This is the accepted manuscript for the publication:
Rees, W., & Rodionova, T. (2015). The Influence of Family Ownership on Corporate Social Responsibility: An International Analysis of Publicly Listed Companies. Corporate Governance: an International Review. 10.1111/corg.12086
The Influence of Family Ownership on Corporate Social Responsibility: An International Analysis of Publicly Listed Companies
William Rees and Tatiana Rodionova
* Address for correspondence: Tatiana Rodionova, The University of Edinburgh Business School, 29 Buccleuch Place, Edinburgh EH8 9JS, UK. Tel.: +44(0)131 6503789; E-mail: .
Acknowledgements
The authors greatly appreciate helpful feedback received at the paper development workshop of the Family Business Conference 2014 in Singapore. The authors are also thankful to the participants of British Accounting and Finance Association (BAFA) Annual Conference 2014, to workshop participants at the universities of Amsterdam and WHU and to Prof. Jo Danbolt and Susan Hancock for helpful comments and advice. Finally, the authors would like to particularly thank the Associate Editor of the special issue and three anonymous reviewers for their insightful and useful suggestions.
The Influence of Family Ownership on Corporate Social Responsibility: An International Analysis of Publicly Listed Companies
Manuscript Type: Empirical
Research Question/Issue: We investigate the impact of family equity holdings on three indicators of corporate social responsibility: environmental, social and governance (ESG) rankings. We further evaluate how firm governance mediates the effect of family ownership on environmental and social improvements and how national governance systems influence the response of family holdings to ESG.
Research Findings/Insights: Based on a sample of 23,902 firm year observations drawn from 2002 to 2012 covering 46 countries and 3,893 firms, our findings show that both closely held equity and family ownership are negatively associated with ESG performance. When we control for governance, closely held equity is no longer associated with environmental and social rankings but family ownership retains a significant negative association. These results are strong and consistent across liberal market economies (LME) whereas coordinated market economies (CME) exhibit generally weaker results and considerable diversity. Japan stands out as different from the other countries examined in depth.
Theoretical/Academic Implications: Our results are consistent with agency relationships driving decisions concerning ESG commitment in LMEs. They also emphasize the role of institutional differences given the weak and variable association between ownership and ESG in CMEs. We show that families may be able to influence decisions, possibly through participation in management, despite normally effective governance constraints. As the impact of ownership and governance varies across economies and ownership type this implies that both agency and governance should be evaluated in the context of the economic environment.
Practitioner/Policy Implications: Our results offer insights to regulators and policy makers who intend to improve ESG performance. The results suggest that encouraging diversified ownership is particularly important in LMEs, that improvements in governance may benefit social and environmental performance where equity is closely held by institutions, but that governance may be less effective in the presence of family ownership.
Key words: Corporate Governance; Corporate Social Responsibility; Environment; Family Firms; Closely Held Equity.
INTRODUCTION
We examine the impact of closely held equity, and in particular family held equity, on corporate social responsibility as reflected in publicly available scores of the environmental, social and governance (ESG) performance of firms. These issues have been receiving increased attention from regulators, investors and businesses, and the question of what drives or hinders environmental, social and governance performance is gaining prominence (Aguilera, Rupp, William, & Ganapathi, 2007; Campbell, 2007). Prior work has addressed this question mostly from an institutional (Aguilera & Jackson, 2003; Campbell, 2007; Ioannou & Serafeim, 2012; Kang & Moon, 2012; McWilliams & Siegel, 2001) or a resource perspective (Arora & Dharwadkar, 2011). However, much of the influence on ESG investment comes from the shareholders of the firm, particularly influential blockholders who monitor management (Barnea & Rubin, 2010; Shleifer & Vishny, 1986). While there is some evidence of the negative impact of blockholdings on corporate social responsibility, little attention has been given specifically to the influence of family equity holdings (Barnea & Rubin, 2010; Mackenzie, Rees, & Rodionova, 2013; Rees & Rodionova, 2013). This is potentially important as family owners often retain control of the company management thereby increasing their influence on decision-making (Faccio & Lang, 2002; La Porta, Lopez-de-Silanes, & Shleifer, 1999; Le Breton-Miller & Miller, 2006; Silva & Majluf, 2008).
