat), the signaling of CSP reputation through the publication of index inclusion (labeled signaling), and the establishment of an engagement dialogue regarding required CSP levels (labeled engagement). We discuss each in turn, but first distinguish between two key concepts in our framework: legitimacy and reputation. In the literature on corporate reputation, the economic perspective emphasizes individual firm attributes such as efficiency or financial performance, whereas the institutional perspective focuses on a firm’s relative status among its peers (Doh et al., 2010; Love & Kraatz, 2009; Rindova et al., 2005). Legitimacy, a concept of central concern in institutional theory, can be seen as a precursor to reputation (Deephouse & Carter, 2005; Doh et al., 2010; Rindova et al., 2005). From this perspective, legitimacy is derived from the adherence to commonly accepted standards and norms of corporate behavior in an organizational field (Deephouse & Suchman, 2008; Scott, 2001; Suchman, 1995). Corporate reputation may be derived from adherence to the same norms and standards that define legitimacy, but additionally is derived from relative performance against the firm’s peers in that field (Deephouse & Carter, 2005; King & Whetten, 2008). Thus, the evaluation of adherence to commonly accepted norms and standards plays a central role in both conferring legitimacy and signaling reputation (Deephouse & Suchman, 2008). The conferring of legitimacy through index inclusion is the most widely acknowledged role of RI indices. For example, Consolandi et al. (2009) argue that inclusion in RI indices confers legitimacy as it shows compliance with global CSP standards to investors and stakeholders of the company. RI indices can be used by investors for identifying target companies for engagement, especially with those companies that are not included (Oulton, 2006). In addition, the indices have also become a tool for a wider group of actors within the CSP industry. Non-governmental organizations (NGOs) use them as a tool to identify “good” companies to partner with or “bad” companies to campaign against, whereas consultants may identify excluded companies as profitable potential clients (Slager, Gond, & Moon, 2012).

The conferring of legitimacy is a powerful incentive to improve CSP for those companies that do not meet the mark set by the intermediary (Consolandi et al., 2009). Chatterji and Toffel (2010) find that firms will improve their performance in response to poor ratings from RI intermediaries, to mitigate the threat of stakeholder sanctions. Yet research suggests that the threat of taking away this legitimacy, once conferred, may provide even stronger incentives for change. Doh et al. (2010) find that the effect on share price is limited to those firms that are deleted from the Calvert Social Index, and attribute this to asymmetric information on poor CSP performance. They argue that because of this asymmetry, deletion will come as more of a surprise to investors than inclusion (Doh et al., 2010). In line with this argument, Scalet and Kelly (2009) find that most firms do not communicate about negative events that lead to an exclusion from ratings. The idea that bad news attracts more attention than good news may also underlie the fear of being excluded from RI indices reported by managers in a survey of the FTSE4Good Index (Collison et al., 2009).

Most RI indices change their criteria for good CSP over time, to take into account current developments and changing global standards relating to environmental, social, and governance performance (Bertels & Peloza, 2008). For example, both the FTSE4Good Index and the DJSI have added assessment criteria for issues such as climate change and supply chain standards in the last decade (FTSE, 2011; RobecoSAM, 2013). This addition means companies that may have previously met the intermediaries’ criteria for good CSP need to improve their CSP to ensure their continued conformance. This process of raising the bar for index inclusion entails a threat to those companies that in the past met the CSP criteria, but now face exclusion. Following the logic that exclusion has stronger effects than inclusion, we propose that companies that face this threat are more likely to work toward compliance with the intermediaries’ CSP criteria to avoid the threat of exclusion:

Hypothesis 1 (H1): Those companies that are threatened by exclusion from RI indices are more likely to comply with the intermediary’s general CSP criteria for index inclusion in subsequent years.

The inclusion of a firm in a RI index may act as the endorsement (Rhee & Valdez, 2009) of that firm’s CSP because the evaluation by RI intermediaries is perceived to be objective and in accordance with global standards (Doh et al., 2010). RI indices provide “a normative benchmark for firms seeking to achieve a positive reputation for corporate social responsibility and [ . . . ] a guidepost for audiences concerned about the socially responsible practices of companies they invest in” (Doh et al., 2010, p. 1470). Although often ignored in studies on intermediaries, it is important to acknowledge the active role of firms themselves in signaling reputation through using intermediaries’ endorsements (Deephouse & Suchman, 2008). Although inclusion in the indices itself could be seen as signal of meeting norms, firms often enhance this signaling effect by communicating inclusion in indices in corporate materials. For example, many firms use the logos of RI indices in their CSP disclosure as a certification of good CSP (Consolandi et al., 2009), and the legitimacy of these signals are thus co-constructed between firms and intermediaries (Durand & McGuire, 2005). These firms use the assessment by RI intermediaries and resultant inclusion in RI indices as a “stamp of approval” for CSP performance, and, in the absence of other global standardized assessment measures of CSP, this stamp is often valorized as such by other parties, such as NGOs, the media, and consultants (Slager et al., 2012). Research on the use of third-party assessments in the field of higher education shows that their ubiquitous nature in organizational communications increases their reputation effects (Sauder, 2006, 2008). Nevertheless, not all companies will recognize and use index inclusion to signal good CSP performance. Given that companies may choose to use intermediary endorsements for reputation signaling purposes, we expect that companies that use RI indices to signal CSP reputation will ensure their continued conformance with the standards set by RI indices, as their failure to do so will lead to reputational damage. In addition, we expect that these companies have higher levels of CSP in subsequent years:

Hypothesis 2a (H2a): Those companies that signal CSP reputation to their stakeholders through intermediary endorsements are more likely to comply with the intermediary’s general CSP criteria for index inclusion in subsequent years.

Hypothesis 2b (H2b): Those companies that signal CSP reputation to their stakeholders through intermediary endorsements are more likely to have higher levels of CSP in subsequent years.

