Does ESG Negative Screening Work?

Robert Eccles is Visiting Professor of Management Practice at Oxford University Said Business School; Shivaram Rajgopal is Roy Bernard Kester and T.W. Byrnes Professor of Accounting and Auditing at Columbia Business School; and Jing Xie is Assistant Professor of Finance at Hong Kong Polytechnic University School of Accounting and Finance. This post is based on their recent paper.

Related research from the Program on Corporate Governance includes The Illusory Promise of Stakeholder Governance (discussed on the Forum here) by Lucian A. Bebchuk and Roberto Tallarita; Does Enlightened Shareholder Value Add Value? (discussed on the Forum here) by Lucian A. Bebchuk, Kobi Kastiel, and Roberto Tallarita; Companies Should Maximize Shareholder Welfare Not Market Value (discussed on the Forum here) by Oliver Hart and Luigi Zingales; and Reconciling Fiduciary Duty and Social Conscience: The Law and Economics of ESG Investing by a Trustee (discussed on the Forum here) by Max M. Schanzenbach and Robert H. Sitkoff.

Negative screening is broadly the process of finding and excluding stocks of companies, whose operations are seen as “unsustainable” from an environmental, social or a governance (ESG) standpoint (The U.S. SEC does not define a poor ESG stock. The European Sustainable Finance Disclosure Regulation (SFDR), on the other hand, defines a sustainability investment as “an investment in an economic activity that contributes to an environmental objective”). A popular version of negative screening, that is widely practiced by institutional investors sensitive to ESG considerations, is to exclude stocks of firms involved in the production of alcohol, tobacco, and gaming, and fossil-fuels such as coal and gas or oil (labeled as “sin stocks” or “excluded industries”). Reasons for exclusion vary and include moral values, the belief that this will put pressure on them to change or even put them out of business, or the conviction that the industry’s prospects are grim.

There is a relatively long literature suggesting that stock returns of sin stocks, traditionally covering alcohol, gaming, and tobacco, outperform the market (see Salaber 2007; Fabozzi, Ma, and Oliphant 2008; Hong and Kacperczyk 2009; Statman and Glushkov 2009; Durand, Koh, and Tan 2013). Hong and Kacperczyk (2009) identify firms in the alcohol, tobacco, and gaming industries as sin firms. Fabozzi, Ma, and Oliphant (2008) identify sin stocks as those classified in the six industries of alcohol, tobacco, defense, biotech, gaming, and adult services. Statman and Glushkov (2009) and Durand, Koh, and Tan (2013) define sin firms as ones operating in alcohol, tobacco, gaming, defence and weaponry industries. Following prior literature (e.g., Hong and Kacperczyk, 2009), we define a sin stock as a firm involved in the alcohol, gaming and tobacco industries. In addition, we also classify firms operating in weapons (gun industry) and carbon-intensive industries (i.e., coal and gas and oil) as sin stocks (Teoh et al., 1999). The intuition is that the investors who are willing to hold these screened investments expect a higher rate of return because of the social opprobrium attached to them and the exclusions they are facing from other investors which reduces the pool of capital available to them. However, relatively little is known about whether such negative screening by institutions impacts the valuation and fundamentals (e.g., operating, investing, and financing activities) of these companies. That is the question we study in this paper.

We begin by documenting that institutions indeed hold less equity in excluded industries, especially in alcohol, tobacco, and gas and oil in the past two decades, consistent with influential prior work (Hong and Kacperczyk, 2009). We focus on the 20 years since 2000 in the analyses of this paper to offer up-to-date insights into sin stocks. Negative screening has seemingly increased in the recent decade. However, a more careful matched sample approach suggests that, after controlling for basic firm fundamentals such as size, return on assets, firm age, capital expenditure, R&D, risk, past performance, dividend yield, and book to market ratio, the gap in institutional ownership between sin firms and other firms disappears. Hence, some institutions do not potentially hold sin firms, on account of intuitive fundamentals-based considerations, rather than purely on account of negative screening rules.

If exclusions were to hurt excluded industries, keeping firm characteristics constant, we should have observed lower firm valuations (e.g., Tobin’s Q). We find that the Tobin’s Q for sin stocks is similar to that of the rest of market after controlling for firm characteristics. The result is a bit nuanced in that sin stocks are associated with lower valuations relative to non-sin stocks in the recent decade since 2010 in an un-matched sample. However, in the matched sample, sin stocks have significantly higher Tobin’s Q than that of control firms. This enhanced valuation was driven by gas and oil firms, and it exists in both the recent and the earlier decade. Relatively exogenous shocks related to divestment (i.e., when a firm switches from a non-excluded industry to an excluded industry) and addition of stocks to a “good” ESG index confirm the finding of no apparent impact of negative screening on the Tobin’s Q of excluded industries. Our findings of a similar Tobin’s Q in the full sample and a higher valuation in the matched sample are different from Hong and Kacpercyzk (2009) who document that sin stocks suffer a lower valuation. Two possible reasons for that include: (i) the publication of their paper might have highlighted an arbitrage opportunity that was eventually realized by investors; (ii) our evidence is sensitive to the inclusion of gas and oil industry, a category left out by Hong and Kacpercyzk (2009). When we only focus on sin firms in the consumer sector (i.e., alcohol, gaming, and tobacco), as Hong and Kacpercyzk (2009) did, the valuation of sin stocks is statistically indistinguishable from that of matched firms.

Next, we examine how the industries that are subject to negative screening differ from those of other industries in other aspects. We find that sin stocks are associated with larger capital expenditure to sales, equity issuance, and acquisitions relative to other firms. In addition, sin stocks are associated with lower sales per unit of assets, lower real sales, and fewer employees, indicating that that the omitted variables could significantly alter conclusions about the fundamental and stock-related performance of sin stocks relative to other stocks. In addition, we find that sin stocks do not earn an alpha or abnormal monthly return estimated with Fama-Macbeth regressions. We also find that sin stocks in general do not differ from other stocks in their cost of new equity. Moreover, sin stocks do not differ in the likelihood of exiting the public market, the cost of raising new equity, and in the announcement returns around negative ESG news relative to non-sin stocks, casting further doubt on whether negative screening hurts sin stocks. However, the cost of new debt is higher for sin stocks. Investors submit more ESG proposals and a larger number of ESG proposals are passed, suggesting that investors in sin stocks likely exert higher pressure on firm management to improve its ESG performance or, alternatively, engage in virtue signalling. It is an open question how effective the latter is. themselves. In sum, evidence for the oft-stated hypothesis that negative screening hurts sin stocks depends on the design the researcher uses.

The complete paper is available for download here.

Both comments and trackbacks are currently closed.