This paper aims to fill a gap in the literature by directly investigating how family ownership affects the ESG practices of firms. We expect families to differ from other closely held blockholders as they have their personal wealth invested in the firm and have both financial and socio-emotional incentives with regards to the firm’s performance and viability (Kappes & Schmid, 2013). Further, we are interested to see whether and how corporate governance and the national economic system influence the relationship between family ownership and environmental and social performance. Corporate governance aims to balance interests of different shareholders and stakeholders of the firm and has been shown to influence corporate social responsibility (Aguilera, Williams, Conley, & Rupp, 2006; Jo & Harjoto, 2011). It can therefore affect the power that family owners have over the decision-making in the firm, and may alter the impact of family equity on environmental and social investments. We also investigate the impact of differences in the national economic systems that have been shown to influence corporate behavior with regards to social, environmental and ethical issues (Campbell, 2007; Kang & Moon, 2012). Whilst the national governance systems may influence the base line ESG performance in an economy, we also investigate whether these differences influence the impact of equity ownership on ESG.
Our sample consists of 23,902 firm/years drawn from 2002 to 2012 for 46 countries, mainly representing developed economies. The initial results are based on conventional regression techniques where the social, environmental or governance performance is modelled against the test variable identifying the closely held equity and a set of control variables accounting for year, industry, country, leverage, profitability, market-to-book and capitalization. We find closely held equity, and even more so family shareholdings, to be associated with lower ESG levels.
However, simply demonstrating an association between blockholdings and ESG scores does not show that the ownership structure causes the low ESG levels. In the case of family ownership reverse causality would imply that low levels of ESG attract family investors (or would encourage them to retain their family ownership) while high ESG levels prompt family owners to reduce their positions in the company. The latter implies that high ESG expenditures (expressed in high ESG ratings) could not be countered by direct action from powerful entrenched owners. We argue that in the case of influential and entrenched family ownership this is unlikely. We also address the causality issue empirically by using quantile regressions to investigate the relationship between ownership and ESG levels for cases where the ESG level is high. Here we obtain a stronger negative impact of family ownership on ESG rankings for firms with relatively high ESG scores. This is inconsistent with the argument that low ESG levels attract family ownership. Our results from the regression models are confirmed when we use a propensity score matching (PSM) approach that attempts to mitigate the endogeneity difficulty inherent in conventional regression modelling (Rosenbaum & Rubin, 1983).
This study offers several contributions. Firstly, we add to the literature that seeks to understand the relationship between ownership, governance and social responsibility. In particular, investor influence on corporate social responsibility is a growing but still an under-researched area, with most attention hitherto paid to institutional ownership (Cox, Brammer, & Millington, 2004; Sjöström, 2008). Our analysis highlights the role of closely held ownership, and in particular family ownership, as an important influence on ESG investments across an internationally diverse sample of firms.
Further, we contribute to the literatures on family governance and the interaction between internal governance, national governance and influential monitoring owners. Emerging literature suggests that there is an increasing trend for corporate governance to reflect the interests of stakeholders rather than solely shareholders of the firm (Jo & Harjoto, 2012; Ricart, Rodriguez, & Sanchez, 2005; Spitzeck, 2009). Here our findings highlight that family owners differ from other blockholders. While high levels of internal governance counterbalance the influence of the closely held equity on environmental and social investment, family owners are able to circumvent governance and preserve their influence regardless of the governance system. We also show that the impact of ownership on ESG varies depending on economic and institutional environments. In particular the negative impact of ownership on ESG is concentrated in LMEs. These findings are consistent with recent arguments suggesting that more attention should be paid not to a particular governance mechanism per se but to its relationship with other internal governance provisions and the national governance system (Aguilera et al., 2006; Yoshikawa, Zhu, & Wang, 2014).
