This summer, the Department of Labor proposed a rule that seeks to reaffirm the principle that pension plans’ main focus should be the financial well-being of retirees, and asked for comments from interested parties. As Labor Secretary Eugene Scalia put it in the Wall Street Journal: “The department’s proposed rule reminds plan providers that it is unlawful to sacrifice returns, or accept additional risk, through investments intended to promote a social or political end.”
The rule shifts the burden of proof onto pension trustees to show that “socially responsible” Environmental, Social, Governance (ESG) factors actually benefit retirees before pension assets can be put into them.
Here is my (lightly edited) comment letter submitted to Secretary Scalia:
Dear Mr. Secretary:
My name is Jerry Bowyer and I am the founder and President of Bowyer Research. BR is what many would call a “quant shop,” which specializes in data analytics. For the past decade we have consulted extensively specifically on the use of social factors in country-level allocation decisions, and governance factors when it comes to company-level allocation decisions, both in equity and debt instruments. Such analysis is a core function for our business and this analysis has been used to create rules-based ETFs and has also been used in asset allocation decision-making processes. Our interactions with the data, with data providers, and with the ESG industry have made me an ESG skeptic, and my views are therefore supportive of the proposed rule, which would protect retirees and the integrity of America’s pension system. I do wonder, though, if the proposal goes far enough in reigning in the ESG industry’s tendency to put what appears to be a politically driven agenda ahead of the interests of retirees whose plans are covered under ERISA.
Comments on Proposed Department of Labor Rule Change Regarding ESG and ERISA
ESG Reform Is a Moral Imperative
Pension Trustees are to have one objective: funding retirement plans.
This is a moral principle first, from which a legal principle has been derived. It predates current law and undergirds it, and the idea of a steward who holds and manages assets in trust for another is well-established in our most ancient moral sources, which speak extensively on the matters of stewardship, trust, and the dangers and pervasiveness of steward/trustees putting self-interest ahead of the interest of the beneficiaries.
“In this case, moreover, it is required of stewards that one be found trustworthy.” 1 Cor. 4:2
Pension trustees are stewards of other people’s capital.
The temptations of being a steward include a lack of diligence (Matthew 25:14-30). But there are other ways of putting one’s own interests/concerns ahead of the interests of the owners. This can be a matter of putting the trustees’ financial interest ahead of owners (Luke 16:1-13), or it can be a matter of putting reputational interest ahead of those who are morally entitled to financial benefits in retirement (Matthew 15:3-9), because leaders, more so than others, are highly motivated to be seen in a positive light. (Matthew 23:5-7).
ESG investing presents moral challenges consistent with each of these examples. Advocates of ESG as well as its companion, “stakeholder capitalism,” receive a great deal of positive press, because it is a highly popular trend and is popular especially with young people in polls (though it is much more highly supported in polls than it is in practice, despite many vehicles and wide publicity). This added social status is then monetized in the form of additional fees — which detract from after-fee performance. Thus ESG investing creates the temptation “to be seen of men” doing good, to be “called great in the marketplace,” and to be “given dinners in their honor” for offerings made to a revered cause at the expense of the dependent elderly.
For this reason, ESG investing represents a high-risk environment for “agency risk,” specifically in the form of virtue-signaling with other people’s money for an additional fee.
Some Logical Problems with ESG as a Return Enhancer
Before getting into the actual observed realities of ESG investing, which are often highly problematic, there are a few reasons to be suspicious that it puts the interests of retirees ahead of the interests of trustees prima facie.
Any constraint which is placed on portfolio construction comes at the expense of other factors.
If a portfolio follows, for example, a valuation discipline, and an ESG screen is applied to the available universe of stocks, the ESG screen diminishes the valuation rule. In other words, if a company is the most attractively valued of two alternatives, but it fails the ESG test and is excluded, then valuation discipline is eroded.
Whatever the method of using ESG, whether a screen-out, or a screen-in or a partial version of either, it will dilute rules which are very well-established in the literature, in the data and in practice. The only way that ESG would not dilute established return-seeking rules would be for the ESG filters to be so broad as to not filter out any constituents of the portfolio as constructed according to rules, which are aimed solely at maximizing risk-adjusted return. In other words, any form of ESG which is more than just for show, meaning that it makes no or few adjustments to the portfolio, is overriding more well-established rules which are designed to maximize the financial benefits for the workers. ESG rules either do little-to-nothing, serving as mere virtue-signaling marketing material, or, if they are materially altering the portfolio they are overriding systems of return maximization. These systems are backed up by decades of academic research, widespread professional practice and basic human reasoning, such as those that underly value discipline (buy low, sell high), Efficient Market Hypothesis (the hypothesis that markets involving many people will tend to incorporate more information than any single expert), or various behavioral schools which look at certain irrational tendencies in human nature and seek to exploit those for return.
ESG’s origins are in political institutions, not market institutions, suggesting that it is, by nature, a political movement with political/social goals, not a genuine financial return-maximization strategy.
In his recent Wall Street Journal op-ed submission, Secretary Scalia mentioned that “More than 3,100 institutional investors and businesses that provide investor services have signed the United Nations Principles for Responsible Investment.” As the name implies, this document finds its origins in the United Nations, coming out of a summit convened in 2005 by then Secretary General of the United Nations, Kofi Annan (whose term in office was marred by serious financial scandal and, ironically, non-sustainable fiscal management). Its origins are in multilateral institutions (such as the UN), political activism and academic social-science departments for the most part. It was, from the beginning, a movement to promote political and social change, not higher market performance. Suggestions that this fashionable social change would also serendipitously coincide with better performance is an add-on to the political goals.
ESG investing faces a headwind that places those whose assets are managed under it at an automatic disadvantage: it adds costs to the transaction. Whichever side of the “make/buy” decision ESG managers and trustees come down on, costs are incurred. Either stewards of retirement assets gather data themselves, a very complex and labor-intensive project, or they pay license fees to some other provider. Either way, substantial costs are incurred. And I can say from personal experience as a member of several investment committees and also as the head of a data analytics firm which spends a great deal of time curating and analyzing data, including ESG data, that even when the data is provided by an outside party, there is a tremendous amount of responsibility involved in deciding between the shockingly inconsistent standards between different ESG rating services, and making decisions about which of the various ESG factors to use, and in which combinations and with which weightings. Then there’s the issue of data accuracy and integrity.
[Within the limitations of confidentiality, I can say from personal experience that the ESG data sources we’ve dealt with have been marred by numerous gaps and inaccuracies in the data as well as extreme time limitations to the historical data, making the data insufficient for an adequate study of statistical relationships and meaningful back-tests.]
Unless ESG is just a badge to be worn, or a bumper sticker to be displayed, (in which case data accuracy is largely beside the point), then any business which takes data integrity seriously (rather than treating ESG as a merit badge or a marketing gimmick) must expend substantial labor at the data analytics level and also at the senior management level where divisive social issues and clashing value systems are hashed out. In the end, who pays for all that? The retirees. One way or another, it comes out of the assets: the retirees’ assets.
So, even if the ESG factors tend to be additive, they must add more additional return (more alpha) than the costs incurred to choose a provider, procure the data, choose the methodology to apply the data, and ensure data integrity and continuity. And even if that bar is cleared, a client-centered approach must account for the asymmetry between the certainty of cost and the uncertainty of benefit. In other words, ESG might enhance returns, but it will increase costs. Some adjustment must be made to balance the certainty of added costs to the plan against the, at best, probable benefit of the use of ESG.
All of that, of course, assumes that ESG is additive at all.
Alex Edmans, Professor of Finance at London Business School and Academic Director of the Centre for Corporate Governance, in his well-documented new book Grow the Pie, shows that socially responsible funds tend to underperform the market, and refutes the idea that ESG in general is additive. In fact, he shows that some practices which are often flagged as problematic by much of the ESG community can, in fact, be additive to shareholder returns.
Rule is Asking for Very Little — Perhaps Too Little
Perhaps my concerns are overblown. Maybe ESG really does work, not just with the advantage of testing hundreds of E and S and G metrics with backward-looking advantage and cherry-picking a few which happen to work by random coincidence; but rather, maybe the ESG proposition as a whole is sound. Maybe the whole ES&G proposition is sound instead of E and S riding in the cart pulled by G, but not weighing the cart down so much as to eliminate all the G advantage.
And maybe the ESG advantage is so great that it generates more added return than the hard costs it incurs, and does it so reliably that the retiree is still ahead of the game once one adjusts for the dead certainty of the costs against the uncertainty of the benefits. Maybe something which started as politics, not finance, and has made an incursion into finance has actually discovered some new key to improving performance and a key which is so much better than the other approaches out there (value investing, momentum, index investing at absolute minimal fee incursion) that it should be used to override or dilute those other approaches.
Perhaps . . . perhaps . . . perhaps. But even if all this is true, all the new rule really amounts to is to ask ESG trustees to prove it. Well, not even to prove it, but to at least back up these claims with solid analysis. The ESG industry has made a lot of claims. My analysis leads me to believe that they are overstated at best and are too often the reverse of the truth. But if the ESG industry is right, then requiring it show the work is really not too much for America’s retirees to ask for, is it?
Suggestions for Improvement
If I have a criticism of the rule it is that it doesn’t seem to go far enough. It seems like more drastic reform is needed.
For example, the ESG industry should show not only that its factors are additive, but that they are additive net of costs, not just the hard costs of data, but the additional costs of internal review, any added transaction costs due to possible turnover, and any opportunity costs compared to operating the funds according to the rules excluding the ESG rules. In other words, a momentum fund plus ESG should do at least as well as the same momentum methodology without the ESG component, over and above a full accounting of the added costs of incorporating ESG.
Furthermore, performance data should include attribution analysis. In other words, ESG should not get credit for excess performance to the degree that it happens to coincide with sector, size or style characteristics. If the large tech growth sector happens to do better during a time period studied and also happens to be more progressive with social policies, that should be controlled for to see whether it is the sector or the ESG characteristics that is providing the advantage. Attribution analysis is not fool-proof; it does involve judgment calls, but it’s hardly unusual. Such a standard tool should make ESG show that it is standing on its own, not riding on top of a big tech bull market.
The DOL should consider going further than requiring trustees to show only “risk-adjusted return” advantage, it should require the ESG industry, in addition, to show both the “risk” (typically volatility is used here) and the excess return, i.e. alpha. If any alleged ESG advantage in risk-adjusted return is due to decreased volatility and not increased returns, long-term investors (who tend to be in a better position to ride out the short-term vicissitudes of volatility) should know that. If ESG ends up giving them a less bumpy ride but at the cost of a lower return, that should be disclosed.
Finally, a closer look should be taken at proxy voting reform. This would make it easier for retirees to get the highest return that they can and still speak out on social issues. Companies should be chosen for probability of return. Then the owners, whether the corporate pension plan or individual defined-contribution participants, can speak out on issues of concern about the corporations’ social policies as owners, not as forced investment boycotters.
Having spent the past two years investigating proxy advisory services and seeing the inner workings of the system, I have seen that it suffers from severe transparency problems and also shows no evidence of political even-handedness. Pension reform and proxy reform are part of the same agenda, putting retirees back in charge of their money.