1. Introduction
The pension plan at my institution – the Université de Montréal – is opaque. Its annual reportsFootnote 1 only provide information on about half the assets that the total volume of approximately C$ 4 billion is invested in. The information reported is limited to the top 10 investments by category: Canadian shares, US shares, and so on. This opaqueness is troubling.
Among the reported investments are several that I would prefer not to have in my portfolio: General Dynamics, a US arms producer, has been among the top 10 US shares; the pension plan owns shares worth about C$200 million in two of the biggest tar sands companies, Suncor and Canadian Natural Resources; the list could go on. Many of my colleagues share my dissatisfaction with this state of affairs.
Of course, our pension fund’s communications put a lot of emphasis on its policy for socially responsible investment (SRI). The annual report now contains a full four pages on the matter, and the fund’s managers insist on the fund’s active role in several investor activism campaigns to hold corporations to account. However, the fund refuses to go beyond engagement strategies by divesting from certain companies or sectors altogether. A survey among 1318 professors at Université de Montréal (463 responses, which corresponds to a response rate of 35.1%) revealed that a large majority of respondents are in favour of divestment from certain sectors.Footnote 2
The situation at our pension fund is not unique, but representative for many pension plans around the world, both in the private and public sector and independently of whether they are defined-benefit plans or defined-contribution plans. What should we make of this situation from an ethical perspective? What, if anything, is owed to the participants in collective pension plans in terms of transparency, socially responsible investment, and input into the investment strategy of their pension fund? Can the concept of fiduciary duties on the part of fund managers bear the weight that is put on it in this context, or could it be unacceptable that a principal’s money is managed by an agent without consultation of the principal? If it turned out that participants should have an input into the management of their retirement funds, what institutional arrangements are available to put this requirement into practice? These are the questions at the heart of this paper.
The argument proceeds in three steps. Against the background of a short survey of various SRI strategies, section 2 critically analyses the dominant paradigm in SRI, which favours engagement strategies over screening or divestment strategies. The upshot of this discussion is that the relative effectiveness of different SRI strategies is much more of an open question than a look at current practices suggests. This provides an argument for a more diversified strategy and mix of SRI instruments.
Whereas section 2 assumes that there is consensus about what is the right thing to do – positively influencing corporate behaviour – and that the controversy is limited to the instrumental question of how best to achieve this goal, section 3 relaxes this assumption. Here, I argue that reasonable pluralism exists about what constitutes the right thing to do in the context of SRI. I distinguish between different ethics profiles of investors: notably, between a broadly consequentialist profile, which indeed tends to share the objective stipulated in section 2 of positively influencing corporate behaviour; and what I call the clean-hands profile, which characterizes people who prefer not to hold certain types of investment independently of the consequences of holding them. In the face of reasonable pluralism, section 3 argues, collective pension plans that do not give their participants some kind of influence – through either voice or exit options – on their investments are unacceptable. While section 2 focuses on the relative effectiveness of different SRI strategies, section 3 focuses on their legitimacy in the context of collective pension plans.
What institutional options are available to give people an input into how their pensions are managed? Section 4 takes up this question. In a first step, it shows that re-interpreting the notion of fiduciary duties represents a necessary but in itself insufficient change. I then discuss four possible policies, one of which gives participants in collective pension plans voice in the investment strategy of their plan, whereas the other three centre on the idea of providing exit options and thus alternatives for investors.
I should emphasize up front that this paper does not aim to provide an answer to the substantive question of what constitutes a responsible investment and what does not. Instead, its goal is to develop the framework in which individual investors as participants in collective pension plans should be able to answer that question for themselves.
2. The effectiveness of SRI
Let us get some of the low-hanging fruit out of the way first: There is no justification for a lack of transparency in the management of pension funds, and some big pension funds do indeed practice full disclosure vis-à-vis their members.Footnote 3 Participants in pension plans have a right to know what their money is invested in. Since it is their money,Footnote 4 they are ultimately responsible for how it is invested, and transparency is a necessary condition for them to be able to live up to this responsibility.Footnote 5 Note that this requirement holds independently of what our substantive understanding of responsible investment might be. If all questions in the context of socially responsible investment were as uncontroversial as the one concerning transparency, this paper would be rather short.
Moving on to more difficult questions, what does it mean for an investment to be made in a socially responsible way? Defined broadly, SRI ‘means integrating non-financial factors’ – often categorized into environmental, social and governance (ESG) issues – ‘into the investment process’ (Foo Reference Foo2017: 3). To render this definition operational, more needs to be said on what this integration should look like and to what end it should be deployed. While a diversity of views has been formulated in this context, the arguably prevailing interpretation today sees SRI as an investment that makes for more socially responsible business practices.Footnote 6 Underpinning this consequentialist outlook on SRI is the idea that ‘corporations engage in morally impermissible behaviour that imposes harm on others’, ‘that individual investors typically share in the responsibility for the harms imposed by corporations in which they invest, and that they therefore have a moral obligation’ (Daskal Reference Daskal2013: 147) to invest in ways that live up to this responsibility.Footnote 7 Given its practical influence today, this consequentialist interpretation will be taken as given in the present section. However, as section 3 will show, we should not ignore other possible, non-consequentialist interpretations of SRI when it comes to the design of collective pension plans.