From the theoretical perspective, our study enriches understanding of the motivations of family ownership. Family owners are thought to be extremely long-term in outlook and to be particularly concerned about the relationships with stakeholders to ensure firm survival and well-being (Le Breton-Miller & Miller, 2006; Le Breton-Miller & Miller, 2009; Yoshikawa et al., 2014). Our results, however, indicate that, when it comes to investments that foster social good but do not guarantee financial returns, families tend to act as financial wealth maximizers and tend to hinder such expenditures.
PRIOR RESEARCH AND HYPOTHESES
Closely Held Ownership and ESG Performance
Improving environmental, social and governance performance has become a major challenge for the corporations. Some ESG developments may advance operational performance (Berry & Rondinelli, 1998; Edmans, 2011), and academic evidence suggests that excellent ESG performance can be a source of competitive advantage (Aguilera et al., 2006; McWilliams & Siegel, 2001; Porter & van der Linde, 1995). For example, ESG projects may bring strategic benefits by improving relationships with stakeholders, including consumers, suppliers and employees (Becker-Olsen, Cudmore, & Hill, 2006; Bénabou & Tirole, 2010; Brekke & Nyborg, 2008; McWilliams & Siegel, 2001; Siegel & Vitaliano, 2007; Turban & Greening, 1997). Such developments have been shown to increase the market value of the firm (Jo & Harjoto, 2011; 2012). However, ESG projects require substantial investment, e.g. where a company aims to reduce toxic pollution or restrain the use of pesticides in the supply chain. Chatterji and Toffel (2010) argue that companies are particularly likely to improve their environmental practices where they offer ‘low-hanging fruit’ opportunities. However, many ESG developments may be negative NPV investments, e.g. withdrawal from drilling in an ecologically sensitive area, or price restraint by pharmaceutical firms. Agency theory suggests that managers may overinvest in such projects to serve their personal interests including a ‘warm-glow effect’ and favorable professional reputation (Barnea & Rubin, 2010; Bénabou & Tirole, 2010). Consequently, whether or not the firm invests in ESG improvements becomes a source of the conflict of interest between managers, shareholders and broader stakeholders of the firm (Cespa & Cestone, 2007; Orlitzky, Schmidt, & Rynes, 2003).
Academic evidence has argued that blockholders have the incentives and ability to monitor management to ensure that their interests are satisfied (Burkart, Gromb, & Panunzi, 1997; Demsetz & Lehn, 1985; Jensen & Meckling, 1976; Shleifer & Vishny, 1986). These large owners bear the cost of investment in ESG projects (Cox et al., 2004) and have the influence to constrain ESG initiatives. Prior literature offers some, albeit inconclusive, evidence of this negative relationship (Barnea & Rubin, 2010; Ioannou & Serafeim, 2012; Rees & Rodionova, 2013). Our main aim is to isolate the influence of family holdings specifically and compare it to the influence of other closely held equity, but we first revisit the question of the relationship between all closely held stock and corporate ESG levels.
Hypothesis 1. There is a negative relationship between closely held equity and ESG scores.
Families and ESG Performance
Family ownership is thought to be the most common ownership structure (La Porta et al., 1999). Faccio and Lang (2002) find that 44.29 percent of firms in 13 Western European countries are family firms while one third of public US firms can be regarded as controlled by families (Anderson & Reeb, 2003). Families and business groups are also seen to dominate corporate control in emerging markets (Silva & Majluf, 2008). Family owners are unique as they have very long investment horizons, they are undiversified and often occupy senior management positions (Anderson & Reeb, 2003). They also have a complex nexus of economic and personal motives with regards to the firm (Andres, 2008).
We argue that family block ownership will have a stronger negative impact on environmental, social and governance performance of firms than closely held equity in general. Our main rationale is that, as families have large and long-term ownership stakes in the firm, they will be particularly opposed to excessive ESG investment because it may not bring personal benefits. Firstly, this is consistent with the agency theory suggesting that large powerful owners channel the firm’s activity towards their benefits (Shleifer & Vishny, 1986). For example, families may prioritize stable cash flow in order to sustain a privileged lifestyle and ensure high dividend payments (Barth, Gulbrandsen, & Schone, 2005; Kappes & Schmid, 2013). Further, DeAngelo and DeAngelo (2000) find that family preferences for special dividends may influence their investment plans.