Shareholder engagement is becoming increasingly popular as a RI strategy in Europe (Clark & Hebb, 2004; Southwood, 2003) and the United States, where it has been termed “Dialogue” to distinguish it from more aggressive shareholder activism (Logsdon & Van Buren, 2009). Due to the private nature of shareholder engagement, which often takes place behind the closed doors of the company boardroom, little is known about the organizational context and process of engagement (Gond & Piani, 2013). From existing qualitative, small N studies, we can distil four factors that are common in shareholder engagement processes, and which, as we will show below, also feature in engagement undertaken by intermediaries such as FTSE in relation to their FTSE4Good Index. First, in their dialogue with corporate management, shareholders often frame their concerns regarding poor CSP in terms of reputational risk and potential impact on long-term financial performance (Gond & Piani, 2013). Second, there is an emphasis on building trust in the dialogue with corporate management, with antagonistic tactics generally being eschewed (McLaren, 2004; Vandekerckhove, Leys, & Van Braeckel, 2007). Third, the goals of the dialogue may simply be information exchange on the issue of concern or include more extensive influencing of corporate policies or reporting that are drawn up in response (Gifford, 2010; Gond & Piani, 2013). Finally, dialogues are often drawn out over extensive periods of time, and the threat of more aggressive tactics such as filing shareholder resolutions or selling of shares is often only used implicitly (Gifford, 2010; Logsdon & Van Buren, 2009). Sullivan and Mackenzie (2008) argue that objectives of this private dialogue with shareholders may be to raise the profile of CSP with corporate management, legitimize debate about contentious issues, and encourage management to meet certain CSP standards. Logsdon and Van Buren (2009) highlight that, compared with shareholder activism, effective engagement requires different skills and tactics, including patience, effective dialogue, and trust (Logsdon & Van Buren, 2009).

Although shareholders may have more salience than other stakeholder groups, engagement tactics are not the sole domain of shareholders. Similar tactics may be employed by RI intermediaries. Slager et al. (2012) show that engagement efforts are central to the construction of the FTSE4Good Index, both in terms of convening stakeholders to set CSP criteria, and through direct dialogue with companies when CSP criteria are ratcheted up. Index engagement shares some of the main characteristics of shareholder engagement described above: It consists of a long-term dialogue, framed around information exchange on CSP performance and influencing corporate policies or reporting on CSP issues. Similarly, index engagement requires skills, time, and resources that can be accumulated over time (Slager et al., 2012). It focuses on those companies under threat of exclusion from the Index that are invited to engage in dialogue about their CSP and the Index criteria. This invitation aims to raise levels of CSP over time and enables them to remain included in the FTSE4Good Index as new criteria on specific issues, such as countering bribery (CB), are introduced. Engagement is voluntary, and companies may choose not to respond to the intermediary. We argue that companies that do respond and participate in engagement are more likely to comply with the new criteria and to have higher performance in the respective area of CSP following the guidance given by the intermediary during the engagement process. Just like with shareholder engagement, effective index engagement requires time and active participation from companies. The more both parties actively engage in dialogue, the more likely it is this will affect levels of CSP. Thus, engagement presents a third mechanism that may be employed to influence levels of CSP on specific issues of engagement:

Hypothesis 3 (H3): Those companies that take part in higher levels of engagement with the intermediary are more likely to have higher levels of CSP in subsequent years.

Method

For this study, we collected a unique data set relating to the FTSE4Good Index, consisting of private correspondence, CSP data, and corporate data. Companies that meet the FTSE4Good Index CSP criteria are automatically included in the FTSE4Good Index: They do not self-select to be included. During the period analyzed in this study the information on CSP was gathered and scored independently by social research agency EIRIS, which collects and evaluates information from company reports, webpages, and information directly provided by companies. Twice a year the FTSE4Good Policy Committee convenes to review the Index CSP criteria and include or exclude companies. The FTSE4Good Index has the dual objective to serve investors interested in RI and to incentivize companies to improve their CSP through raising the bar for Index inclusion (Mackenzie, Rees, & Rodionova, 2013; Slager et al., 2012). Additional inclusion criteria have been introduced throughout the years since the Index was launched, on topics as diverse as climate change, supply chain labor standards, and anti-bribery practices.

This article studies the effects of the exclusion threats, signaling, and engagement specifically related to the FTSE4Good CB criteria, first announced in July 2006. This new category of CSP criteria require “high-risk” companies operating in corrupt environments to have in place policies and management systems to counter bribery and corruption in their operations, and to report publicly on their policies and management systems. The risk categorization is based on industry sectors, countries of operation, and involvement in government contracts. The appendix lists the high-risk categorization and criteria indicators in full.

The FTSE4Good Index is managed on a day-to-day basis by the FTSE RI team. During the introduction of the new CB criteria, the FTSE RI team first provided information regarding the criteria to companies; second, it provided warnings to those companies which were in the Index but were not meeting the new criteria; and subsequently it tried to engage in dialogue with these companies. In the engagement process, companies had the chance to provide evidence to show they are working toward meeting the criteria, a process that may take several months to years. The FTSE4Good Policy Committee ultimately decides whether a company should be excluded from the Index after this period of engagement, based on an evaluation of information provided by EIRIS and the FTSE RI team.

For this study’s sample, we selected all listed European companies that were categorized by EIRIS as operating in an environment with high risk for encountering bribery. Fifty-seven percent of companies in the sample were included in the FTSE4Good Index in the observation period, hence roughly 40% were not included in the Index.¹ The resulting panel is unbalanced due to mergers, acquisitions, and so on, and is presented as a pooled panel. The total sample includes 254 companies and 813 observations. The observation period began in 2007 and ended in 2010. This period was chosen because we expected the full effects of the mechanisms identified (exclusion threat, signaling, and engagement) to be observable within this period. Table 1 describes the variables used in the study and their data sources.

Table 1.

Variables Used in the Study.

	Description	Data source	Coding/proxy	Testing
Dependent variable
Compliance	Binary variable that indicates whether a company meets the FTSE4Good Index criteria	FTSE4Good database	1 = the company meets the Index criteria, 0 otherwise, for period 2007 to 2010	Model 1, H1, H2a
CB practices	Describes the total score given to corporate CB practices, reflecting high levels of CSP in this area	EIRIS CSP data	Summed score for quality of corporate policy, management system, and reporting. Based on converted numerical EIRIS scores, range between 0 (low) and 9 (high), for period 2007 to 2010	Model 2, H2b, H3
Independent variables
Exclusion threat	The company is threatened with exclusion for not meeting the FTSE4Good CB criteria	FTSE4Good database	1 = the company is included in the Index, but does not meet the CB practices, 0 otherwise, for period 2007 to 2010	H1
Signaling	The company communicates its inclusion in the FTSE4Good Index to stakeholders	Corporate CSP reporting	1 = the company uses the name/logo of the Index in its communications, 0 otherwise, for period 2007 to 2010	H2a, H2b
Engagement	The intensity of the engagement process with the FTSE RI team	FTSE correspondence	Number of months the dialogue lasted plus number of company actions undertaken during dialogue; first quartile coded “low,” second quartile coded “medium,” third and fourth quartile “high”	H3
Control variables
CSP reporting	The company uses the GRI to report on CSP	GRI Database	1 = the company uses GRI, 0 otherwise, for the period 2007 to 2010
Size	The size of the company	Datastream	Natural logarithm of number of employees
Return on Assets (ROA)	Financial performance of the company	Datastream	Return on assets
Risk	Management risk tolerance	Datastream	Long-term debt/total assets
Intangibles	Intangible assets including goodwill, patents, and copyright	Datastream	Natural logarithm of intangible assets
Industry	ICB classification (supersector)	FTSE/EIRIS	Including oil & gas, basic materials, industrials, consumer goods, health care, telecom, utilities
United Kingdom	Dummy variable indicating a company is listed in the United Kingdom	FTSE/EIRIS	1 = listed in the United Kingdom, 0 otherwise

Note. CB = countering bribery; CSP = corporate social performance; GRI = Global Reporting Initiative; ICB = Industry Classification Benchmark.

