Institutional ownership, environmental, social, and governance performance and disclosure – a review on empirical quantitative research

Since the financial crisis of 2008–2009, nonfinancial-related shareholder activism increased, as public interest entities (PIEs) should strengthen their environmental, social, and governance (ESG) activities. This study aims to determine whether institutional ownership (IO) impacts ESG performance and disclosure and vice versa. Moreover, IO’s moderating and mediating influence on the relationship between ESG and firms’ financial consequences is included. This is the first literature review focusing on IO and ESG, describing IO as independent, dependent, moderator, and mediator variable. A structured literature review with 81 empirical-quantitative (archival) studies on that topic is presented based on an agency theoretical framework. Regarding the main results, long-term IO leads to increased ESG performance. Moreover, ESG performance promotes the ratio of institutional investors. Other relationships are rather heterogeneous and too low in an amount yet, stressing major research gaps.


INTRODUCTION
Institutional investor activism has become a major topic during the annual general meetings of Public Interest Entities (PIEs) (Dangaard, 2019; Villalonga, 2018). Since the financial crisis of 2008-2009, a rapid increase in institutional ownership can be found from an international perspective (Sethi, 2005). Shares of most large corporations are owned by institutions rather than individuals (Cox, Brammer, & Millington, 2004), especially in developed countries (Dyck, Lins, Roth, & Wagner, 2019). For example, the proportion of U.S. public equities managed by institutions has risen steadily over the past six decades, from about 7 or 8% of market capitalization in 1950 to about 67 % in 2010 (Aguilar, 2013). In contrast to private investors, institutional investors are companies or organizations that invest money on behalf of other people or organizations. The main examples are mutual funds, pensions, and insurance companies. Institutional investors buy and sell a significant amount of stocks, bonds, or other securities. Many institutional investors fulfill an active monitoring function within the corporate governance system due to their main shareholder influence, strategic goals, and increased financial experience and expertise (Bebchuk, Cohen, & Hirst, 2017).
During the last years, there has been a major increase in the importance of environmental, social, and governance (ESG) issues among these investors, e.g., ESG ratings like the Dow Jones Sustainability Index or the FTSE4Good Index and Principles for Responsible Investment (PRI) guidelines by the United Nations (UN) (Dyck et al., 2019). ESG can be classified as "a business organization's configuration of principles of [environmental,] social [and governance] responsibility, processes of [environmental,] social [and governance] responsiveness, and politics, programs and observable outcomes as they relate to the firm's societal relationships" (Wood, 1991, p. 693). The literature stresses that both positive and negative relationships between institutional ownership (IO) and ESG can be realistic (Faller & Knyphausen-Aufseß, 2018). Traditional agency theory assumes that institutional investors are only interested in the short-term financial performance, and they will prevent intensive ESG expenditures to enhance their shareholder value (Bebchuk et al., 2017; Barnea & Rubin, 2010). However, institutional investors will demand ESG issues if it lowers the risk of an investment (Mahoney & Roberts, 2007). As ESG risks have a major impact on firm value (e.g., climate risks), institutional investors may prefer firms with high levels of both financial and ESG performance (Graves & Waddock, 1994).
Given the current relevance of the topic, the following paper synthesizes and discusses empirical research on the impact of IO on ESG and vice versa, as also other ownership types (e.g., managerial ownership) and other stakeholder groups (e.g., customers, suppliers, employees) can foster the ESG activities of a firm. Given the dominance of empirical research on institutional investors (Obermann & Velte, 2018) and its huge power, IO's concentration is justified in this literature review. Therefore, empirical-quantitative (archival) articles are included that analyze 1) the impact of IO on ESG performance and disclosure, 2) the impact of ESG performance and disclosure on IO as a bi-directional relationship, and 3) the moderating and mediating influence of IO on the firms' financial consequences of ESG. To achieve academic quality, 81 double-blind, peer-reviewed journal articles were considered.
The empirical relationship between IO and ESG is very complex. According to principal agent theory, institutional investors are classified as a homogeneous shareholder group with a clear focus on financial performance. It is not surprising that prior meta-analyses dominantly analyze the relationship between ownership structure and financial performance (e.g., Sundaramurthy, Rhoades, & Rechner, 2005; Sanchez-Ballesta & Garcia-Meca, 2007). IO represents a key monitoring instrument as part of the external corporate governance system. ESG issues will not be important for every type of institutional investor and thus not automatically pushed. As prior research stated contradictory empirical results (Faller & Knyphausen-Aufseß, 2018), institutional investors' preferences are heterogeneous in business practice. A more detailed analysis of IO and their different strategic goals are necessary. In contrast to prior research, a clear differentiation between IO ratio, nature, and type as main categories of institutional investors is conducted in the present literature review. Regarding institutional investors' nature, long-term versus short-term, active versus passive, and financial versus sustainable IO can be found in practice. Sustainable (non-financial) and long-term oriented institutional investors will include ESG issues in their decision-making in line with financial goals.
because of conflicts of interests between both parties (Harris & Bromiley, 2007). To decrease those agency conflicts, the investors will implement monitoring and incentive-alignment mechanisms, e.g., a useful management compensation system.
As institutional investors are interested in maximizing firm value, they may also include ESG goals, e.g., an appropriate ESG performance and a transparent ESG report. Many researchers, e.g., Goranova and Ryan (2014), stress that IO and ESG are interdisciplinary topics. Modern agency-theoretical approaches, e.g., stakeholder agency theory (Hill & Jones, 1992), neglect the assumption of homogeneity within institutional investors. Thus, it depends on the nature and the special type of institutional investors, whether IO is significantly related to ESG or not. In line with portfolio theory, shareholders' investment decisions are related to risk and return (Hoq, Saleh, Zubayer, & Mahmud, 2010). As institutional investors have a main focus on financial results and investment risks, socially responsible investors (SRI) and long-term investors as special IO nature, explicitly consider ESG aspects in their investment decisions (Clark & Hebb, 2004). Thus, long-term versus short-term investors, SRI versus purely financial investors, active versus passive institutions fulfill heterogeneous investment strategies.
In line with IO nature, specific IO types are focused on quarterly earnings and act as traders, whereas others include non-financial aspects in their decisions, as it influences the risk-return-profile in the long run (Johnson & Greening, 1999). Mutual funds, unit trusts, investment trusts, and investment banks are normally short-term institutional investors with average holding periods of fewer than two years (Graves & Waddock, 1994). In contrast to this, pension funds and life insurance companies are normally long-term investors and hold large shares in portfolio companies (Fauzi, Mahoney, & Rahman, 2007). These investors may have problems to find new beneficial investments, as they have typically diversified holdings across a broad number of firms (Hoq et al., 2010). Thus, long-term investors are active monitors of the management; they are expected to prevent man-agement activities that could erode firm value (Cho, Lee, & Pfeiffer, 2013).

RESEARCH FRAMEWORK
As this study relies on IO's impact on ESG and vice versa, IO can be included as independent, dependent, moderating, or mediator variable. This literature review's first IO variable is the simple ratio of institutions within total corporate ownership (IO ratio). Most of the studies in this review still rely on the IO ratio compared to other ownership types with heterogeneous measures (e.g., top institutions, top three, top five).
In line with IO ratio, heterogeneity within institutional investors can be addressed by its nature and type. Empirical research on different types of institutions and their impact on ESG and vice versa is growing since the last decade. In this literature review, different IO nature proxies are recognized to analyze whether institutional investors demand a high ESG performance and disclosure. Most of the included studies on IO nature focused on long-term versus short-term IO. A classical time-based separation has been made by Bushee (1998). Based on the past investment patterns in portfolio turnover, diversification, and momentum trading with nine variables, institutions are classified into transient, dedicated, and quasi-indexers. Bushee (1998) assumes that transient institutional investors mostly rely on short-term goals in contrast to dedicated institutions' longterm motivation. Regarding the range of activity, Brickley, Lease, and Smith (1988) have also conducted a well-used proxy of IO nature. The authors differentiate between pressure-sensitive (insurances, banks, and non-bank trusts), pressure-resistant (public pension funds, mutual funds, endowments, and foundations), and pressure-indeterminant institutions (private pension funds, brokerage houses, investment counsel firms, miscellaneous financial service firms, and unidentified institutions). According to this differentiation, pressure-sensitive institutions are passive, and pressure-insensitive are active institutions in their monitoring strategy. Pressure-sensitive institutions have current or potential business relations with corporations that create potential conflicts of interest with their fiduciary obligations more frequently than other institutional investors. Thus, monitoring activity is reduced.
Rather, current classification of IO nature is the explicit recognition of ESG goals as a complement to institutional investors' traditional financial focus. Dyck et al. (2019) differentiated institutional investors by whether they signed the UN Principles for Responsible Investments (PRI) or not and classified the signatories as socially responsible investors (SRI). Signing the UN PRI commits institutional investors to active engagement and the consideration of ESG issues in their investment decisions. Socially responsible investors possess a homogeneous set of ethical values, according to which they engage in active oversight regarding ESG strategies. As invested financial stakeholders, they are incentivized to monitor managerial behavior in alignment with a multi-attribute value function regarding firms' financial and ESG performance.
In comparison to IO nature, IO type as the third main category of IO proxies focuses on special groups of institutional investors, e.g., pension funds, mutual funds, or investment funds, and their different investment strategies.
ESG proxies are separated in ESG performance and ESG disclosure. Prior research recognized a variety of different proxies. ESG performance or sub-pillars, e.g., environmental performance, social performance, or carbon performance, are normally related to external databases (e.g., Thomson Reuters). Few studies also include ESG spendings, corporate philanthropy donations as ESG performance, or a simple dummy variable (e.g., CDP participation, or categorization as ethical or sin firm). ESG disclosure proxies and their sub-pillars, e.g., carbon disclosure, environmental disclosure, and integrated reporting, mostly rely on individual content analysis and scoring methodology.
The first research question examines whether IO ratio, nature, and type have an impact on ESG performance and disclosure. The second research question investigates the effect of the opposite relationship between ESG performance and disclosure and IO. Finally, empirical research on the firms' financial consequences of ESG performance and disclosure is included by recognizing IO as a moderator or mediator variable. Figure 1 presents a research framework to analyze the main streams of research. Table 1 gives an overview of the included IO and ESG proxies in this literature review.

