ESG investing with your pension money bends the fiduciary duty owed you

There are three words in the law of trusts that carry more moral weight than most speeches in Congress: solely, loyalty and prudence. They describe what a person owes when he holds someone else's money for someone else's future. They do not say mostly. They do not say in the spirit of the times. They say solely. A good deal of the modern argument about environmental, social and governance investing, ESG for short, turns out to be an argument about whether that word still means what it says.

I will argue that it does, that investing retirement money to advance a cause, however worthy, is a breach of the duty owed to the person whose money it is, and that the cure is not a ban on any idea but a firm rule about whose interest comes first. I will also give the best case on the other side its full due, because it is not a silly one.

The terms

ESG stands for environmental, social and governance factors. As an investing label it covers a range of practices, from simply asking whether a company's climate exposure or legal risk could affect its earnings, to screening out whole industries, to voting shares in order to change corporate behavior. A fiduciary is a person who holds a duty of trust and loyalty to another, like a trustee, a pension board member or a plan manager. The duty of loyalty requires the fiduciary to act in the beneficiary's interest and not his own or anyone else's. The duty of prudence requires him to act with the care a knowledgeable person would use. Pecuniary means related to money: financial return and risk.

My thesis has two parts. First, a retirement fiduciary must make decisions on pecuniary grounds, which means that a factor like climate or governance is legitimate if, and only if, it bears on risk and return. Second, when a manager uses retirement assets to pursue a goal that is not about risk and return, he is spending the savers' money on his own causes. Here are the premises.

1. Retirement savings belong to the savers, who depend on them for their old age.
2. The law places a duty of loyalty on those who manage them.
3. That duty is a duty to the savers' financial interest, which is the sole purpose of the arrangement.
4. Using the savings to advance a social or political aim, where the aim is not a means to better returns, is therefore a breach, even when the aim is popular.

What ERISA says

The Employee Retirement Income Security Act of 1974, known as ERISA, is the federal law that governs private-sector pension and benefit plans. Its fiduciary provisions are blunt. A fiduciary must discharge his duties with respect to a plan solely in the interest of the participants and beneficiaries and for the exclusive purpose of providing benefits to them and defraying reasonable expenses of administering the plan. He must act with the care, skill, prudence and diligence of a prudent person familiar with such matters, and he must diversify the plan's investments unless it is clearly prudent not to.

Read that slowly. "Solely." "Exclusive purpose." Those are not words that leave room for a second mission. Congress passed ERISA after a series of failures in which workers had lost promised pensions, and the point of the statute was to put the interests of those workers ahead of every other consideration, including the interests of the employer, the union, the plan's own managers and, I would add, the fashions of the day.

ERISA applies to private plans. Government employees' pensions, such as those of state and local workers, are governed by state law and by their own plans' rules, and the details vary. But the principle is the same one and is found in the law of trusts generally: a trustee must be loyal to the beneficiaries.

The Labor Department's back-and-forth

The Department of Labor writes the regulations under ERISA, and on ESG its position has swung with each administration. In 2020, near the end of the first Trump administration, the Department issued a rule on the selection of plan investments that required fiduciaries to base their decisions on pecuniary factors and not to subordinate the interests of participants to other goals. The rule did not ban the consideration of environmental or social factors. It said that if they bear on financial risk and return, they count; if they do not, they do not.

In 2022, the Biden administration issued a new rule that loosened this. It made clear that a fiduciary may consider climate change and other ESG factors when they are relevant to a risk-and-return analysis, and it loosened the 2020 rule's limits on using non-financial factors as a tiebreaker. Congress passed a resolution to overturn the rule in 2023, and President Biden vetoed it, the first veto of his presidency. Rules of this kind can change again with any administration, and the reader who wants the operative text should consult the Department of Labor and the Federal Register directly, because I do not want to describe a rule that no longer applies.

What the back-and-forth shows is that the line between "relevant to returns" and "advancing a goal" is a matter of judgment, and judgment is where the trouble comes in. A rule that says "consider whatever is relevant to risk and return" is correct, and also elastic. A manager who wants to favor a cause can always find a risk story to tell about it.

What the states did

States have gone further. Texas, in 2021, passed a law that bars state governmental entities from large contracts with companies that boycott energy companies, and directed the comptroller to keep a list of financial companies that do so, which state entities are restricted from dealing with. Other states have passed their own measures, some limiting ESG considerations in the management of public funds and some, in the opposite direction, requiring them. Be careful about the term "ban" here: these laws are best understood as an insistence that public fiduciaries use pecuniary criteria, or that the state not do business with firms that boycott a lawful industry. A reader should check his own state's statute on its legislature's website, because the details differ from one state to the next.

Why would a state want such a rule? For the same reason ERISA does. A teacher in Texas who has contributed to a pension plan for thirty years has a claim on its returns. If the board decides to divest from an industry that is a major employer in her state, on the ground that the industry is unfashionable, and the divestment costs the fund returns, the loss falls on her and on the taxpayer who must make up any shortfall. The board member who made the decision does not bear it.

The voting power

There is another, quieter dimension. The largest index-fund managers, which hold the shares of most large public companies on behalf of millions of savers, also vote those shares at corporate meetings. A handful of firms, among them BlackRock, Vanguard and State Street, hold large stakes in most big companies and cast the votes that come with them. This is a form of concentrated power that the Founders would have recognized, and probably distrusted, because those who exercise it are not the owners of the money.

