Investment horizon and corporate social performance: the virtuous circle of long-term institutional ownership and responsible firm conduct

Abstract

We investigate the relationship between corporate social performance and institutional ownership. We distinguish between long-term and short-term institutional investors using holdings-based measures which directly capture the investment horizon of each institution. Our analysis shows that long term institutional investment is positively related to corporate social performance (mainly by an avoidance of investing in firms with significant controversies) whereas short-term institutional investment is negatively related to corporate social performance. Further investigation reveals that increased holdings of a firm by long-term investors are positively associated with its future corporate social performance. Hence, we provide evidence of a ‘virtuous circle’ between long term investment and social responsibility.

KEYWORDS:

• Corporate social responsibility
• CSR
• CSP
• sustainability
• institutional investors
• investment horizon

JEL Classifications:

• G23
• G31
• M14

Acknowledgments

The authors would like to thank attendees at the 5th Paris Financial Management conference and the PRI Academic Network Conference 2017, Chris Brooks and two anonymous referees for comments that helped improve the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Ioannis Oikonomou http://orcid.org/0000-0002-9935-8525

Notes

1 The acronym SRI is nowadays also used for Sustainable and Responsible Investing. Though subtle differences can be argued to exist between the two terms, the concept is largely the same.

2 US SIF: The Forum for Sustainable and Responsible Investment is a United States-based membership association that promotes environmentally and socially sustainable investment practices: http://www.ussif.org/about

3 The term CSP is usually used to capture the outcome-based measurement of a firm's stance towards CSR-related issues. In this paper, we will use CSR as the acronym for the main concept and CSP for variables related to its measurement.

4 ESG stands for Environmental, Social and Governance performance and is often used instead of CSR or CSP in recent literature. CFP stands for Corporate Financial Performance and it is a generic term used in the literature to encapsulate all the financial metrics that researchers have employed and tested whether they are correlated with CSP.

5 For a full list of the PRI Principles the interested reader is directed to https://www.unpri.org/about/the-six-principles

6 However, if the stock market is efficient, the stock price at any time point equals its fundamental value. As a result, the interests of short-term investors and long-term investors are aligned perfectly, which are to maximize the present value of all future cash flows of the firm. Therefore, under the efficient market assumption, investment horizon does not really matter.

7 The Bushee classification uses a factor and cluster analysis approach to classify institutional investors, based on a large number of trading behaviour variables and portfolio characteristics (e.g. portfolio turnover, diversification, and momentum trading).

8 Starting with the S&P 500 Index firms and the Domini 400 Social Index firms in 1991, KLD has expanded its coverage to incorporate the largest 1,000 US companies by market value since 2001, an expansion which advanced further in 2003 with the inclusion of the 3,000 largest US firms.

9 The Security and Exchange Commission (SEC) requires that all institutions operating in the US with discretion over 13F securities worth $100 million or more report all equity holdings greater than 10,000 shares (or $200,000) to the SEC at a quarterly frequency.

10 There are several reasons which could lead to a nominal institutional ownership rate being higher than 100% for a given firm. First of all, when investors share investment discretion, the security may be double counted (once for each institution). Secondly, when investors short sell a security, it will be recorded as a holding for both the lender and the borrower (short-seller) which will also lead to an overstatement of ownership. Thirdly, sometimes a firm's financial reporting date and institutional investors reporting date will not match perfectly. In this case, if a firm's total shares outstanding changed dramatically during this time gap, the base of ownership calculation could cause some data errors (Striewe, Rottke, and Zietz 2013). To minimize the effects of these factors, we follow the same treatment as in Yan and Zhang (2009). Our results are robust when keeping those observations with more than 100% total institutional ownership in our sample.

11 Following Servaes and Tamayo (2013), we exclude corporate governance from our CSP construction because it is a mechanism that aligns the interest between shareholders and managers rather than a concern that deals with social objectives and stakeholders other than shareholders. Human rights has also historically considered to be too broad of a category within KLD and not related to a particular group of stakeholders so is also excluded from the analysis.

12 Following Hillman and Keim (2001) and Oikonomou, Brooks, and Pavelin (2012), we assume that each type of social action is given equal weighting so that employee programmes, for example, are considered just as important as product safety and quality. Though not a perfect solution, in the absence of up to date data on the relative importance of each dimension this is what the literature has been employing.

13 Analytical descriptions of all the key dependent and independent variables have been placed in Appendix 1 for the sake of parsimony.

14 Including the lagged dependent variable could result in biased coefficient estimates if the true data generating process is static. To alleviate this concern, we remove the lagged dependent variable and re-estimate all regression specifications. Our (unreported) results are robust to the changes. We thank an anonymous reviewer for pointing this out to us.

15 See Waddock and Graves (1997), Neubaum and Zahra (2006), and Cao, Liang, and Zhan (2016).

16 The three examples are: (1). In 2017, PGGM engaged in dialogue with Tyson to improve its wastewater management; (2). PGGM engaged in dialogue with various companies in the mining, oil and gas sectors, including Glencore and FreePort McmoRan, to improve their assessment of potential human rights violations; (3). PGGM also voted against the excessive remuneration policy of McKesson. See https://www.pggm.nl/english/what-we-do/Documents/Summary_Responsible_Investment_Annual_Report.pdf

17 To corroborate our results and further show that increase in LIO leads to increase in CSR performance, we conduct regression analysis looking at the change of long-term IO and the change of CSR scores. The change regression results show that indeed long-term institutional investors increase CSR performance and they do so by immediately reducing controversies (see Appendix 3 Table A3.1).

18 Note that NLIO (short-term + medium-term IO) has a negative impact on almost all CSP dimensions.

19 For the detailed explanations of the Russell Index reconstitution, please refer to www.ftserussell.com/research-insights/russell-reconstitution.

20 Before 2003, only around 1100 firms were covered by KLD and therefore the overlap between KLD-covered firms and the Russell 2000 firms was limited. To address this concern, we re-estimate Table 10 using a subsample spanning from 2003 to 2012. The results become even stronger (see Appendix 3, Table A3.3).