Data selection
This structured literature review of empirical articles focuses on the bi-directional relationship between IO, ESG performance, and disclosure. Moreover, IO is recognized as a moderator or mediator variable of the relationship between ESG and firm value. Several keywords are used, e.g., 'Corporate Social Responsibility (CSR)', 'Environmental, Social and Governance (ESG)', 'Environmental Performance', 'Carbon Performance', 'Financial Performance', 'Environmental Reporting', 'Carbon reporting' and related expressions in connection with 'institutional investors', 'institutional ownership', 'sustainable responsible investors', and related expressions. The limitation of the period of empirical research on that topic was not useful.
In a second step, and as in other related literature reviews (Obermann & Velte, 2018), the keywords are applied to five major academic journal databases -ISI Web of Science, ScienceDirect, SAGE Journals, Emerald Insight, and Wiley Online Library, Google Scholar. For the field of business economics, only peer-reviewed journal articles in the English language were considered. This procedure resulted in 185 potentially relevant empirical papers. Contents of their abstracts were examined, and 69 articles whose topics did not fit the research question were excluded. Another 35 articles were dropped as these studies did not refer to the respective research method (empirical-quantitative (archival) research). The final sample of 81 papers was stated. An overview of the cited studies can be found in Table 2, organized by publication year, by region, by journal, by content, by IO variable(s), and by regression model(s). Note: * some studies include more than one IO variable and regression model.

Content analysis
The first study in the review was performed in 1994, and the field has grown considerably during the last few years (Panel A). Most of the included studies have been conducted for the U.S.  (10) and Corporate Social Responsibility and Environmental Management (9). Furthermore, accounting and corporate finance journals have often been used as a publication medium (32). Table 2 also shows that prior research mainly focused on the impact of IO on ESG (55) and not the other way round (14) (Panel D). Moreover, only a few studies include IO as moderator variables (11), and only one study in our sample addresses IO as mediator. The most relevant IO proxies are IO ratio (49) and IO nature (38), as Panel E indicates. As institutional investors are heterogeneous in their ESG interests, the use of an overall IO ratio is not precise. The research topic is related to major endogeneity concerns, especially due to reversed causality and omitted variable bias. While Panel F indicates that OLS (40) and panel regressions (24) are the most common, advanced regression models, e.g., two-stage least squares The included studies in the literature review are summarized in detail in Tables 3-5. Table 3 in- cludes only papers on the impact of IO on ESG. Table 4 focuses on empirical-quantitative research on the relationship between ESG performance, disclosure, and IO, assuming a bi-directional relationship. Finally, Table 5 lists the studies that include IO as moderator and mediator variable on IO's impact on firms' financial consequences.