If the goal of SRI is to change corporate behaviour for the better, what tools are available to pursue this goal? We can distinguish the following four instruments here:Footnote 8
(1) Shareholder engagement: Through votes at the general assemblies of the corporations that pension funds are invested in, but also through more targeted initiatives, funds attempt to influence corporate behaviour.
(2) Negative screens: Certain kinds of investments, e.g. corporate sectors or individual companies, are excluded from the portfolio entirely, because the contrast between investor convictions and corporate behaviour is particularly stark and/or because the likelihood of change in corporate behaviour is particularly low.
(3) Positive screens: Rather than excluding corporations with undesirable activities, the portfolio is biased in favour of corporations that engage in socially desirable behaviour.
(4) Best in class: In each sector of the economy, only those corporations that best respond to the demands of SRI are included in the portfolio.
Let me make four observations on this taxonomy. First, viewed through the prism of the distinction between voice and exit (Hirschman Reference Hirschman1970), shareholder engagement falls into the former category, whereas negative screens and, to a lesser extent, best in class follow the logic of the latter. Positive screens are harder to categorize in terms of this distinction.
However, and this is the second observation, it would be misleading to present the four categories as mutually exclusive. Take the relationship between shareholder engagement and negative screens for example. Shareholder engagement without a credible threat of divestment in case of inertia is likely to be ineffective. Conversely, as Kolstad usefully highlights, a negative screen does not exclude engagement with corporations about a possible re-investment if certain changes in corporate behaviour are put in place (Kolstad Reference Kolstad2016: 48–9).
Third, I should make it explicit that I believe an SRI strategy based on any combination of the above instruments to sometimes be costly to investors as far as their financial returns are concerned. This is not to say that SRI will not also sometimes overlap with the strategy designed to maximize financial return (see Freshfields Bruckhaus Deringer Reference Freshfields2005). For example, if those anticipating that oil producers will turn into stranded assets in the medium term are correct, then divesting these assets will be both the right and the financially prudent thing to do. However, timing is important in investment, and it would be simply absurd to claim that maximizing one’s return will never include investing in undesirable activities. Thus, tensions do exist between return maximization and doing the right thing. As Irvine poignantly puts it, ‘[t]he inconveniences of being moral … in no way lessen our obligation to be moral, and this is a point investors must not forget’ (Irvine Reference Irvine1987: 242).Footnote 9
Fourth and finally, the investment strategies listed above obviously do not represent the only ways to influence corporate behaviour. Let me mention two examples of approaches that are situated further upstream, and one that is located downstream from investment narrowly defined. Among the approaches upstream from investment are the direct regulation of corporate activity as well as the influence on corporate activity through fiscal incentives. For example, in order to reduce carbon emissions, we might impose emissions caps on production or, alternatively, rely on a carbon tax. Since these strategies represent a stronger lever on corporate activity than the investment strategies discussed above, they are likely to be more effective compared to exit or voice strategies. However, to the extent that individual investors want to go further than governments – which is plausible given the tepid use of regulation and carbon taxes by most governments to date – the spotlight is on the four investment strategies listed above. In addition, as we shall see in section 2.1, the two approaches are connected in the sense that more widespread use of SRI strategies by individuals will plausibly increase the political feasibility of directly influencing corporate behaviour.
As to an example for an approach situated further downstream, consider the individual investor who, instead of trying to influence corporate behaviour, will maximize her return, but then donate the proceeds to causes that correspond to her ethical convictions (Zweig Reference Zweig1996: 64). Note that this individual has adopted a more general objective than stipulated thus far: Rather than aiming at changing corporate behaviour for the good, she strives to maximize positive consequences, period. For present purposes, we will stick to the objective of influencing corporate behaviour, in part because it keeps our focus squarely on investment as such.Footnote 10
Table 1 summarizes the place of investment strategies in the larger context of what one might subsume under the slogan ‘making the world a better place’ or, put more prosaically, under the heading of promoting positive consequences in an economic context.
Different approaches for ‘making the world a better place’ in the context of investment, and in the economy more broadly defined.