Dependent Variable

Data regarding the new FTSE4Good CB criteria started to be collected by EIRIS from 2006 onwards following a consultation led by FTSE to define methodology. Data exist for most companies categorized as high risk from 2007. The data for 2007 are therefore taken as a baseline in the study. The data cover the EIRIS evaluation of the quality of anti-bribery policies, management systems, and corporate reporting. These three elements are rated by text grades: no evidence, limited, intermediate, good, and advanced. In line with previous use of the EIRIS database by Brammer and Millington (2008), the text gradings are converted into numerical scores. Because the number of companies rated good is low (N < 10), the categories of good and advanced are merged into one, leading to numerical scores between 0 and 3 on the each of the three elements. As the “intermediate” grade was designed to reflect the FTSE4Good index inclusion threshold, this grade still provides enough information for the analysis.

The analysis employs two dependent variables: compliance with the overall FTSE4Good Index inclusion criteria and the quality of CB practices in particular. The first dependent variable, compliance, is a binary variable that identifies companies that have received the minimum score that is needed to remain in the FTSE4Good Index. Thus a 1 signifies the company complies with the Index criteria, and has the level of CSP that is required to remain in the index. Data on compliance for the observation period are taken from the FTSE4Good database. The compliance dependent variable is used to test H1 and H2a. Although the first dependent variable provides a binary indication of broad compliance with the Index criteria, we create a second dependent variable to test a fuller range of performance on a specific CSP issue. The second dependent variable, CB practices, is the summed total score for the three CB criteria elements, as given by EIRIS. A company with elaborate and high-quality CB practices receives a higher score than those companies that disclose little about their CB practices or lack appropriate policies and management systems. Thus higher levels of the dependent variable reflect higher levels of CSP in this area. We include this second dependent variable in our analysis in particular to test the effects of engagement on the new CB criteria that took place during the observation period (see below); it is thus used to test H2b and H3. Although subsuming CSP aspects into one aggregate score has been criticized as theoretically and empirically unsound (Griffin & Mahon, 1997), here the aggregated score is less problematic because it measures only one aspect of CSP.

Independent Variables

The independent variables are created based on FTSE4Good archival data and corporate data. The FTSE4Good archival data collected include a longitudinal database that is used to identify all companies eligible for Index inclusion based on their CSP scores. It also includes private correspondence between companies and the FTSE RI team, such as formal letters, emails, and minutes of meetings. The correspondence is normally started by the FTSE RI team when a company is warned regarding not meeting the CB criteria. The subsequent dialogue would include, for example, discussions about definitions of whistle blowing or facilitation payments, companies providing more information about anti-bribery managing systems, or questioning appropriate forms of corporate reporting. The dialogue is continued until the company meets the criteria or is deleted from the Index. The data collected from FTSE4Good and corporate sources were used to create three independent variables as proxies for the three mechanisms identified.

First, we create a binary measure of exclusion threat. Data regarding the compliance of European companies with the CB criteria were extracted from the FTSE4Good archives. A company is coded 1, if it does not meet the CB criteria, but continues to be included in the Index in that year. It is coded 0 otherwise. This binary measure corresponds to those companies facing the threat of exclusion from the Index due to not meeting the CB criteria at that point in time. It serves as a precursor to the process of engagement (see below), which may serve to incentivize compliance with the Index criteria in subsequent years. To allow for this time effect, exclusion threat is lagged by 1 year in the analysis.

Second, we create the measure of signaling by examining CSP disclosure. If a company in the FTSE4Good Index advertises the fact that it is included in the Index, this advertising can signal reputation for “good CSP,” and act as a reputation differentiator (Deephouse & Suchman, 2008). FTSE does not advertise the complete list of Index constituents. Instead, companies may opt to undertake this signaling through promoting Index inclusion as a de facto certification of their CSP (Consolandi et al., 2009; Slager et al., 2012). To assess whether a company was using FTSE4Good as a signaling tool, stand-alone CSP reports and, where no stand-alone CSP report was published, sections of annual reports reporting on CSP were examined for the period between 2007 and 2010. Signaling was coded 1 if the company included the name or logo of the FTSE4Good Index in its reporting that year, and 0 otherwise. Although the logo is widely used in annual CSP communication, not all companies use the logo. Just over half (58%) of the companies included in the Index used signaling in the observation period. Signaling is lagged by 1 year in the analysis to capture its effect on compliance and levels of CSP in subsequent years.

Third, we create the measure engagement based on the FTSE archive of correspondence with companies regarding the CB criteria. As highlighted above, companies are not automatically excluded from the Index when they do not meet the criteria. Instead, they have the opportunity to enter into engagement with the FTSE RI team on the specific criteria they are not meeting. Only if a company is not responsive to the opportunity for discussion, or it is clear that it cannot or does not want to meet the criteria, will it be excluded at the initial deadline. Where a company is making progress and makes a formal written commitment to FTSE, the FTSE4Good Policy Committee will normally approve an extension of the company’s deadline. As the exclusion threat acts as a precursor to engagement, the variables are not independent and we analyze their effects in separate models. We include the measure of engagement to test the effects on a specific topic of CSP, in this case CB. We collected the archived correspondence between the FTSE RI team and companies engaged on the CB criteria (500+ emails), and summarized and coded their contents. To create an ordinal measure of engagement we coded for the length and intensity of the correspondence. The number of months the company is in engagement plus the number of company actions (such as sending more information, or requesting a meeting, etc.) are summed into a total score for engagement intensity. Engagement intensity is coded “low” for the first quartile, “medium” for the second quartile, and “high” for the third and fourth quartiles of the scores generated by coding the correspondence. A “low” engagement score means the engagement lasted less than 6 months, which is the time between two Index reviews, and the company undertook relatively little action. More intensive engagement can be characterized by a flurry of company actions, engagement over a prolonged period of time, or both. Those classified in the high engagement category would have higher levels of company activity over a prolonged period of time. This ordinal measure of engagement corresponds to the idea that companies in intensive engagement are likely to benefit more from the dialogue between the FTSE RI team and corporate management. Again, engagement is lagged by 1 year to enable capturing its effects on CB practices in subsequent years.

Control Variables

As Doh et al. (2010) point out, the degree of information asymmetry regarding CSP can be reduced by intermediaries or by the firm itself. Transparency and good-quality CSP reporting reduces information asymmetry and facilitates evaluation by intermediaries, as well as other stakeholders. Companies are more likely to provide information on relevant aspects of CSP if they have a history of being open about CSP practices and have set up corporate structures to report CSP and deal with information requests. We control for the quality of existing CSP reporting by extracting data from the Global Reporting Initiative (GRI) Sustainability Disclosure database. The GRI Sustainability Reporting Guidelines have become the de facto standard for meaningful, high-quality CSP reporting (Etzion & Ferraro, 2010). First introduced in 1999, the current version of the guidelines includes recommendations for disclosure on bribery and corruption. A company is coded 1 if it has reported to use the GRI and 0 otherwise, for each of the years in the observation period. If the company is using GRI, it is assumed that information asymmetry is lower, and that existing reporting systems may enable a faster response to new requests for CSP data from intermediaries.

We also control for firm characteristics that have been identified as introducing variance in CSP, such as organizational size, financial performance, risk, and industry sector (Margolis, Elfenbein, & Walsh, 2007; Orlitzky & Benjamin, 2001; Orlitzky, Schmidt, & Rynes, 2003). Financial data were taken from Datastream. Organizational size was controlled for by taking the natural logarithm of the number of employees. An accounting-based measure of financial performance, return on assets, was used to control for financial performance, to reflect the company’s use of resources and financial strength. The long-term debt to total assets ratio was taken as a proxy for financial risk (Waddock & Graves, 1997). The industry sectors represented in the sample were restricted to those considered high risk for encountering bribery and corruption as per the FTSE classification (see the appendix). We expect that companies from high-risk industry sectors will undertake more action to comply with CSP standards in this area, while we want to control for intra-industry differences. The number of industry sectors represented in the sample was coded following the Industry Classification Benchmark. The selected reference group, the technology sector, has above average scores in the Transparency International Bribe Payers index (meaning less reported instances of bribery; Transparency International Bribe Payers index, 2011). Thus, the industry sectors in the analysis were deemed higher risk for corruption and bribery relative to the reference group, technology. According to McWilliams and Siegel (2000, 2001, 2011), CSP can be viewed as a form of product innovation (the creation of new socially responsible product features or categories) or process innovation (the use of a socially responsible production process; McWilliams & Siegel, 2011). We thus control for R&D expenditure by collecting data on intangible assets, including goodwill, patents, and copyright from Datastream. The natural logarithm of intangibles is used in the data set. Finally, the degree of government regulation may also influence how firms respond to activists’ pressures (Chatterji & Toffel, 2010; Reid & Toffel, 2009). A dummy variable was created to control for companies based in the United Kingdom. Within Europe, the United Kingdom can be considered to have the most advanced regulation regarding corruption and bribery in the form of the U.K. Bribery Act (Osuji, 2011). Although the introduction of the U.K. Bribery Act was delayed and it only came into force in 2011, consultations regarding the act first started in 2002 and 2005. Therefore, companies in the United Kingdom might have anticipated the forthcoming regulation.

Analysis

Two models were estimated to test the hypotheses. The logit analysis in Model 1 tests the effects of the exclusion threat and signaling on the general compliance with the FTSE4Good Index inclusion criteria, thus testing H1 and H2a. The Tobit analysis in Model 2 tests the effects of signaling and engagement on the quality of CB practices, testing H2b and H3. The dependent variable in Model 2 is censored in the sense that it can only have values between 0 and 9, and cannot take negative values. This means a censored regression technique is necessitated because ordinary least squares (OLS) estimation can provide both biased and inconsistent parameter estimates (Greene, 2008, Greene & Hensher, 2010). The most commonly adopted solution to these types of data is to estimate a Tobit model using by maximum-likelihood estimation (see Brammer & Millington, 2006, 2008). The Tobit model is suitable when the dependent variable is zero for a non-trivial proportion of the sample and roughly continuously distributed over the positive values (Greene, 2008).

Both models are pooled to allow for the examination of variance over time. A pooled model effectively ignores individual effects to explore situations in which the main interest is in the effect of an intervention (in this case, the introduction of the CB criteria in the Index), the cases do not constitute a random sample of the population and the panel is unbalanced (Fiss & Zajac, 2006; Hsiao, 1985; Petersen, 1993). Time effects can be captured in pooled analysis by including year dummies. We have included dummy variables for the years in our observation period. All independent and control variables were lagged by 1 year to avoid reverse causality. This also makes sense substantially; we expect the effects of engagement on CSP to develop over time. All variables were estimated with robust standard errors.

Results

The models were estimated using Stata 13. The descriptive statistics are provided in Table 2. Table 3 provides the results of Model 1.

Table 2.

Summary Statistics and Correlations.

	M	SD	1	2	3	4	5	6	7	8	9	10	11	12
1. Compliance	0.57	0.49
2. CB practices	3.93	2.25	.810***
3. Exclusion threat	0.11	0.31	.113**	.091*
4. Engagement low	0.04	0.20	.105**	.072*	.686***
5. Engagement med	0.02	0.14	.084*	.072*	.455***	−.039
6. Engagement high	0.03	0.15	−.010	.005	.500***	−.043	−.028
7. Signaling	0.24	0.43	.365***	.357***	.197***	.148***	.127***	.048
8. CSP reporting	0.31	0.46	.403***	.407***	−.048	.008	−.043	−.060	.267***
9. Size	9.58	1.71	.390***	.443***	−.054	−.015	−.060	−.023	.169***	.323***
10. Intangibles	13.83	2.18	.411***	.474***	−.033	−.016	−.071*	.025	.250***	.394***	.681***
11. Risk	0.76	1.15	.084*	.062	.073*	.118***	−.072*	.043	.125***	.085*	−.082*	.134***
12. ROA	7.25	9.77	−.003	.014	.018	−.021	.068	−.002	.051	−.046	.081*	−.023	−.081*
13. Risk	0.38	0.48	−.160***	−.144***	.062	.025	.140***	−.048	−.078*	−.253***	−.375***	−.367***	−.093**	.015

Note. CB = countering bribery; CSP = corporate social performance.

p < .05. **p < .01. ***p < .001.