ESG performance
Most of the included studies in this literature review are related to IO's influence on ESG performance. IO

ESG disclosure
In line with ESG performance, most of the included studies address IO ratio, assuming homogeneity within institutional investors. A positive impact of IO ratio on ESG disclosure (          . As these studies rely on IO ratio, there is a low amount of knowledge on IO nature's impact and type on integrated reporting (quality) and vice versa. Moreover, IO variables should be included as moderator and mediator variables on the firms' financial consequences of integrated reporting. Assuming that integrated reporting decreases the risks of greenwashing and information overload by the integrated thinking approach, IO activism might put pressure on the top management to compile an integrated report of high quality (Raimon et al., 2020). Given the low research density, future studies should re-assess SRI investors' effect or examine whether there is an association between investors' signing of the UN PRI and the publication of an integrated report with high quality.
In line with integrated reporting, institutional investors demand a transparent and credible ESG disclosure. Top management can provide external independent assurance of ESG reports by a third party voluntarily. International standard setters on ESG disclosure, e.g., the Global Reporting Initiative (GRI, 2018), explicitly recommend implementing assurance to increase the decision usefulness of ESG reports. Assurance of ESG disclosure has also been established as a key topic in empirical research. Prior literature reviews (e.g., Velte and Stawinoga 2017b) stressed the complexity of corporate governance-related influences of sustainability assurance within this field. However, there is a lack of knowledge on the impact of IO on the decision to conduct an ESG assurance, the choice of the assurance provider (accountants versus consultants), and the range of assurance (reasonable versus limited assurance level). Up to now, few studies included ownership concentration in empirical research on ESG assurance. Miras-Rodriguez and Di Pietra (2018) assume that block holders demand assurance of ESG disclosure and found positive results. In contrast to this, Kuzey and Uyar (2017), De Beelde and Tuybens (2015), Castelo Branco, Delgado, Gomes, and Pereira Eugenio (2014) and Ruhnke and Gabriel (2013) did not state any significance between ownership concentration and ESG assurance.

CONCLUSION
Institutional investor activism has become a major corporate governance issue from research, regulatory, and practice view (Bebchuk et al., 2017). While prior empirical research stressed the heterogeneity of IO and found heterogeneous results (e.g., Faller & Knyphausen-Aufseß, 2018; Friede, 2019), there is no literature review up to now on the bi-directional relationship between institutional investors and ESG and possible moderating influences of IO. Thus, this paper aims to discuss the connection between IO ratio, nature, and type as institutional investor categories and ESG performance and disclosure. 81 empirical-quantitative (archival) studies published in peer-reviewed journals are included. In contrast to prior literature reviews, a clear differentiation between 1) the impact of IO on ESG variables, 2) the influence of ESG on IO proxies, and 3) the moderating and mediating influence of IO on firms' financial consequences of ESG activities is included.
Regarding IO nature, long-term, sustainable, and active investors are assumed to strengthen ESG performance and disclosure, and those investors are also attracted by high ESG performance and disclosure to invest their money in a sustainable firm. Moreover, specific groups of institutional investors, e.g., pension funds or mutual funds, should have a different attitude on ESG performance and disclosure. According to the literature review, contradictory and insignificant empirical results in the review sample and a low amount of studies in specific research questions (e.g., on mediator analyses and ESG disclosure impact on IO) can be found. However, according to this literature review, there are indications that 1) long-term IO increases ESG performance, and 2) ESG performance leads to a higher IO ratio.
This analysis is not only relevant for research but also standard setters and business practice. First, longterm institutional investors have a huge impact on ESG performance, and ESG performance leads to a higher IO ratio. Recent regulations on shareholder rights and ESG from an international perspective, e.g., the European "Green Deal", will promote this IO activism in the future. Executive directors and audit committees should be aware of the increased power of institutional investors in ESG activities. Thus, ESG-driven monitoring of top managers is not only conducted by non-executives within the board, but also by some groups of institutional investors. This research topic is not only relevant in US-American one-tier systems with a traditional shareholder value focus but also two-tier systems with a stakeholder value view. As part of their investor relations management, executives should carefully analyze the current status and the development of IO within the firm, especially its nature and types. Given the current climate change discussion and the Fridays for future debate, many institutional investors (e.g., pension funds, SRI) connect financial and climate issues. Thus, financial performance and ESG performance represent relevant figures for investor relations management. The demand for successful integration of financial and ESG key performance indicators (integrated reporting) is likely to increase during the next years.