2.1. Questioning the engagement paradigm
The above sketch of the conceptual landscape surrounding SRI will now serve as background to interpreting SRI practice in recent years. Focusing on the distinction between shareholder engagement on the one hand, and various exit strategies on the other, I think it is fair to say that the current SRI paradigm favours the former over the latter. Despite some important divestment campaigns over the years, such as the divestment from South Africa to protest against apartheid, divestment from tobacco or, more recently, divestment from fossil fuels, it is fair to say that these campaigns still represent the exception rather than the rule.Footnote 11 Kolstad (Reference Kolstad2016: 53–4) provides a number of possible, and complementary, explanations for this move towards engagement strategies. They range from game-theoretic ones – SRI might be ‘a coordination game, where once a critical mass of investors have developed an engagement approach, it is rational for the others to follow suit’ (Kolstad Reference Kolstad2016: 54) – via strategic considerations – the non-binding character of engagement strategies might make it easier for corporations to engage in ‘greenwashing’, that is, continue their bad ways but sell them as socially responsible – to the idea that there are bureaucratic incentives that favour engagement.
My goal here is neither to suggest that shareholder engagement is generally ineffective,Footnote 12 nor to contest or endorse any of the explanations of its dominance cited in the previous paragraph, but rather to critically discuss three other potential reasons for why one might prefer engagement to screening strategies. These reasons are frequently cited to justify why exclusionary instruments are not part of SRI strategies. I will show that none of the three arguments holds water. If correct, this reasoning would seem to represent a pro tanto argument for a more diversified approach to SRI.
(a) Reduction of the investment universe? Footnote 13 Diversification is the golden rule of investment. Diversifying one’s portfolio offers better protection against risks, and thus will tend to deliver a better return on investment over time. Viewed from this angle, it is easy to see why negative screens look like a bad idea compared to shareholder engagement: Negative screens reduce diversification. By contrast, so the argument runs, engagement does not entail a reduction in the investment universe and is thus preferable, other things being equal.Footnote 14
Kolstad convincingly argues that this analysis is too superficial. If engagement is indeed successful in influencing corporate behaviour, this will lead to ‘a homogenisation of the investment universe in terms of activities and practices’ (Kolstad Reference Kolstad2016: 47). In other words, rather than being invested in fewer companies as would be the case under negative screens, successful engagement results in being invested in the same companies but with a less varied set of activities. Thus, the impact on diversification is arguably comparable between the two strategies. Given our stipulation above that SRI does at times call for a sacrifice in terms of returns, this conclusion does not come as a surprise either. On the contrary, what would have been surprising is the prospect of being able to have our cake (engage corporations to change their behaviour) and eat it (maximize returns), too. To be sure, Kolstad’s critical analysis does not question the virtues of diversification in any way, but it shows that engagement comes at a cost in terms of diversification, too.
(b) Negative screens ineffective in changing corporate behaviour? The basic idea of divestment, recall, is that by withdrawing your money from corporations that engage in morally problematic behaviour, you reduce their capacity to engage in this behaviour. But what if this is simply not true? What if negative screens are unlikely to have any impact whatsoever on corporate behaviour? This would be a troubling observation for the ethically motivated investor, and it would provide strong arguments to favour more effective strategies, one of which might be shareholder engagement. In what follows, I will consider and put in perspective three different versions of the idea that divestment does not have any impact on corporate behaviour.Footnote 15
First, consider what Irvine (Reference Irvine1987: 238) calls the ‘old-stock objection’. When an investor buys shares on secondary markets, they do not in fact give money to the listed corporation in question, but merely buy existing shares from other investors. Thus, when an investor buys ‘old’ stock in this sense, they are not directly enabling the corporation in question to do anything.
But there is a response to this objection (see Irvine Reference Irvine1987: 239). If the divestment of shares today makes it harder for the corporation to raise new capital tomorrow, then this will count as a causal link from divestment to reducing morally objectionable moral behaviour.
There is a catch 22, however, which leads us to the second version of the objection, dubbed the ‘small-purchase objection’ by Irvine (Reference Irvine1987: 239). Since the share purchases of most investors are negligible as a proportion of the overall capitalization of corporations, their purchases will not have any significant influence on the share price. If their purchases do not influence the share price, then a fortiori they do not influence the future capacity to raise capital either, and thus fail to rein in morally objectionable corporate behaviour.
Again, there is a response available. ‘The real question, from a moral point of view, is not whether [the investor’s] one purchase affects the ability of the company to conduct its business, but rather whether his purchase, if imitated by many other investors, would affect the ability of the company to conduct its business’ (Irvine Reference Irvine1987: 239). In other words, as an investor, I should ask myself whether my behaviour, if universally applied by all investors, would impact corporate behaviour. This nimble use of the universalization principle might hold in theory and be sufficient to convince some philosophers, but the problem in practice is precisely that not all investors will behave in this way. This might in part be due to a collective action problem, but in part it results from different convictions about what morality requires.
