--- slug: fiduciary-duty type: concept summary: "The legal and governance obligation owed by anyone managing assets, authority, or decisions for someone else's benefit, and the duty map that decides which standard governs a given pool before impact-first capital can be deployed." created: 2026-06-25 updated: 2026-06-25 related: investment-policy-statement: relation: informs note: The IPS is where duty-sensitive allocation rules, concession budgets, and claim boundaries become written policy the committee can be held to. decision-rights-charter: relation: informs note: The charter names who has authority to make or ratify a fiduciary decision, which is half of answering whose duty is at stake. investment-committee: relation: bounds note: Committee members owe duties of care and loyalty when they approve allocations; the duty stack is the standard their process must satisfy. private-trust-company: relation: bounds note: A PTC's directors owe trust-law duties of prudence, loyalty, and impartiality to current and remainder beneficiaries. dynasty-trust: relation: bounds note: A long-duration trust sharpens the impartiality problem because the trustee owes duties to beneficiaries who are not yet born. mission-related-investment: relation: bounds note: MRIs from a foundation endowment remain investments subject to prudent-investor standards, which is the duty boundary the entry maps. program-related-investment: relation: bounds note: A PRI is governed by the section 4944 jeopardizing-investment rules, a different duty boundary from an ordinary endowment investment. family-office-exclusion: relation: complements note: The office's regulatory perimeter shapes who owes which duties; the exclusion answers the registration question, not the fiduciary one. impact-washing: relation: detects note: Impact-washing starts when the public claim outruns what the fiduciary file can support; the duty record is the evidence test. family-mission-statement: relation: informed-by note: A values statement can evidence beneficiary consensus in a prudent-process record, but it does not replace the legal duty analysis. five-capitals: relation: informed-by note: The Five Capitals frame the family's purpose; the duty stack decides how much of that purpose a given pool is permitted to pursue. --- # Fiduciary Duty > **Concept** > > Vocabulary that names a phenomenon. *The legal and governance obligation owed by anyone managing assets, authority, or decisions for someone else's benefit, and the duty map that decides which standard governs a given pool before impact-first capital can be deployed.* *Also known as: fiduciary obligation, fiduciary standard, trustee duty, the duty stack.* ## What It Is Fiduciary duty is the obligation a person or body takes on when it manages assets, authority, or decisions for someone else's benefit. In a family-office setting the fiduciaries are not abstract. They are the trustees of the family trusts, the directors of a [Private Trust Company](private-trust-company.md), the directors of the private foundation, the members of the [Investment Committee](investment-committee.md), the delegated [Outsourced Chief Investment Officer](outsourced-cio.md), and sometimes the officers of the office itself. The recurring vocabulary is small. The duty of **prudence** (also called the duty of care) requires the fiduciary to manage with the skill, caution, and diligence a prudent person would use. The duty of **loyalty** requires the fiduciary to act in the beneficiary's interest, not the fiduciary's own and, in the strictest formulations, not anyone else's. The duty of **impartiality** requires a trustee to balance the interests of current beneficiaries against those of remainder beneficiaries, including beneficiaries not yet born. The duty of **obedience** requires a charity's directors to keep the institution's assets dedicated to its stated charitable purpose. The standards are not identical across settings, and the difference is the whole point. A private trust is governed by trust law, the Uniform Prudent Investor Act in most U.S. states, and the trust instrument itself. A charitable endowment is governed in most states by the Uniform Prudent Management of Institutional Funds Act (UPMIFA), which lets the board weigh the charitable purpose alongside return. A private foundation's investments run under IRC section 4944, which penalizes jeopardizing investments but, under [a Program-Related Investment](program-related-investment.md), exempts capital deployed primarily for charitable purpose. An ERISA pension applies the strictest "sole interest" loyalty rule. A retail-advised account applies the Advisers Act fiduciary standard. The practical reading: there isn't one fiduciary duty. There is a duty *stack*, and which layer governs depends on which pool the capital sits in, what document created the pool, and which statute the jurisdiction applies to that pool. ## Why It Matters Impact-first capital often fails in the conversation before it ever reaches the investment memo, because the parties use "fiduciary duty" as a veto word rather than as an analysis. A trustee assumes that any concessionary or values-aligned investment violates the duty of loyalty, and refuses the proposal on principle. A foundation board assumes that its mission lets it ignore ordinary business care, and approves a structure it cannot defend if the IRS examines it. A family council asks for a climate, racial-equity, or place-based allocation without naming which pool is governed by private-trust law, by UPMIFA, by the section 4944 jeopardizing-investment rules, or by an internal investment policy. Each party is reasoning from a different layer of the stack without knowing it, and the conversation stalls. The phrase that resolves the stall is not "fiduciary duty allows this" or "fiduciary duty forbids this." It is a four-part question: what duty is owed, to whom, under which pool, and what documentation turns a values preference into a prudent decision record. Carry that question into the room and the veto word becomes a worklist. The field has been moving for two decades toward treating impact and ESG factors as inside fiduciary analysis rather than outside it. The UNEP FI, PRI, and Generation Foundation "Fiduciary Duty in the 21st Century" project argued that failing to consider material sustainability factors is itself a failure of prudence for a long-horizon investor. The Freshfields-authored *A Legal Framework for Impact* extended the argument to when an investor may, or should, pursue sustainability-impact goals. None of this converts fiduciary duty into a mission blank check. It moves the burden: the fiduciary who ignores a material factor now has something to defend, and the one who considers it has a documentation discipline to follow. That's the shift the operator needs to read correctly. For the operator, then, fiduciary duty is not a lawyer-only abstraction. It is vocabulary for a design constraint that governs which pool can hold a concessionary commitment, who must approve it, and what the file has to contain before the office can claim the decision was sound. ## How to Recognize It You are watching fiduciary duty handled well when the office can name, for any given commitment, which pool holds it and which standard governs it, and can produce a prudent-process record rather than a verbal assurance. The test is documentary, not rhetorical. The duty map usually answers the following: | Pool | Governing standard | What the standard permits | |---|---|---| | **Private family trust** | Trust instrument plus the Uniform Prudent Investor Act in most states | Prudence, loyalty, and impartiality across current and remainder beneficiaries; values may inform process where the instrument and beneficiary consensus support it. | | **Charitable endowment** | UPMIFA in most states | The board may weigh the charitable purpose, mission, and the institution's other resources alongside return when managing the fund prudently. | | **Private foundation, ordinary investment** | IRC section 4944 jeopardizing-investment rules plus state nonprofit law | Ordinary business care and prudence; managers may consider the investment's relationship to charitable purpose under IRS Notice 2015-62. | | **Private foundation, PRI** | IRC section 4944(c) | A primary charitable purpose with no significant income or appreciation purpose; counts toward the distribution requirement and is exempt from the jeopardizing-investment penalty. | | **ERISA plan** | ERISA "sole interest" rule | The strictest loyalty standard; collateral benefits are tightly constrained. | Two distinctions are easy to miss and expensive to get wrong. The first is the [Mission-Related Investment](mission-related-investment.md) versus the [Program-Related Investment](program-related-investment.md). An MRI is a real investment from the endowment and must satisfy the prudent-investor or UPMIFA standard; a PRI is a charitable deployment under section 4944(c) that the foundation can make at concessionary terms precisely because charitable purpose, not return, is the primary aim. Confusing the two is the most common foundation-side fiduciary error. The second is the difference between "best interests" and "sole interest." A charitable pool managed under UPMIFA operates on a best-interests analysis that can accommodate mission. A private trust and an ERISA plan operate closer to a sole-interest analysis that treats collateral benefits with suspicion. The same proposed investment can be prudent in one pool and a breach in another. > **⚠️ Process over outcome** > > A fiduciary is generally judged on the quality of the decision process at the time, not on the investment outcome with hindsight. A documented record showing the objective, the expected-return analysis, the mission link, the monitoring plan, and the approval is the defense. A good outcome with no record is luck; a poor outcome with a strong record is usually a prudent decision that did not work, which is a very different exposure. ## How It Plays Out Consider a U.S. single-family office holding a $180M private foundation, a $90M family dynasty trust, and a $40M donor-advised fund, sitting inside a broader $1.1B balance sheet. The family council, fresh from ratifying a [Family Mission Statement](family-mission-statement.md) that commits the family to regional climate resilience, asks the office to "put the trusts and the foundation behind the mission." The CIO brings three proposals to the [Investment Committee](investment-committee.md): a $20M mission-related allocation from the foundation endowment, a $5M concessionary first-loss commitment to a regional climate fund, and a request that the dynasty trust shift 15% of its public-equity sleeve into a values-screened strategy. Read as one impulse, the request sounds simple. Read through the duty stack, it is three different problems. The $20M foundation MRI is an investment. It is governed by UPMIFA and the section 4944 jeopardizing-investment rules. The committee can pursue it, but the file has to show a real expected-return analysis, a risk assessment, and the mission link, and the return