Secondly, families may oppose ESG investments as being value destroying (Barnea & Rubin, 2010; Rees & Rodionova, 2013). For example, the entrenchment view suggests that activities related to corporate social responsibility may be used by the entrenched management to advance their agenda by appealing to non-financial stakeholders (Cespa & Cestone, 2007). Given their large equity positions and financial interest in the firm, family owners may then be expected to constrain these developments.
Thirdly, family owners may consider themselves to be better monitors and to have better knowledge of the business. Indeed, their long historical presence in the firm and personal attachment can give them unique knowledge of the business and can make them well-placed to monitor decision-making (Anderson & Reeb, 2003; Le Breton-Miller & Miller, 2006; Sraer & Thesmar, 2007). In order to keep control and ensure family participation in management, family owners may forgo governance improvements (Andres, 2008; Hillier & McColgan, 2009).
Finally, although we previously argued that closely held equity in general will be associated with lower ESG performance, we argue that this effect will be less strong than in the case of families. Indeed, some blockholders may have motivations to resist ESG developments to a lesser extent. For example, the state has to address the issues of the quality of life and environmental concerns and may therefore not entirely oppose the efforts of the corporations regarding ESG improvements (Dam & Scholtens, 2012; Rees & Rodionova, 2013). More diversified equity holders, such as investment institutions, rely on multiple markets and may be affected by the economic consequences of political and social instability or environmental damage caused by a weak ESG position (Gjessing & Syse, 2007). Further, they have to preserve their reputation to sustain competition for funds, and exhibiting higher ESG involvement is an important way to improve corporate image (Godfrey, Merrill, & Hansen, 2009). Finally, pension funds are facing increasing pressure from their beneficiaries for socially responsible investment (Cumming & Johan, 2007). Conversely, family owners typically invest their wealth in the firm (Anderson & Reeb, 2003). Consequently, they do not have the reputational pressure for social and environmental responsibility from the beneficiaries.
Based on the evidence above, we argue that, as families have private wealth invested in the firm and have long-term commitment to this investment, they will be guided by personal benefits and will have less motivation to take ESG issues into consideration. We then predict the following relative relationships between ownership and ESG:
Hypothesis 2. The negative relationship between closely held ownership and ESG scores is stronger for family shareholdings than for those of other blockholders.
The Mediating Impact of Internal Governance
Agency theory suggests that internal governance aims to align interests of managers and investors, and to ensure that corporate decisions are made to offer shareholders the return on their investment (Gillan & Starks, 2007). Such a view implies that management would not support excessive ESG (Barnea & Rubin, 2010). More recently, it has been suggested that good governance serves (or should serve) the interests of various stakeholders (Aguilera et al., 2007; Spitzeck, 2009). Strong governance would then prevent corporations from having a detrimental impact on their stakeholders, e.g. via lost pensions or social instability, and improve their corporate social responsibility. Jo and Harjoto (2011; 2012) tested the relationship between corporate governance and corporate social responsibility and found support for the stakeholder interpretation of governance. Their rationale suggested that engaging in corporate social responsibility ensures favorable relationships between the company and its stakeholders and helps to avoid costly conflicts (Jo & Harjoto, 2012).
When a company has powerful blockholdings, these owners are likely to resist excessive ESG, viewing it as a costly investment with uncertain benefits to the company (Barnea & Rubin, 2010). Strong governance, however, is argued to prevent large blockholders from imposing their agenda and ensures that the power of different investors in the company is balanced (Shleifer & Vishny, 1986). Many investment funds are increasingly viewing corporate social responsibility as a fiduciary duty towards their trustees (Aguilera et al., 2007). Other diversified investors may view environmental and social programs as an important risk-management tool or they may favor reputational benefits of enhanced environmental and social performance (Gjessing & Syse, 2007). Consequently, if the governance system is strong, managers have more opportunities to take the interests of other financial and non-financial stakeholders into consideration, and the influence of large owners with closely held equity decreases: