The second day’s session focused on the role of ESG investors in achieving net-zero goals. As stressed by the discussion chair, Professor Andreas Engert (Freie Universität Berlin and ECGI), although ESG investing has recently risen in prominence, there is still a concern that climate-conscious investors will divest from brown business targets, letting less climate-conscious investors seize profitable ‘brown’ investment opportunities. The principal issue is therefore how ESG investors can best promote a transition to a non-carbon economy.

Second Session. Climate Change: Exit or Voice

3. Exit vs Voice

This, amongst other issues, was addressed in the core presentation based on the paper ‘Exit vs Voice’ by Eleonora Broccardo, Oliver Hart and Luigi Zingales. The presenter and co-author, Professor Oliver Hart (Harvard University and ECGI) suggested as a point of departure a divergence in the motives of ESG-minded investors and consumers—whilst some act for moral reasons through divestment or boycotts, others aim to achieve a particular effect (so-called ‘consequentialists’). The strategies available to consequentialist socially responsible agents (CSR investors) may be split into two categories: ‘exit’ (divesting from polluting companies, refusing their products, refusing to work for them) and ‘voice’ (engaging with management, e.g. by voting). The core conclusion of the research is that exit and voice may respectively be less and more effective than initially anticipated.

To demonstrate the logic of this conclusion, the authors used a model under which a company is established at time t0, production decision is made at t1, while t2 is when production and consumption take place. The model works on the assumption that consumers and investors act in a ‘replica economy’ competitive environment, in which investors are constantly risk averse and hold fully diversified portfolios, by way of holding marginal stakes in all companies pro rata. Pollution becomes relevant at t1 (production decision) where a firm discovers a climate change problem and the concomitant fixed costs it may incur to mitigate or eliminate the problem and become ‘clean’. Although government strategies may prescribe investor behaviour in such a scenario, the paper analyses what happens when choices regarding social action are left to individual investors or consumers.

Divesting and boycotting options available to investors would have similar outcomes in the model. In an interior equilibrium, ‘dirty’ and ‘clean’ firms have the same market value but different profitability forecasts, because clean firms must incur the cost of becoming ‘clean’. Accordingly, by divesting from a brown asset and buying a clean one a CSR investor faces a loss. An investor, it is argued, can justify such loss if it is offset and exceeded by the ‘gain’ that investor may reap from any positive effect the divestiture will have on parties external to it (such as other investors, consumers, or the environment at large). Whilst divesting may benefit the price of shares in ‘clean’ firms, increasing such firms market capitalisation and thus persuading investors to divest in order to optimise their portfolios, the authors argue that this effect is negligible in determining the likelihood of investor decisions. Thus, an investor is only likely to divest its holdings in brown firms in instances where, notwithstanding the monetary loss it will incur, it derives value from the benefits divestiture has on the environment which are equal to or greater than the losses incurred, such that divesting has a net positive effect for that investor. Accordingly, if investors are not altruistic and pollution is not overly inefficient (so that the benefits of divesting are less than the monetary losses faced by an investor), the only equilibrium is one where CSR investors do not divest from brown firms.

An alternative to divesting and boycotting is investors engagement, primarily through voting. Before a CSR investor takes a decision to convert the firm into a clean one, the economic effects of the decision will be considered, including the effects on (1) investors, (2) other shareholders, (3) consumers, and (4) the environment. The effect on investors, represented by their respective capital losses, is negligible as investors are well-diversified (per the founding assumption of the paper). However, the majority shareholders as a whole will incur a loss via the payment of a fixed cost by the firm in order to become ‘clean’. Consumers’ condition will not change, since the price for a unit of product is fixed. The effect of the decision to divest on the environment is positive. Accordingly, the investor would vote in favour of converting the firm to ‘clean’ if the positive effect thereof on the environment is greater than the fixed cost of conversion which it bears in its capacity as shareholder. Unlike divesting and boycotting cases, this conclusion remains correct for any investors whose degree of social responsibility is greater than zero.

Professor Hart concluded that although the model may (i) omit some important benefits of exit or boycotting campaigns, (ii) disregard scenarios where voice is not an option, and (iii) ignore the possibility of voice limitation due to regulation, it nevertheless shows that engagement may be more efficient than exit or boycotting campaigns. Thus, opportunities for engagement should be sought and enhanced.

The discussant, Professor Laura T. Starks (McCombs School of Business and ECGI) highlighted some key take-aways from the paper, including the importance of the distinction between socially responsible, consequentialist investors and traditional, moral investors, a distinction that explains the preference for voice, rather than exit strategies. She then focused her discussion on a number of questions that, according to her, would require further elaboration, such as: what are the costs of exit for climate change reasons? Outside of the US are governments more efficient at dealing with externalities than investors? What is the effect of worker boycotts versus recruitment, retention and consumer boycotts? Professor Starks also raised the issue of the cost of divesting from an entire industry compared to that of divesting from a single firm and questioned the role of mutual funds in the model as being the primary submitters of shareholder proposals, suggesting that such funds usually play rather a reactive than a proactive role. Finally, Professor Starks pointed out the need to take social issues into account, in addition to climate change problems, and suggested to elaborate on the notion of socially responsible consequentialists.

The discussion continued with questions and comments from the audience, raising various issues, including (i) how to reduce voting costs (through decision-making delegation, technologies etc.) so that shareholders are incentivised to vote on climate issues, and (ii) an imperfect competition scenario where large shareholders bear increased costs and are less incentivised to vote in favour of loss-making ‘clean’ decisions. The participants also pointed out that firms’ workers may have an important role in incentivising the firm’s conversion to cleanliness.


