In 1830 a Massachusetts court refused to list the investments a trustee could safely make, and the law spent the next century assembling the list anyway

Anyone who manages money for somebody else works under a standard of prudence. The question worth asking is not whether a manager must be prudent but what the unit of judgment is: a single holding, or the whole collection. The two give opposite answers about the same decision. A diversified portfolio contains losing positions by design, because the point of spreading money across assets that do not move together is that some fall when others rise. Judge the holdings one at a time and you condemn the practice you meant to encourage.

The rule everyone cites begins with a loss. John McLean's will of 1823 left fifty thousand dollars in trust for his wife, the remainder to Harvard College and the Massachusetts General Hospital, and gave his trustees discretion to invest in safe and productive stock, whether in the public funds, bank shares or other stock. The trustees put a large part of it into manufacturing and insurance company shares. By the time the accounts were settled the fund had fallen to twenty-nine thousand four hundred and fifty dollars, though substantial dividends had been paid to the widow.² The two institutions waiting at the end of the trust objected.

The Massachusetts court held the trustee not liable. Justice Samuel Putnam's opinion in Harvard College v. Amory told trustees to observe how men of prudence, discretion and intelligence manage their own affairs, with regard both to the income to be derived and to the safety of the capital.¹ Two things in it are worth separating, because the second is almost always dropped. It states a standard of conduct, measured against what careful people do rather than against outcomes. And it declines to confine trustees to an approved list of investments, as English practice did.¹² The famous sentence is the first half. The refusal is the half that did the work.

The refusal did not survive. In King v. Talbot, decided in New York in 1869, executors had put children's legacies of fifteen thousand dollars into corporate stocks and bonds, and the court held those investments improper.³ The opinion is not itself a retreat into rule-making: it describes the English confinement of trustees to government debt and loans secured on real estate, rejects it as a purely arbitrary rule with no application in this country, and states the duty as one of fidelity, diligence and prudence.³ Yet the case is remembered, correctly, as restrictive, read by commentators as narrowing the acceptable investments and tilting trustees towards government bonds and mortgages rather than common stock.³ By the twentieth century the great majority of American trust funds were governed by statutory legal lists.⁴ A rule born as a refusal to specify content had become content.

How that happens is a feature of adjudication rather than an accident of doctrine. A case arrives as one investment that lost money, so a standard of conduct gets applied backwards to a particular holding, and the holdings that generate litigation are the ones that fell. Precedent then generalises from the facts, and a standard calling for judgment in all the circumstances accumulates subrules prescribing which investments a trustee may make.⁴ Nobody chose the unit of review. The procedure chose it.

The correction changed the unit and left the standard almost intact. The Restatement of the law of trusts adopted a portfolio standard, and the Uniform Prudent Investor Act of 1994, drafted on it, provides that a trustee's decisions about individual assets are evaluated in the context of the whole portfolio and as part of an overall strategy suited to the trust.⁵ Both disavow the categorical bans the older rule had produced: prudence becomes a matter of how risk is managed rather than of results, no investment is imprudent in itself, and diversification becomes the default duty unless it is prudent not to diversify.⁵ What changed was the level of aggregation at which a decision is reviewed, and that was the whole reform.

The choice of unit is a choice about which error to prefer, and neither option is neutral. Asset-level review catches the trustee who gambles but punishes the trustee who diversifies, because a diversified portfolio always offers a loser to complain about. Portfolio-level review protects the diversifier and makes almost any single bad holding defensible as one leg of a strategy. A legal system has to decide which failure it would rather have.

The question is live in the largest pool of fiduciary money there is. The American statute governing retirement plans requires a fiduciary to act with the care, skill, prudence and diligence a prudent person familiar with such matters would use under the circumstances then prevailing, and the associated duty is to diversify so as to minimise the risk of large losses unless it is clearly prudent not to.⁶⁷ In Hughes v. Northwestern University, decided on 24 January 2022, the Supreme Court held unanimously that an appeals court had erred by relying on the breadth of a plan's menu, and on participants' freedom to choose within it, to defeat a claim that some options were imprudent. A fiduciary has a continuing duty to monitor the plan's investments and remove imprudent ones.⁸

That is asset-level review, reinstated for retirement menus long after trust law moved away from it, and on the analysis above it is right. The reason is who holds the risk. In a trust the trustee holds the portfolio, so reviewing a decision in its context matches what the beneficiary owns. In a defined-contribution plan the fiduciary assembles a menu and each participant holds whichever funds they picked. A menu is not a portfolio, and the badness of one option is not cured by the goodness of another a saver did not buy. The unit of review should follow whoever bears the undiversified exposure. Two apparently contradictory lines of authority then turn out to be answering different questions.

The strongest objection is that this makes a story about procedure out of what was really a story about markets. The nineteenth-century legal lists had an obvious rationale: equity markets were thin, corporate accounts unreliable, information scarce, and a rule keeping trustees in government bonds and mortgages may have been sound advice rather than a drafting failure. If the lists were substantively right, their content was a judgment about securities markets, carried by a rule whose stated authority was a case that had declined to make such judgments. Where a standard's language and its operative content come apart, the content is doing the work.