Table 3.

Binary Logit Analysis of Compliance With the FTSE4Good Criteria.

	Model 1
Exclusion threat	2.875**
Signaling	5.702***
CSP reporting	3.410***
Size	1.608***
Intangibles	1.176**
Risk	1.139
ROA	0.983
Oil and gas	2.801**
Basic materials	1.127
Industrials	1.349
Consumer goods	1.023
Health care	8.715***
Telecommunications	0.402
Utilities	1.253
United Kingdom	1.386
2008	2.082**
2009	3.354***
2010	3.426***
N	813
Pseudo R²	.313
Wald χ²	186.32***
df	18
Log pseudo likelihood	−387.18

Note. CSP = corporate social performance. Odds ratios reported.

p < .05. **p < .01. ***p < .001.

Model 1 illustrates the effects of exclusion threats and signaling on compliance with the FTSE4Good Index inclusion criteria. The coefficients in logit models are not straightforwardly interpreted as they are reported as the log of odds. To enable more intuitive interpretation, we present the results of the logit model in Table 3 in odds ratios, which are interpreted in this model as the odds of compliance with FTSE4Good criteria or odds of non-compliance with FTSE4Good Criteria. This allows a more intuitive interpretation of the results of the logit model. The results confirm H1, which predicted companies that are threatened by exclusion from RI indices are more likely to comply with the intermediary’s general CSP criteria for index inclusion in subsequent years. Exclusion threat is positive and significant: The odds ratio shows that those companies that have encountered an exclusion threat in the previous year meeting FTSE4Good criteria is 2.875:1; companies that face exclusion threats are nearly three times more likely to meet the FTSE4Good criteria in subsequent years than those that do not. The results also confirm H2a, which predicted that companies that signal CSP reputation through intermediary endorsements are more likely to comply with the intermediary’s CSP criteria in subsequent years. The results of the logit model shows the strength of the signaling effect, as the odds ratio for signaling can be interpreted as companies that signal inclusion in the FTSE4Good index in CSP disclosure are nearly six times more likely to meet the criteria the following year.

Table 4 reports the results of Model 2, the Tobit model that examines the effects of signaling and engagement regarding CB issues on the quality of the CB practices, as measured by the EIRIS scores. The results confirm the positive and significant effects of signaling, supporting H2b. H3, which predicted that those companies that take part in higher levels of engagement with the intermediary are more likely to have higher levels of CSP in subsequent years, is partially supported. We find that although the medium level of engagement is significant and positive, low and high levels of engagement are not significant. For the control variables, we find that the coefficients for CSP reporting, size, intangible assets, the oil and gas, health care and utilities industries (in Model 2 only), and the year dummies are all positive and significant (p < 0.01).

Table 4.

Tobit Analysis of the Quality of Countering Bribery Practices.

	Model 2
Signaling	1.231***	1.268***
Engagement low	0.513
Engagement med	1.241***	1.185***
Engagement high		0.381
CSP reporting	0.872***	0.884***
Size	0.458***	0.464***
Intangibles	0.204***	0.199***
Risk	0.0419	0.0545
ROA	0.00238	0.00194
Oil and gas	1.217***	1.172***
Basic materials	0.0669	0.0271
Industrials	0.608*	0.564*
Consumer goods	−0.276	−0.305
Health care	1.582***	1.530***
Telecommunications	−0.282	−0.375
Utilities	0.982**	0.963**
United Kingdom	0.528**	0.549***
2008	1.165***	1.208***
2009	1.595***	1.609***
2010	1.670***	1.663***
N	798	798
Pseudo R²	.134	.133
F	37.94***	37.96***
Log pseudo likelihood	−1,532.07	−1,533.03

Note. CSP = corporate social performance.

p < .05. **p < .01. ***p < .001.

Discussion

Previous research has taken a rather one-dimensional view of intermediaries as information providers, which evaluate corporate practices against set standards, with improved information provision as the main outcome of this process. Within these studies, the improved information provision translates into signals regarding corporate reputation and legitimacy that may affect financial performance (Doh et al., 2010; King & Whetten, 2008; Rindova et al., 2005). However, these studies have not examined a number of additional roles that intermediaries may undertake. Our findings suggest RI intermediaries do more than providing legitimacy to companies when providing a benchmark for the evaluation of “good” CSP. They may play an active role in fostering corporate sustainability. First, inclusion in RI indices, such as the FTSE4Good Index, creates a level playing field of legitimate companies with good CSP. By raising the bar for “good CSP” through the introduction of new Index criteria, the FTSE4Good Index has incentivized companies to improve CSP. When FTSE introduces new Index inclusion criteria, this is taken as a signal that the relevant issue has become part of the CSP agenda and needs to be addressed accordingly. The results support the view that the FTSE RI team is able to successfully convince companies to comply with new Index inclusion criteria, by using the promise of continued index inclusion as a carrot and the threat of index deletion as a stick.

Second, companies may also wish to leverage their perceived legitimacy in this field to signal CSP reputation, by differentiating their CSP disclosure from non-included companies by displaying the logos of RI indices. Whether focusing on overall compliance or performance on a specific issue such as CB, we found that the effects of signaling are strong. The finding supports the idea that the FTSE4Good Index is used by companies as a certification of good CSP practices to strengthen reputation (Consolandi et al., 2009; Doh et al., 2010).

Third, RI intermediaries may undertake index engagement, which shows similar characteristics to shareholder engagement: a process of dialogue with company management regarding CSP performance that often requires significant investment in time and resources from both parties (Gifford, 2010; Gond & Piani, 2013; Logsdon & Van Buren, 2009). Index engagement presses companies for more transparency on CSP, and also serves to provide companies information and guidance on index criteria. It may be used as new Index criteria, such as the CB criteria that are introduced to large numbers of companies. Our findings show index engagement may be effective at encouraging companies to improve CSP. However, results were not consistent across all categories of engagement. The results show that although medium engagement produces higher CSP scores in subsequent years, lower and higher levels of engagement do not. This relationship suggests that those companies that are either in short or protracted dialogues are less likely to ultimately improve their practices on specific CSP issues. It seems there is an optimum period of effective engagement (between 7 and 15 months in this case), after which the dialogue is less likely to lead to improved CSP practices regarding bribery and corruption. This medium intensity engagement is most likely to produce results for intermediaries.

The findings further imply that using standardized templates for transparency and reporting on CSP facilitate evaluation by external rating agencies and reduce information asymmetries with other stakeholders. Using the GRI reporting standards signals that reporting systems are in place to accommodate information requests from intermediaries. However, care should be taken in interpreting this finding, as the GRI provides only guidelines for reporting, not rigid standards. Companies that use the GRI may still display variance in their CSP disclosure to the public, including on issues related to CB. With regards to the control variables, larger firms and those with greater intangible assets are more likely to have better CB practices. The results show perhaps unsurprisingly that companies in industry sectors known for corruption scandals (e.g., Shell in Nigeria or the unfolding case of GlaxoSmithKline in China) are more likely to have better anti-corruption systems. It was also found that companies in the United Kingdom, where there is stricter anti-corruption legislation in place, are more likely to have higher performance on CB practices. This concurs with the findings by Chatterji and Toffel (2010) and Reid and Toffel (2009), who find that higher levels of government regulation lead to higher levels CSP. We also find time effects, showing that as more time has passed since the introduction of the CB criteria in 2007, companies are more likely to be compliant and show higher levels of CSP on this issue.

Implications and Future Research

Our study has multiple implications for the study of corporate reputation and the role of intermediaries in signaling reputation in situations of information asymmetry, such as those related to CSP. First, standards for CSP do not appear out of nowhere, and intermediaries may play an active role in setting or promoting the norms, standards, and behavioral guidelines that determine the legitimacy of corporate behavior in a given field (Brunsson & Jacobsson, 2000; Suchman, 1995). As can be seen in the case of FTSE4Good, intermediaries may use the Index as both a carrot and stick to change corporate behavior by continuously raising the bar for what is considered to be good CSP. In doing so, intermediaries must negotiate substantial cultural, legal, and institutional differences in norms and standards for CSP around the world. Intermediaries’ role in setting and diffusing globally accepted standards merits more critical attention in research on CSP and reputation.

Second, the research on corporate reputation rarely takes into account the feedback role of reputation intermediaries. Yet a growing stream of research on third-party evaluation of organizational performance through rankings, ratings, certification, and accreditation shows that the process of social evaluation incentivizes changes in organizational behavior, especially when reputational stakes are high (Elsbach & Kramer, 1996; Sauder & Espeland, 2009). This feedback effect indicates the existence of multidirectional relationships between corporate practices, evaluation, and reputation, which deserve further scrutiny.

Third, intermediaries may choose to play an active role through engagement with their participants of evaluation. Engagement by intermediaries may range from the soliciting of information needed to evaluate performance to the active lobbying of management to comply with the criteria of evaluation. In the case of the FTSE4Good Index, engagement was necessitated by the need to minimize turnover of Index constituents and the explicit objective for the Index to incentivize change in CSP (Slager et al., 2012). Although the FTSE4Good Index has been public about its engagement strategy, there may be examples of intermediaries where engagement is more obscured from public view. For example, certification or accreditation may only prove a valuable legitimating signal once a critical mass of participating organizations has been reached, and certification bodies may need to engage with their target organization to convince them of the worth of certification (Durand & McGuire, 2005). This engagement may be hidden from view, but, as our findings show, its effects on corporate practices need to be taken into account when studying the role of intermediaries.

Finally, there is increasing recognition of the active role firms play in managing the reputational signals emitted by intermediaries. Elsbach and Kramer (1996) show that when rankings produced by intermediaries are perceived to be a threat to organizational identity, organizational members may choose to highlight other types of evaluation, which align more closely with their values (Elsbach & Kramer, 1996). With the multitude of rankings and lists that exist nowadays to measure CSP, it is easy for firms to promote only those that show them in good light. New research in institutional theory recognizes organizational actors play an active role in establishing, maintaining, or disrupting institutions (Lawrence & Suddaby, 2006; Lawrence, Suddaby, & Leca, 2009), and this may include a more active response to market intermediaries that uphold institutions. For example, whereas securities analysts have traditionally been seen as a source of institutional pressure in financial markets (Beunza & Garud, 2007; Zuckerman, 1999), current research have emphasized the active response of companies in offsetting or manipulating analysts’ recommendations (Benner & Ranganathan, 2012). Further research could examine whether and how CSP signals from intermediaries are managed by companies.

Our study also has implications for the study of engagement processes. Previous research on shareholder engagement has emphasized the unique and individual nature of interaction between shareholders and company management, indicating that each dialogue requires sufficient attention to be paid to specific company contexts as well as the CSP issue of concern. This finding may to some degree be an artifact of the research methods in these studies, which often employ a limited number of qualitative case studies with little comparative analysis (Gifford, 2010; Gond & Piani, 2013; Southwood, 2003). Of course, due to the private nature of shareholder engagement dialogues, large data sets are hard to come by for researchers. Based on to the FTSE4Good data set collated for this study, we show that index engagement can effectively involve large numbers of companies. Classifying companies into risk categories based on their exposure to CSP elements such as bribery and corruption means that a degree of inter-firm differences can still be taken into account, even when engaging large numbers of firms. The findings imply that index engagement may use the threat of exclusion as a stick and reputation signaling as a carrot to incentivize companies to comply with set CSP standards. Finally, the findings indicate that there might be an optimal time frame for engagement, and that protracted engagement might be less effective in achieving high levels of CSP, indicating other mechanisms such as exclusion should be considered at that point.

Our study has a number of limitations. The relatively short time span of the observation period (2007-2010) tells us little about longer term effects. Future work could analyze responses to index inclusion and engagement over longer periods of time. Furthermore, CB practices form only one part of the spectrum of activities, processes, and structures that make up CSP. Furthermore, work needs to be done before the findings might be generalized to other areas of CSP. For example, what is the role of regulation and voluntary initiatives in areas such as the protection of human rights or labor standards in the supply chain? The current analysis also isolates the companies identified as being at high risk for encountering bribery and corruption. More research could identify whether the companies that face lower risk are more or less likely to respond to index inclusion and engagement and investigate the interactions between risk, reputation, and engagement effects further. Finally, our study looked at a specific type of intermediary only. Graffin and Ward (2010) distinguish between different types of intermediaries, such as certification, rating, and accreditation agencies and suggest that each will have different effects on corporate reputation under situations of uncertainty. They suggest that these differences will depend on whether the outcome of evaluation is based on a relative rank ordering or a measure of absolute performance against the standard in question (Graffin & Ward, 2010). Further research could explore different types of intermediaries, including those using absolute or relative performance benchmarks, to examine their effects.

Conclusion

Previous research shows that RI indices act as reputation intermediaries by evaluating information regarding CSP for investors and other stakeholders (Consolandi et al., 2009; Doh et al., 2010). Our research shows that the role of intermediaries is not confined to information provision. We identify three mechanisms that may influence CSP: exclusion threats, signaling, and engagement. By explicating these mechanisms, we provide a more comprehensive view of the role of intermediaries in conferring legitimacy and reputation for CSP. The identified mechanisms may be used to study the effects on CSP beyond the corruption and bribery context, as well as providing a useful framework to study the effects of other forms of engagement.

Footnotes

Appendix

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research on which this paper is based was partly funded by the Economic and Social Research Council, UK, as a PhD CASE studentship.

Notes

Author Biographies

Rieneke Slager, PhD in business and management, Nottingham University Business School, is assistant professor in strategy and sustainability at Nottingham University Business School (NUBS) in the United Kingdom. Her research examines the link between corporate social responsibility (CSR) and responsible investment markets, the influence of CSR metrics on organizational behavior, and institutional work in the context of CSR, including anti-bribery and corruption initiatives. Her work appears or is forthcoming in Business & Society, Organization Studies and Policy & Politics.

Wendy Chapple, PhD in industrial economics, Nottingham University Business School, is deputy director of the International Centre for Corporate Social Responsibility and associate professor in industrial economics at NUBS in the United Kingdom. Her research includes the analysis of corporate social performance, corporate impact and competitiveness, comparative CSR, and institutions in both developed and developing country contexts. Her work has appeared in journals such as Business & Society, Journal of Business Ethics, Journal of Productivity Analysis Managerial Economics and Decision, and Research Policy.

References

Benner

M. J.

Ranganathan

R. A. M.

(2012). Offsetting illegitimacy? How pressures from securities analysts influence incumbents in the face of new technologies. Academy of Management Journal, 55, 213-233.

Bertels

Peloza

(2008). Running just to stand still? Managing CSR reputation in an era of ratcheting expectations. Corporate Reputation Review, 11, 56-72.

Beunza

Garud

(2007). Calculators, lemmings or frame-makers? The intermediary role of security analysts. In Callon

Millo

Muniesa

(Eds.), Market devices (pp. 13-39). Malden, MA: Blackwell.

Brammer

Millington

(2006). Firm size, organizational visibility and corporate philanthropy: An empirical analysis. Business Ethics: A European Review, 15, 6-18.

Brammer

Millington

(2008). Does it pay to be different? An analysis of the relationship between corporate social and financial performance. Strategic Management Journal, 29, 1325-1343.

Brunsson

Jacobsson

(2000). A world of standards. Oxford, UK: Oxford University Press.

Capelle-Blancard

Couderc

(2009). The impact of socially responsible investing: Evidence from stock index redefinitions. Journal of Investing, 18, 76-86.

Chatterji

Toffel

M. W.

(2010). How firms respond to being rated. Strategic Management Journal, 31, 917-945.

Cheung

A. W. K.

(2011). Do stock investors value corporate sustainability? Evidence from an event study. Journal of Business Ethics, 99, 145-165.

10.

Clark

G. L.

Hebb

(2004). Pension fund corporate engagement: The fifth stage of capitalism. Relations Industrielles/Industrial Relations, 59(1), 142-171. Retrieved from http://www.erudit.org/revue/ri/2004/v59/n1/009130ar.pdf

11.

Collison

D. J.

Cobb

Power

D. M.

Stevenson

L. A.

(2008). The financial performance of the FTSE4Good Indices. Corporate Social Responsibility and Environmental Management, 15, 14-28.

12.

Collison

D. J.

Cobb

Power

D. M.

Stevenson

L. A.

(2009). FTSE4Good: exploring its implications for corporate conduct. Accounting, Auditing & Accountability Journal, 22, 35-58.

13.

Consolandi

Jaiswal-Dale

Poggiani

Vercelli

(2009). Global standards and ethical stock indexes: The case of the Dow Jones Sustainability STOXX Index. Journal of Business Ethics, 87, 185-197.

14.

Curran

M. M.

Moran

(2007). Impact of the FTSE4Good Index on firm price: An event study. Journal of Environmental Management, 82, 529-537.

15.

Deephouse

D. L.

Carter

S. M.

(2005). An examination of differences between organizational legitimacy and organizational reputation. Journal of Management Studies, 42, 329-360.

16.

Deephouse

D. L.

Suchman

(2008). Legitimacy in organization institutionalism. In Greenwood

Oliver

Suddaby

Sahlin-Andersson

(Eds.), The SAGE handbook of organizational institutionalism (pp. 49-77). Thousand Oaks, CA: Sage.

17.

Doh

J. P.

Howton

S. D.

Howton

S. W.

Siegel

D. S.

(2010). Does the market respond to an endorsement of social responsibility? The role of institutions, information, and legitimacy. Journal of Management, 36, 1461-1485.

18.

Durand

McGuire

(2005). Legitimating agencies in the face of selection: The case of AACSB. Organization Studies, 26, 165-196.

19.

Elsbach

K. D.

Kramer

R. M.

(1996). Members’ responses to organizational identity threats: Encountering and countering the Business Week rankings. Administrative Science Quarterly, 41, 442-746.

20.

Etzion

Ferraro

(2010). The role of analogy in the institutionalization of sustainability reporting. Organization Science, 21, 1092-1107.

21.

Fiss

P. C.

Zajac

E. J.

(2006). The symbolic management of strategic change: Sensegiving via framing and decoupling. Academy of Management Journal, 49, 1173-1193.

22.

Fombrun

Shanley

(1990). What’s in a name? Reputation building and corporate strategy. Academy of Management Journal, 33, 233-258.

23.

FTSE. (2011). FTSE4Good 10 years of impact & investment. Retrieved from http://www.ftse.com/products/indices/esg-historical

24.

Gifford

(2010). Effective shareholder engagement: The factors that contribute to shareholder salience. Journal of Business Ethics, 92, 79-97.

25.

Gond

J.-P.

Piani

(2013). Enabling institutional investors’ collective action: The role of the Principles for Responsible Investment initiative. Business & Society, 52, 64-104.

26.

Graffin

S. D.

Ward

A. J.

(2010). Certifications and reputation: Determining the standard of desirability amidst uncertainty. Organization Science, 21, 331-346.

27.

Greene

W. H.

(2008). Econometric analysis. Upper Saddle River, NJ: Pearson.

28.

Greene

W. H.

Hensher

D. A.

(2010). Modeling ordered choices: A primer. Cambridge, UK: Cambridge University Press.

29.