4. Panel Discussion

A panel discussion led by Professor Georg Ringe (Hamburg University and ECGI) followed the paper presentation, with contributions from Professor Ann-Kristin Achleitner (TUM School of Management), Professor Clemens Fuest (Ifo Institute for Economic Research), Professor Rajna Gibson (University of Geneva and ECGI) and Professor Jeffrey N. Gordon (Columbia Law School, University of Oxford, and ECGI).

Rajna Gibson addressed the question of whether institutional investors actually implement ESG strategies which they pledge to. She referred to the paper ‘Do responsible investors invest responsibly?’, co-authored by herself, Simon Glossner, Philipp Krueger, Pedro Matos and Tom Steffen. The paper analyses the UN-sponsored Principles for Responsible Investment (PRI) signatories’ commitment to implement ESG goals. The researchers computed ESG scores for investors’ equity portfolios and used the PRI Reporting Framework to identify the intensity of investors’ commitment to ESG principles. PRI signatories were divided into 3 groups: the fully committed, the partially committed, and the non-committed (or non-reporting). One of the main findings of the paper is that in the US, unlike in Europe, non-signatories have better ESG results than non-committed signatories (which indicates greenwashing), consistent with another paper finding, namely that the main driver for signing and compliance for some (namely, the non-committed) US signatories is commercial.

Clemens Fuest switched the focus of the discussion to the question of the role of public policy. As he illustrated with reference to CO2 pricing regulation in the US, public policy may be designed such that self-serving investors may still ‘do the right thing’. The main concern is that private investors may search for their own, distinct ESG policies, raising two questions: if private investors do care about the consequences of their investments, does this mean that the current public policy economic approach is wrong? And if so, how should public policy be designed?   

Ann-Kristin Achleitner reverted to the discussion on voice and exit strategies. Importantly, as she highlighted, in Germany the two-tier board system (with the supervisory board including representatives of workers) may ensure the exercise of a voice strategy without the need to apply boycotting. This is important as the ‘S’ (social) element of ESG gains further significance alongside the ‘E’, as ESG strategies work best if they are combined with employees’ engagement. As different industries face different challenges, voice strategies may be more suitable to understanding those challenges, by providing a means by which the challenges can be effectively discussed. In addition, there is a flip side of each ‘exit’ strategy, which is ‘entry’. Currently, traditional investors are much quicker to exit companies than ESG investors are to enter, which signals that ‘net exit’ issues should receive more attention.

Jeff Gordon offered a different take on the ‘portfolio diversification’ that was pivotal in the ‘Exit vs. Voice’ paper. What drives much of the result is that diversified investors do not face serious costs from a shareholder vote to force a firm to internalize externalities, for a polluting firm to become ‘clean’. But such a firm-specific approach will not succeed in addressing an issue like climate change, because the necessary measures to mitigate climate change risk will result in disruptions at many firms and will require costly adjustments throughout various industrial sectors represented in the portfolio. Investors cannot escape those costs through diversification. Thus the ‘Exit v. Voice’ paper does not offer guidance in addressing the questions that matter most.  Gordon’s view, based on his theory of Systematic Stewardship,’ is that a diversified investor should break down portfolio risk into its two components, expected return and systematic risk, appreciating that the investor’s objective is to maximize risk-adjusted returns. In addressing a risk like climate change, which can result in losses across the entire portfolio, a self-interested investor should see the private value of risk-reduction even in circumstances where firm-specific returns might be lower. The ‘altruism’ that animates the ‘Exit vs. Voice’ paper is necessarily bounded; that’s why diversification is important to the story. But addressing a systematic risk like climate change—which cannot be escaped by diversification—requires a different story, said Gordon, in arguing for a Systematic Stewardship framework.

The panel then discussed the role of minority shareholders in using voice tools and the hurdles they need to overcome to do so (including the absence of a platform to air their concerns, in addition to technical and cost issues). The panellists also agreed that non-structured and non-standardised databases, criteria, and metrics overcomplicate the assessment of outcomes and effects of ESG principles implementation. Thus, despite the criticism levelled at the EU Taxonomy arising from the impossibility of classifying all types of business activity ESG criteria, it was acknowledged that some level of standardisation is needed with respect to climate change issues. A further conclusion was that the role of governments ought not to be excluded from the assessment, although the actions of shareholders do have a role and an impact.

While the main focus of the discussion was necessarily the panellists’ contributions, a number of interesting questions and observations were raised in the conference chat platform. The attendees discussed, amongst other things, whether ESG criteria may impact the value of a clean company regardless of Hart et al’s model’s assumption of equal value for both ‘clean’ and ‘dirty’ firms, as well as how mutual funds whose investors have differing preferences should exercise such investors’ individual ‘voice’. Some participants noted that the model should be further elaborated to clarify its application to different ownership structures.


Gabriel Acuna Csillag is an Advisor on Sustainable Finance at the Financial Markets Commission of Chile, and Founder and Former President of the Oxford Sustainable Finance Students’ Society.

Stepanyda Badovska is an Associate at Baker & McKenzie - CIS, Limited, and Founder and Former Vice President of the Oxford Sustainable Finance Students’ Society.


Part 1 of this post is available here.

This post is part of the series ‘Business Law and the Transition to a Net Zero Carbon Economy’. This series consists mainly of posts summarizing papers presented and presentations made at the 5th Annual Oxford Business Law Blog conference on ‘Business Law and the Transition to a Net Zero Carbon Economy’ which took place online on 25 to 27 May 2021. The recordings are available here.