This leads us to the third, and most troubling version of the objection. As Sandberg (Reference Sandberg, Vandekerckhove, Leys, Alm, Scholtens, Signori and Schäfer2011: 170) points out, financial theory tells us that the equilibrium share price of a corporation is determined by what are called the ‘economic fundamentals’ relevant for the economic situation of the corporation. These fundamentals include demand for its products, but they also include the regulatory environment of the corporation. Sandberg claims that ‘[t]hese fundamentals are not likely to change just because some ethical investors decide to boycott the firm’s shares’ (Sandberg Reference Sandberg, Vandekerckhove, Leys, Alm, Scholtens, Signori and Schäfer2011: 170). If this is true, then if some investors sold the shares of a corporation for ethical reasons leading to a temporarily underpriced stock, other investors with fewer moral qualms would happily scoop them up and pocket the capital gain when the stock price reverts to its equilibrium level. Once again, divestment arguably does not reduce morally objectionable corporate behaviour.
Let me mention two lines of response to this final version of the objection. First, one might wonder how well share prices reflect economic fundamentals in practice. After all, if this were the case, there would be no speculative bubbles on financial markets … But let us put this concern aside here. Second, is it true that the economic fundamentals of corporations are immune to influence from divestment? Contrary to Sandberg, I believe the answer is ‘It depends’. In order to have a real impact, divestment needs to attain a certain threshold.Footnote 16 When it reaches that threshold, several channels of influence on the fundamentals of the corporation open up.
For example, people who are prepared to divest are more likely to boycott the goods of certain corporations as well. Even a slight dent in demand represents an impact on the economic fundamentals. Moreover, consider the potential impact on government regulation. I stated at the beginning of this article that it works on the assumption that governments are not doing enough to rein in morally objectionable corporate behaviour. However, once governments, or political parties more generally, realize that a critical mass of people feel strongly enough about an issue to divest, and that there are thus votes to be gained by taking regulatory action, this again will influence the economic fundamentals of corporations.
These considerations are in line with Miller’s (Reference Miller1991: 34–5) insistence on the fact that the most important aspect of divestment is not necessarily the impact on corporate capacity to raise capital, but the signalling effect it has and the potential knock-on effects on the behaviour of other investors, consumers and governments.Footnote 17
Note an interesting implication of the importance of attaining a certain threshold of divestment. At first sight, you might think that if, at present, my divestment will have no impact because we have not reached the threshold in question, then I have no moral obligation to divest. This seems to follow straightforwardly from ‘ought implies can’. If divestment cannot have the desired impact, then I cannot be under a moral obligation to divest. However, what if my divestment today contributes to enabling further divestment tomorrow to pass the threshold and thus to start having an impact? In this case one might say that I have a dynamic duty (Gilabert Reference Gilabert2012: 241–4) to divest.
In sum, while the effectiveness of negative screens certainly depends on a number of empirical assumptions to fall into place, it would be premature to conclude that divestment is never effective. The relevant question is Compared to what? As long as negative screens are more effective than other available SRI strategies in some contexts, they should be part of the toolbox. We shall come back to this issue.
(c) Pervasiveness of investment and the austere conclusion? Kolers (Reference Kolers2001) has presented the following challenge to negative screens as an SRI strategy. Suppose that I consider the emissions generated by corporations exploiting the Albertan tar sands as imposing harms on present and future generations, and that I therefore wish to divest from such corporations.Footnote 18 Having sold the shares, what do I do with the money? Suppose I buy shares in the Canadian financial sector instead. Canadian banks finance corporations active in the tar sands. So my strategy has failed. Suppose I leave the money in my bank account instead. Again, the bank might lend it out to corporations in the tar sands, undermining my strategy. This is what Kolers (Reference Kolers2001: 438ff.) calls the ‘pervasiveness of investment’.
According to Kolers, the pervasive character of investment makes it very difficult if not impossible to invest in ways that do not involve enabling harm by corporations in one way or another. It leads, he argues, to ‘the Austere Conclusion: morality requires either total divestiture, or adherence to debilitatingly high, iterated ethical screens that would eliminate practically everything from our portfolios’ (Kolers Reference Kolers2001: 439).
Is the Austere Conclusion justified? While I agree that the pervasiveness of investment poses a serious challenge to SRI, I believe that Kolers is too pessimistic for two reasons. First, note that the Austere Conclusion depends on a lack of reliable ethical screens. Kolers himself acknowledges that there might be some financial institutions – he cites community banks (Kolers Reference Kolers2001: 441) – where one’s money only runs a negligible risk of enabling immoral behaviour. The more investment options are available that satisfy not only a prima facie negative screen, but that ensure that the money will not end up enabling bad corporate behaviour through the back door, the less justified the Austere Conclusion appears.
Second, and more importantly, even if we could never get around the problem of the pervasiveness of investment altogether, negative screens can plausibly reduce bad corporate behaviour to some extent.Footnote 19 Surely, reducing morally objectionable corporate behaviour is a good thing, and preferable to doing nothing.
At this point, once again the relevant question becomes the one of the relative effectiveness of negative screens compared to other SRI strategies, notably compared to shareholder engagement. The current engagement paradigm in SRI presumes that engagement is more effective than negative screens. The considerations of this section suggest that this conclusion is premature. I have argued that engagement also implies a loss of diversification, that negative screens can be effective, and that Kolers’ Austere Conclusion is too strong.
To be sure, these arguments do not provide a conclusive case in favour of negative screens either. Instead, they suggest that the issue of relative effectiveness of SRI strategies is more complicated than often assumed. It seems fair to say that it is an open question, dependent on a host of empirical issues as well as on context, whether voice or exit strategies, or a mix thereof, represent the adequate instrument for SRI. There is a certain irony in the fact that diversification at the level of investment strategies be resisted by an investment community in which financial diversification has quasi-religious status.
If the relative effectiveness of different SRI strategies is an open question, then we should revisit the engagement paradigm and adopt a more pluralist approach to SRI that includes all four of the strategies listed at the beginning of section 2, and potentially other ones, too.
3. The ethics profile of investors
The upshot of section 2, within the consequentialist framework assumed there, is an obligation towards participants in collective pension plans to pursue a pluralist SRI strategy. I justified this conclusion by pointing to several difficulties in assessing the relative effectiveness of different SRI strategies, notably of a strategy of voice – through shareholder engagement – versus exit – through negative screens. As long as we are not sure which of the two works better, diversifying our strategy and employing both methods seems the prudent thing to do. While questioning current practice, this is hardly a revolutionary argument.
Does this exhaust the obligations the managers of collective pension plans have towards their members? This section will argue no. To see why not, we need to revisit an assumption I made at the beginning of section 2. There, I argued that the prevailing interpretation of SRI today views it as targeting more socially responsible business practices. This represents a consequentialist position on SRI. I shall now argue that other positions on SRI are equally plausible. If that is correct, if, in other words, there is reasonable disagreementFootnote 20 about what constitutes not just the right thing to do but the theory of the right in the context of investment, then the obligations towards participants in collective pension plans will be much more extensive.
Consider three different motivations that might animate an investor to behave ethically. First, in line with section 2, I might indeed be motivated to invest in ways that reduce corporate wrongdoing. This motivation aligns with a consequentialist theory of the right. Second, I might be motivated by the wish to avoid association with certain kinds of activities, independently of the consequences of dissociating myself from them. This motivation coheres with a deontic or virtue ethics theory of the right.Footnote 21 Third, I might be motivated to invest in ways that support my community, where the latter can be understood in a more narrow (e.g. municipal, regional) or broader (e.g. national) sense. This motivation, which might overlap with either of the two previous ones, coheres with a communitarian theory.Footnote 22
I will call these motivations different ethics profiles of investors:Footnote 23 the consequentialist profile, the clean-hands profile and the communitarian profile.Footnote 24 Note that I do not claim this list to be exhaustive.Footnote 25
Now, two points are important in this context. First, I take it that there is reasonable disagreement about what constitutes the appropriate ethics profile. There is no meta-standpoint from which we could convince an investor with a consequentialist profile to adopt a clean-hands profile, or vice versa. Contrary to the disagreement in section 2 about the most effective SRI strategy, which is a disagreement about empirical questions, the disagreement here is about morality.
Second, and crucially, the recommendations for action that flow from these ethics profiles can sometimes point in very different directions. For example, investors with a clean-hands ethics profile will be much more inclined towards exit-strategies than towards engagement strategies. They will be completely unfazed to learn that their SRI strategy is less effective in influencing corporate behaviour, because this is not something their ethics profile makes them sensitive to.
Note that it is not necessary for my argument to claim that investors neatly fall into the categories of the ethics profiles just sketched. It is plausible to think, as Sandberg and Nilsson’s research (Reference Sandberg and Nilsson2015) suggests, that actual investors often do not consciously conform to one ethics profile or another, and that they in fact have mixed motivations that simultaneously draw on several ethics profiles. It is enough for my argument that some investors object either to a consequentialist or to a clean-hands ethics profile.
These observations lead us to a more radical conclusion. Imposing an SRI strategy onto an investor where the strategy does not match the investor’s ethics profile is deeply problematic. It amounts to ignoring the existence of reasonable disagreement on the ethics of investment, and to imposing a one-size-fits-all ethical position. This is unjust.
To highlight the more radical nature of this conclusion, suppose for a moment that, contrary to my argument in section 2, we could indeed settle the question of what constitutes the most effective SRI strategy to influence corporate behaviour. Suppose shareholder engagement could be shown to be more effective than various exit strategies. This would not change anything with respect to the reasonable disagreement about ethics profiles introduced in this section. In other words, disagreement about the empirics of investment and disagreement about the ethics of investment are independent in this sense. Therefore, what we owe to participants in collective pension plans on these respective grounds is also independent. Even if the most effective SRI strategy could be determined, imposing a one-size-fits-all ethical position and corresponding SRI strategy would still be problematic.