Two things follow for a reader meeting the word prudent in a document about managed money. The standard tells you nothing until you know its unit: prudent in what, judged against what, at what level of aggregation. And here is a prediction with its own test. Litigation over retirement menus after Hughes should drift back towards informal lists, meaning fee ranges and fund categories understood to be safe to offer, for the same reason the lists appeared before. It will show up in settlements and in advice to plan committees rather than in any judgment announcing a list.

Notes

1Harvard College v. Amory, 26 Mass. (9 Pick.) 446 (1830), for Justice Samuel Putnam's direction that trustees observe how men of prudence, discretion and intelligence manage their own affairs with regard both to income and to the safety of the capital, for the court's rejection of the English approach of confining trustees to a list, and for the trustee being held not liable. The outbound network this paper was drafted on refused every request, so the opinion could not be opened; the quoted formulation and the facts of the case were returned in search results, one of which gives the prudence passage as appearing at pages 460-461 of the report. Encyclopaedia entry on the case, a scholarly account of the rule's later distortion and a local history record of the decision

2For the facts of the McLean trust, including the bequest of fifty thousand dollars for the testator's wife with the remainder to Harvard College and the Massachusetts General Hospital, the trustees' discretion to invest in safe and productive stock whether in the public funds, bank shares or other stock, the holdings in manufacturing and insurance company shares, and the decline of the fund to twenty-nine thousand four hundred and fifty dollars despite substantial dividends paid to the widow, see the encyclopaedia entry at note 1 and the account of the prudent man rule in the Office of Thrift Supervision's trust handbook, which also records that the court rejected the legal list in favour of the prudent man rule. Trust handbook and a reference entry on the rule

3King v. Talbot, 40 N.Y. 76 (1869), for the executors' investment of fifteen thousand dollars of legacies in corporate stocks and bonds, for the holding that those investments were improper, for the opinion's own rejection of the English rule confining trustees to government debt and real-estate-secured loans as a purely arbitrary rule with no application in this country, and for its statement of the trustee's duty as fidelity, diligence and prudence. The characterisation of the case as narrowing permissible investments and tilting trustees towards the safety of capital comes from commentary rather than from the opinion, and the commentary is not unanimous: one account says the case overturned the prudent man rule and reverted to the stricter common-law standard, another that it adopted the rule while restricting the options available under it. Both readings are reported rather than reconciled. Full text of the opinion at the New York courts reporter archive, a case brief and commentary on the case's restrictive effect

4For the accumulation of specific subrules prescribing the types and characteristics of permissible trustee investments, and for the statement that the great majority of trust funds came to be subject to legal list statutes, see the account of the rule's history in the New York University National Center on Philanthropy and the Law paper cited at note 1, and the summary of the pre-1994 position in the materials on the Uniform Prudent Investor Act at note 5. Scholarly account

5Uniform Prudent Investor Act (1994), and the Restatement (Third) of Trusts on which it was drafted, for the evaluation of a trustee's decisions about individual assets in the context of the trust portfolio as a whole and as part of an overall investment strategy having risk and return objectives reasonably suited to the trust, for the disavowal of categorical restrictions on types of investment, for prudence being a matter of risk management rather than of results so that no investment is imprudent in isolation, and for diversification as the default duty unless it is prudent not to diversify. The uniform text itself was not reachable, so the Act is cited through two state enactments and an explanatory summary. Sources differ on the Restatement's date, one giving its prudent investor provision as 1992 and another dating the adoption to 1990; no date for it is asserted in the text above. Texas enactment of the Act, the Texas legislature's analysis of the bill, an explanatory summary of the Act and an account of the California enactment

6Employee Retirement Income Security Act, section 404(a)(1)(B), and the Department of Labor's investment duties regulation at 29 C.F.R. 2550.404a-1, for the requirement that a fiduciary act with the care, skill, prudence and diligence under the circumstances then prevailing that a prudent person acting in a like capacity and familiar with such matters would use. The regulation's own portfolio-context language could not be verified in a reachable source, so it is neither quoted nor relied on here. Text of the regulation

7Department of Labor interpretive bulletin on investment duties, for the duty to diversify the investments of a plan so as to minimise the risk of large losses unless under the circumstances it is clearly prudent not to do so. Code of Federal Regulations text

8Hughes v. Northwestern University, decided 24 January 2022, for the unanimous holding that the Court of Appeals erred in relying on the participants' ultimate choice over their investments to excuse allegedly imprudent decisions, for the continuing duty to monitor investments and remove imprudent ones, and for the holding that offering some prudent options does not categorically bar a claim that other options are imprudent. The opinion was written by Justice Sotomayor, Justice Barrett taking no part, and the case was reversed and remanded rather than decided on the merits. Encyclopaedia entry on the case, a law firm account of the decision and a second account