Griffin

J. J.

Mahon

J. F.

(1997). The corporate social performance and corporate financial performance debate. Twenty-five years of incomparable research. Business & Society, 36, 5-31.

30.

Hsiao

(1985). Analysis of panel data. Cambridge, UK: Cambridge University Press.

31.

King

B. G.

Whetten

D. A.

(2008). Rethinking the relationship between reputation and legitimacy: A social actor conceptualization. Corporate Reputation Review, 11, 192-207.

32.

Lawrence

T. B.

Suddaby

(2006). Institutions and institutional work. In Clegg

S. R.

Lawrence

T. B.

Nord

W. R.

(Eds.), The SAGE Handbook of organization studies (pp. 215-254). London, UK: Sage.

33.

Lawrence

T. B.

Suddaby

Leca

(2009). Institutional work: Actors and agency in institutional studies of organizations. Cambridge, UK: Cambridge University Press.

34.

Logsdon

Van Buren

(2009). Beyond the proxy vote: Dialogues between shareholder activists and corporations. Journal of Business Ethics, 87, 353-365.

35.

Love

E. G.

Kraatz

(2009). Character, conformity, or the bottom line? How and why downsizing affected corporate reputation. Academy of Management Journal, 52, 314-335.

36.

Mackenzie

Rees

Rodionova

(2013). Do responsible investment indices improve corporate social responsibility? FTSE4GOOD’s impact on environmental management. Corporate Governance: An International Review, 21, 495-512.

37.

Margolis

J. D.

Elfenbein

H. A.

Walsh

J. P.

(2007, July 26). Does it pay to be good? A meta-analysis and redirection of research on the relationship between corporate social and financial performance. Retrieved from http://ssrn.com/abstract=1866371

38.

McLaren

(2004). Global stakeholders: Corporate accountability and investor engagement. Corporate Governance: An International Review, 12, 191-201.

39.

McWilliams

Siegel

(2000). Corporate social responsibility and financial performance: Correlation or misspecification? Strategic Management Journal, 21, 603-609.

40.

McWilliams

Siegel

(2001). Corporate social responsibility: A theory of the firm perspective. Academy of Management Review, 26, 117-127.

41.

McWilliams

Siegel

(2011). Creating and capturing value: Strategic corporate social responsibility, resource-based theory, and sustainable competitive advantage. Journal of Management, 37, 1480-1495.

42.

Orlitzky

Benjamin

J. D.

(2001). Corporate social performance and firm risk: A meta-analytic review. Business & Society, 40, 369-396.

43.

Orlitzky

Schmidt

F. L.

Rynes

S. L.

(2003). Corporate social and financial performance: A meta-analysis. Organization Studies, 24, 403-441.

44.

Osuji

(2011). Fluidity of regulation-CSR nexus: The multinational corporate corruption example. Journal of Business Ethics, 103, 31-57.

45.

Oulton

(2006). The role of activism in responsible investment: The FTSE4Good indices. In Sullivan

Mackenzie

(Eds.), Responsible investment (pp 196-205). Sheffield, UK: Greenleaf Publishing.

46.

Petersen

(1993). Recent advances in longitudinal methodology. In Blake

Hagan

(Eds.), Annual review of sociology (pp. 425-454). Palo Alto, CA: Annual Reviews.

47.

Rao

(1994). The social construction of reputation: Certification contests, legitimation and the survival of organizations in the American automobile industry: 1895-1912. Strategic Management Journal, 15, 29-44.

48.

Rao

(1998). Caveat emptor: The construction of nonprofit consumer watchdog organizations. American Journal of Sociology, 103, 912-961.

49.

Reid

E. M.

Toffel

M. W.

(2009). Responding to public and private politics: Corporate disclosure of climate change strategies. Strategic Management Journal, 30, 1157-1178.

50.

Rhee

Valdez

M. E.

(2009). Contextual factors surrounding reputation damage with potential implications for reputation repair. Academy of Management Review, 34, 146-168.

51.

Rindova

V. P.

Williamson

I. O.

Petkova

A. P.

Sever

J. M.

(2005). Being good or being known: An empirical examination of the dimensions, antecedents, and consequences of organizational reputation. Academy of Management Journal, 48, 1033-1049.

52.

RobecoSAM. (2013). The sustainability yearbook 2013. Available from www.robecosam.com

53.

Robinson

Kleffner

Bertels

(2011). Signaling sustainability leadership: Empirical evidence of the value of DJSI membership. Journal of Business Ethics, 101, 493-505.

54.

Sauder

(2006). Third parties and status position: How the characteristics of status systems matter. Theory and Society, 35, 299-321.

55.

Sauder

(2008). Interlopers and field change: The entry of U.S. News into the field of legal education. Administrative Science Quarterly, 53, 209-234.

56.

Sauder

Espeland

W. N.

(2009). The discipline of rankings: Tight coupling and organizational change. American Sociological Review, 74, 62-82.

57.

Scalet

Kelly

T. F.

(2009). CSR rating agencies: What is their global impact? Journal of Business Ethics, 94, 69-88.

58.

Scott

R. W.

(2001). Institutions and organizations. Thousand Oaks, CA: Sage.

59.

Slager

Gond

J.-P.

Moon

(2012). Standardization as institutional work: The regulatory power of a responsible investment standard. Organization Studies, 33, 763-790.

60.

Southwood

(2003). Shareholder engagement: Prospects for improving corporate social responsibility. International Journal of Business Performance Management, 5, 223-236.

61.

Suchman

M. C.

(1995). Managing legitimacy: Strategic and institutional approaches. The Academy of Management Review, 20, 571-610.

62.

Sullivan

Mackenzie

(2008). Can investor activism play a meaningful role in addressing market failures? Journal of Corporate Citizenship, 31, 77-88.

63.

Transparency International Bribe Payers Index. (2011). Retrieved from http://www.transparency.org/bpi2011

64.

Vandekerckhove

Leys

Van Braeckel

(2007). That’s not what happened and it’s not my fault anyway! An exploration of management attitudes towards SRI-shareholder engagement. Business Ethics: A European Review, 16, 403-418.

65.

Waddock

S. A.

Graves

S. B.

(1997). The corporate social performance—Financial performance link. Strategic Management Journal, 18, 303-319.

66.

Zuckerman

E. W.

(1999). The categorical imperative: Securities analysts and the illegitimacy discount. American Journal of Sociology, 104, 1398-1438.

67.

Zuckerman

E. W.

(2000). Focusing the corporate product: Securities analysts and de-diversification. Administrative Science Quarterly, 45, 591-619.