So what do we owe participants in collective pension plans in the face of reasonable disagreement about ethics profiles? I shall now argue that this form of reasonable disagreement calls for options of voice or exit at a higher level. In other words, participants in collective pension plans need to have options available to them to make sure that the investments in their pension plan do not conflict with their ethics profile. Generally speaking, these options can take the form of either a say in determining the investment strategy or the availability of alternative investment strategies that conform better to their ethics profile.
Whereas the question of the relative effectiveness of SRI strategies discussed in section 2 concerns the mix of voice versus exit strategies of investors in their portfolios, the reasonable disagreement about ethics profiles in this section concerns the mix of voice versus exit strategies of individual investors within collective pension plans.Footnote 26
The debate about SRI is often conducted in terms that suggest that the only ethical question is the one at the heart of section 2 of this paper. If my analysis is correct, this is a serious mistake. Not only is there another ethical issue – what to do in the face of reasonable disagreement about ethics profiles – but it turns out that this other question has much more thorough-going implications for the obligations we have towards participants in collective pension plans.
Having argued that reasonable disagreement about ethics profiles requires giving participants in collective pension plans options in terms of either voice or exit, the final section of the paper will now analyse the possible institutional implications of this requirement.
4. Institutional lessons
Before analysing how respect for the different ethics profiles of participants in collective pension plans can be institutionally spelt out by giving participants options of either voice or exit, it is worth pausing over the implications of the argument thus far for a central element of pension plans today: the fiduciary duty of pension plan managers as trustees.
The concept of fiduciary duty is standardly defined as ‘requiring the trustees to demonstrate that the [investment] decision is motivated only by the financial interest of the beneficiary. … However, a ‘benefit’ is not necessarily confined to a financial benefit. If beneficiaries share a moral objection to a particular form of investment, it could be construed as for their benefit if the trust avoided that investment, possibly even at the cost of a lower financial return’ (Richardson Reference Richardson2007: 158–9). The arguments of the previous section corroborate this criticism of the narrow standard definition of the fiduciary duty. If the ethics profile of a participant in a collective pension plan conflicts with the return-maximizing strategy adopted by the trustees making the investment decisions, then we encounter both a moral problem of not respecting the ethics profile in question and a legal problem because the fiduciary duty is violated.Footnote 27
The challenge, as we have already seen in the previous section, lies in the fact that there is reasonable disagreement about ethics profiles.Footnote 28 In the face of this disagreement, it prima facie seems impossible for trustees to discharge their fiduciary duty to all participants simultaneously,Footnote 29 which they are required to do by the duty of impartiality.Footnote 30 However, as this section will argue now, there are institutional tools available to overcome this challenge. I will discuss four institutional arrangements, one which falls into the category of voice, and three that fall into the category of exit. None of them is without weaknesses, but they all take seriously the idea that we owe participants in collective pension plans real choices when it comes to SRI. As we shall see, some of them have already been tested in certain jurisdictions.
(a) Democratic pension funds: A first possibility consists in giving participants an active say in the choice of investment policy. Some observers have suggested that trustees could be obliged to consult participants in the plan (Watt 2006: 437). The statutes of the plan could for instance require that investment policies are voted on in the annual general assembly. Requiring democratic accountability seems particularly important ‘because pension fund managers and their delegates are presently subject to counter-incentives to disregard their fiduciary duty to act in the best interest of the beneficiaries’ (Davis Reference Davis2008: 176). One example for such counter-incentives is the maximization of short-term financial gain and, therefore, of the bonuses of the fund managers, rather than of the long-term financial, and potentially also the non-financial, interests of the beneficiaries.
Democratizing pension funds has pros and cons. On the one hand, while voting on investment strategies may not produce a result that fits everyone’s ethics profile, it does give all members of a pension fund an input into the decision-making process. Given that we accept democratically made decisions as legitimate and respectful of citizens’ convictions, there is no reason to reject a similar argument in the context of SRI.Footnote 31 Especially if one thinks that SRI includes a deliberative dimension, where members of pension funds define the contours of what SRI means to them over time, then there is a strong case for democratizing pension funds.
On the other hand, there are some drawbacks, which unsurprisingly are again similar in nature to weaknesses of democratic arrangements more generally. The ‘views of a minority may be overridden where there is no consensus of opinion, and the relative weight that fiduciaries attach to the various views cannot be readily scrutinized’ (Richardson and Cragg Reference Richardson and Cragg2010: 35). Given my insistence on the importance of matching the investment strategy of the collective plan to the ethics profile of individual participants, the first weakness in particular seems problematic. Giving participants voice in this way might be considered as an insufficiently strong instrument to respect their ethics profile. Providing exit options represents a more promising strategy in this regard.
(b) Individual opt-outs: One might say to participants in collective pension plans: ‘If you don’t like our investment strategy, because it does not correspond to your ethics profile, you can leave.’ If I do not like the fact that my pension plan is invested in fossil fuels, say, I can withdraw my money and invest it in a way that better matches my ethics profile.