The saver who has money in an index fund through his plan has delegated the vote. The question is what the manager does with it. If he votes to maximize the long-run value of the company, he is a faithful agent. If he votes on a social agenda that has no clear tie to value, he is using the saver's property to push a cause the saver was never asked about. This is the same agency problem described in Milton Friedman's 1970 essay against stakeholder capitalism still wins, one level up: now it is not the corporate executive but the asset manager who spends someone else's money on his own preferences.

What it costs: the arithmetic

Let me show why a small difference in returns matters, with a hypothetical worked example. I am labelling it a hypothetical: these are illustrative numbers, not a forecast, and this is not advice for your own account.

Suppose a worker's retirement balance is $100,000 and it is left to grow for 30 years. If it earns 6 percent a year, the balance grows to 100,000 × 1.06^30, which is about $574,000. If an investment policy costs one percentage point in returns, so that the balance earns 5 percent, the result is 100,000 × 1.05^30, about $432,000. The difference is roughly $142,000, or about one quarter of the larger figure. A single percentage point, compounded over a working lifetime, takes a quarter of the retirement.

Now, I do not claim that ESG policies cost exactly one point. Some research finds a drag and some finds none, and I would urge readers not to accept anyone's single study as settled. My point is what the arithmetic says about the burden of proof. When the stakes are compounded over decades, the person who proposes to depart from a pure return-based standard ought to show that the departure is free. He should not be allowed to simply assert it.

The steelman

The best case for ESG investing says this. First, many ESG factors really are financial. A company with a bad safety record, a history of legal violations, poor governance or large exposure to climate-related regulation faces real risks, and a prudent investor would look at them. Second, long-horizon investors like pension funds have decades to wait, and risks that look remote to a trader can become material over a working life. Third, the fiduciary rule has always allowed the consideration of any factor that affects returns, so banning the use of ESG factors would itself be an imposition on the manager's discretion. And fourth, when conservatives say that fiduciaries should ignore politics, they ignore the fact that boycotting ESG is also a political act, and that state laws punishing firms for their views are themselves political.

I accept most of this. If a factor is financially material, a prudent fiduciary should weigh it, whatever label it travels under. A fund that ignored a company's legal liabilities because they came under the heading of "governance" would be negligent. And the state laws deserve scrutiny, because a rule that says "no business with firms that disagree with us" can be a political instrument. A principled conservative should apply the same test to both sides.

But here is my reply in four numbered points.

1. The test is the same, and ESG advocates often fail it. The question is whether the factor bears on risk and return, shown by evidence. Many ESG initiatives are justified by their social benefits, and the managers' own marketing says so. If the stated purpose is to change the world, then the purpose is not the saver's return.
2. "Long-term risk" is an elastic phrase. Any cause can be described as a long-term risk to something. The question is whether the claim is specific, measurable and priced by the market. A fiduciary who adopts a broad theory of long-term risk that happens to coincide with his preferences has not escaped the agency problem. He has given it a better disguise.
3. The saver cannot easily leave. A worker in a 401(k) plan or a public pension has limited ability to take the money elsewhere, and a participant in a public plan has none. That makes the duty of loyalty more important, not less, since the usual market discipline of exit is weak.
4. Neutrality is a virtue in a fiduciary. The great value of ERISA's "solely" is that it gives the saver a clear standard that does not depend on who holds office. A rule that says "pecuniary factors only" protects the liberal teacher and the conservative plumber equally. A rule that says "social goals count" protects them only until the goals change.

The Founders and the trustee

The logic here is older than ERISA. The law of trusts descends from English equity, which the American states inherited, and it rests on the idea that a person who holds power over another's property must exercise it for the owner and not for himself. That is the same principle behind the Declaration's complaint against a government that governs without consent, and behind the Constitution's treatment of public power as a trust. A representative holds power for the people; a trustee holds property for the beneficiary. In neither case may the holder redefine the purpose.

I would add one more thing about the political stakes. When pension money is used to push social goals, the decisions are made by a small number of people, none of whom the saver chose, away from the visibility of elections. That is a trend worth watching, in the same way that the growth of unelected administrative power is worth watching, which is a theme the site returns to at the Constitution page and elsewhere. For a related warning about how a trusted institution can be bent to purposes other than its stated one, see Social Security is not your property, and Flemming v. Nestor said so.

Consider also who does the paying when a mistake is made. When a corporation's management wastes shareholder money, a rival can buy it and fix it, as profit is a signal, not a sin, and what a loss tells a company explains. When a pension board wastes savers' money, there is no rival and no exit. The loss is simply absorbed by retirees or by taxpayers, which is one more reason why the loyalty rule has to be strict.

The case in brief

1. ERISA, enacted in 1974, requires a plan fiduciary to act solely in the interest of participants and beneficiaries and for the exclusive purpose of providing benefits.
2. The Labor Department's rules of 2020 and 2022 moved the line on ESG in opposite directions, which shows that "relevant to risk and return" is a standard that can be stretched.
3. State laws such as Texas's 2021 energy-boycott law reflect the same insistence on pecuniary criteria in public funds, though they too deserve scrutiny for political overreach.
4. A one-point drag over 30 years takes about a quarter of a $100,000 balance in the hypothetical above, so the burden of proof rests on whoever proposes the departure.
5. If an ESG factor affects risk and return, a prudent fiduciary should use it; if it advances a cause, it does not belong in someone else's retirement money.

What you can do: ask your plan administrator or pension board for its written investment policy statement and its proxy voting guidelines, and read how it defines the participants' interest.