The problems with opt-outs of this kind are twofold. First, though the specific regulation varies from country to country, ‘there are three significant factors that provide substantial incentive for employees to participate’ in collective pension plans: ‘the relatively limited public provision of support for retirees, the tax-advantaged status of the plans, and the availability of matching contributions from employers’ (Daskal Reference Daskal2013: 151). In other words, under current arrangements, people who choose to opt out of their collective plan would be at a serious financial disadvantage.Footnote 32 To be sure, this first problem with opt-outs could potentially be overcome. Regulation could force employers to continue contributing to pension plans even if they are managed independently,Footnote 33 and other privileges such as tax advantages could apply, too.
But even then, being left alone to manage your retirement savings rather than being part of a collective plan comes with a second set of drawbacks. Most people lack the financial literacy necessary to assess the risks of different investments and to invest prudently. Moreover, even someone who is financially savvy would no longer benefit from the economies of scale that a collective plan offers. Whether it is savings in terms of time spent researching different investment opportunities, or the commissions and fees charged by fund managers, the small investor is structurally disadvantaged in ways that cannot be remedied in obvious ways. Having to incur this kind of structural disadvantage and the resulting financial cost in order to make one’s investments match one’s ethics profile would be discriminatory against some ethics profiles and therefore unjust.
In sum, it is fair to conclude that individual opt-outs do not represent a real alternative to being part of a collective pension plan.
(c) Different ethics profiles for sub-groups: Rather than letting individuals fend for themselves, the logical next option in the same conceptual neighbourhood is to let participants in collective pension plans self-select into sub-groups that match their ethics profiles. I should immediately add that this formulation is slightly misleading:Footnote 34 If this institutional option were pursued, the different groups could no longer be part of the same plan but would instead turn into independent pension plans. After all, it would be unreasonable to ask one group that has chosen a certain ethics profile and selected a certain portfolio to share risks with another group whose members have made different choices. This cuts both ways: If ethical investment turned out more profitable because of the decline of stranded assets, then it would not be fair to ask the ethical investor group to compensate the others; conversely, if it turned out that ethical investment was less profitable than the profile with no ethical constraints, those who chose the latter also should keep their financial advantage. It is hard to tell which of these scenarios is more likely, but that is not the point here.
I see two potential objections against this proposal. First, even though it does not ask individuals to fend for themselves, it nonetheless results in a reduction in the size of pension funds and, thus, in a loss of economies of scale. This objection motivates the fourth and final institutional arrangement to which we will turn in a minute. One possible response to this objection is that today funds mostly invest in the market via pooled funds administered by specialist asset management companies anyway, and thus do not face an economy of scale issue.Footnote 35
Second, and this is a point that applies to both this proposal and the next, which and how many ethics profiles have to be made available? Where do we draw the line between reasonable disagreement and views that falls into the category of the unreasonable? Imagine a pro-lifer who demands investment possibilities that match his ethics profile. Would pension funds be under an obligation to accommodate this view on the present proposal?Footnote 36
Let me make four points in response.Footnote 37 First, even though there is no precise line between the reasonable and the unreasonable, some views clearly fall into the latter category. We have no obligation to accommodate someone searching for investment opportunities to match his racist views.
Second, one way to decide which ethics profiles should receive tailored investment opportunities is by looking at demand for them. This is how investment opportunities are selected on the market: If a sufficient number of people are prepared to invest in a certain kind of fund, a clean technology fund or a pro-life fund, say, a fund manager will likely put them together. Within existing pension funds, this logic is once again subject to the economies of scale constraint. If only 50 people at my employer have a certain ethics profile, accommodating them would come at a serious price tag. The fourth option below will respond to this worry. One potential objection to leaving the selection of investment opportunities to the market is that, in this case, investing power rather than reasons shapes the conception of social responsibility.Footnote 38 If one thinks that the issues of SRI confronting our societies should ultimately be addressed politically, then this would seem particularly troubling.Footnote 39 What we see here is that market versus democratic arrangements have pros and cons that mirror each other. If one is more concerned about the deliberative aspect of finding an SRI strategy, one is likely to favour democratic pension funds; if one is more concerned to see one’s ethics profile respected, the availability of choices between different SRI options on the market will seem more important.
Third, some SRI activists might object to this proposal that it opens the door to ‘irresponsible’ investment. Since there is demand for them, capital markets have put together so-called vice funds, which exclusively invest in things such as arms, tobacco or fossil fuels. Surely, accommodating this kind of thing would undermine the cause of SRI. Yet, within the limits of reasonable disagreement (see point 1 two paragraphs earlier), this is the bullet I am prepared to bite. According to the position defended in this paper, accommodating a diverse range of ethics profiles is what we owe to participants in collective pension plans.
Fourth and finally, especially in smaller pension funds, the critical mass necessary to make accommodating an ethics profile financially feasible might often not be attained. In this case, in practice, the idea of different ethics profiles for sub-groups would probably still leave a non-negligible number of people and their ethics profiles out in the cold. Yet, there is a remedy to this problem.
(d) Pan-institutional ethics profiles: The potential economies of scale constraint of the previous institutional arrangement could be overcome by pooling people with similar ethics profiles from different institutions. Today, people tend to share the financial risks of their nest-egg with others working at the same institution, where what counts as an institution can be defined at various degrees of generality – contrast my university’s pension fund with TIAA in the USA, where teachers from all over the country participate in a collective pension plan. Alternatively, why not pool people across society according to their ethics profiles instead? Of course, this would imply that I might not share the same retirement plan with my office neighbour, but that is not a problem as such. Australia’s Superannuation Act (2005) could be interpreted in this way, since it requires employers to give employees a choice of the fund they would like their pension to be invested in.Footnote 40
Since the critical mass of investors for any given ethics profile could be easily attained in this way, this proposal represents a promising way of accommodating a maximum number of ethics profiles. It no doubt calls for a fundamental change in the landscape of pension funds, but arguably this is the change that is required in our era of reasonable disagreement about ethics profiles in investment.
One important worry has to be flagged about this proposal. Today, within the constraints imposed by the legislator, pension plans tend to be negotiated between employers and the representatives of employees (at the firm-level or on a larger scale). Shifting this negotiation away from the employer–employee context bears a danger of weakening the bargaining position of employees. If this institutional arrangement were to be pursued, measures would have to be taken to guard against this possibility.
In sum, this section has sketched four ways in which the obligation to respect the ethics profiles of participants in collective pension plans could be discharged. The first did so by giving them voice in the determination of the investment strategy, the three others provide them with alternative ways to invest their money, thus favouring exit strategies. As we have seen, none of these institutional arrangements comes without drawbacks. However, only in one case – individual opt-outs – do these drawbacks seem so significant that it seems to rule out the institutional arrangement in question as an attractive way of respecting the ethics profiles of investors. As for the others, someone who deems deliberation important in identifying SRI strategies and who believes that, ultimately, it represents a collective, political choice, will favour democratic pension plans. Someone who prioritizes questions of personal integrity and, thus, the importance of matching investment strategies with the individual ethics profiles of investors, will find different ethics profiles for subgroups or pan-institutional ethics profiles more attractive. This paper does not take a stance on the choice between these arrangements.
5. Conclusion
According to the OECD (2017: 6–7), in several rich countries the ratio of assets under management in funded pension plans to GDP now exceeds 100%. Deciding on how these assets should be managed is no small fry. This paper has argued that we owe the participants in collective pension plans a say in how their retirement money is invested. Underpinning this obligation is the fact of reasonable disagreement about what constitutes the right thing to do in the context of SRI or, put differently, reasonable disagreement among the ethics profiles of investors.
I have argued that it is a mistake to reduce the debate about SRI to the question of the most effective SRI strategy. While section 1 of the paper has shown that this empirical question is more controversial than often claimed, the issue that turns out to have much more fundamental implications for the management of pension funds is reasonable moral disagreement about ethics profiles. For example, if I have a consequentialist ethics profile, the SRI strategies I am likely to favour are very different from those I will choose if I have a clean-hands ethics profile. Imposing a one-size-fits-all SRI strategy on people with different ethics profiles cannot be justified. The last section of the paper has presented different strategies of giving participants in collective pension plans voice or exit options in order to respect their ethics profiles.
A shift towards more ethical behaviour often starts small. Think of the way recycling has taken hold in many societies in recent decades. However, a necessary condition for gaining momentum is that there are no institutional barriers against behaving ethically. Arguably, this is still the case today in the domain of SRI, because participants in collective pension plans are not given sufficient alternatives to the status quo. This paper has provided some arguments for why and how this needs to change.
Acknowledgements
Previous versions of this paper were presented at the Centre de recherche en éthique (CRÉ) in Montréal in spring 2018 as well as at the annual meeting of the Economic Ethics Network in Paris in June 2018. I thank all participants at these events as well as two anonymous referees for their comments. Thanks also to Alexandre Beaudoin Peña for his research assistance on this article.
Peter Dietsch is a Professor in the Département de Philosophie at Université de Montréal. His research focuses on issues of economic ethics, notably on tax justice, normative dimensions of monetary policy, and on income inequalities. Dietsch is the author of Catching Capital – The Ethics of Tax Competition (Oxford University Press, 2015), co-author of Do Central Banks Serve the People? (Polity Press, 2018) and co-editor of Global Tax Governance – What is Wrong with It and How to Fix It (ECPR Press, 2016). Dietsch is a member of the College of New Scholars, Artists and Scientists of the Royal Society of Canada. URL: https://en.philo.umontreal.ca/repertoire-departement/vue/dietsch-peter/.