objective has to be defensible as prudent for the endowment. The committee documents a market-rate target with a defined mission theme, sets a 32-basis-point tracking-budget tolerance against the policy benchmark, and records that mission relevance was a tiebreaker among prudent options, not a license to underwrite a below-market return without naming it. The $5M concessionary first-loss commitment is the one that gets misfiled. If it comes from the foundation, it can't be governed as an ordinary endowment investment, because a deliberately concessionary commitment is hard to defend under the jeopardizing-investment standard. The general counsel restructures it as a [Program-Related Investment](program-related-investment.md): primary charitable purpose, no significant income or appreciation purpose, documented as such, counting toward the foundation's distribution requirement and exempt from the section 4944 penalty under section 4944(c). The same $5M that would have been a fiduciary problem as an investment becomes defensible as a PRI, because the duty boundary changed when the deployment was named correctly. The dynasty trust request is the hardest. The trustee owes a duty of impartiality between the current income beneficiaries and the remainder beneficiaries, some of whom are not yet born. The trust instrument is silent on values screening. The trustee can't simply adopt the council's preference; the council doesn't control the trust. Counsel reviews the instrument, finds a broad investment power but no values mandate, and the trustee concludes that a values-screened strategy is permissible only if it satisfies the prudent-investor standard on its own merits, with the screening treated as one input rather than an override. The trustee documents that the screened strategy carries comparable expected risk and return to the unscreened alternative, records the beneficiary-consensus evidence from the mission statement as supporting context, and proceeds. Had the screened strategy carried a clear return concession, the trustee's impartiality duty to the remainder beneficiaries would likely have blocked it absent instrument language or beneficiary consent. The result is not that the family abandons the mission. It is that the same mission lands in three different pools under three different standards, and each one is governed by a record that would survive examination. The veto word never gets used. The office runs the four-part question on each proposal and produces a file instead of an argument. ## Caveats and Open Questions The U.S. foundation position is genuinely contested, and the book does not pretend otherwise. IRS Notice 2015-62 confirmed that private-foundation managers may consider an investment's relationship to charitable purpose when exercising ordinary business care and prudence under section 4944. That settled less than advocates hoped. It permits consideration; it doesn't compel it, and it doesn't tell a board how much return concession charitable relevance can justify before the investment becomes a jeopardizing one. The MacArthur Foundation's published framework occupies the middle ground that most careful boards now use: impact and ESG factors may be considered when the objective, the expected-return implications, the mission link, the monitoring plan, and the documentation are all clear. That is a process answer to a question the law leaves partly open. The trust side is more conservative than the foundation side, and for a reason. A foundation's charitable purpose is itself a beneficiary interest, so mission can enter the analysis directly. A private trust's beneficiaries are people, the duty of loyalty runs to them, and the duty of impartiality runs across generations of them. Northern Trust's framing of the trust-duty triad as prudence, loyalty, and impartiality is the working model, with a family-values statement serving as evidence of beneficiary consensus rather than as authority to override the standard. The contested question is how far a trustee may go on values when the instrument is silent and the beneficiaries disagree. The institutional-investor arc is the most settled of the three and the least directly applicable to a family trust. The UNEP FI, PRI, and Generation Foundation work, and the Freshfields *Legal Framework for Impact*, were written largely for pensions, insurers, and large asset owners with long horizons and many beneficiaries. Their conclusion, that ignoring material sustainability factors can be a prudence failure, transfers cleanly to a foundation endowment and a long-duration trust. Their further conclusion, that an investor may sometimes pursue sustainability impact as an end, transfers far less cleanly to a private trust governed by a sole-interest reading of loyalty. The open question underneath all three is jurisdictional. UPMIFA, prudent-investor adoption, and trust-law detail vary by state and by country; the analysis that holds in one jurisdiction can fail in another, and a cross-border family with trusts in multiple jurisdictions may owe different duties on functionally identical pools. ## Consequences The benefit of carrying the duty stack as vocabulary is that the conversation changes shape. A council request that used to trigger a reflexive yes or a reflexive no now triggers the four-part question, and the office can say yes faster when a proposal fits a pool's standard and no more cleanly when it does not. The fiduciaries stop treating "fiduciary duty" as a single wall and start treating it as a map with named layers. The discipline also makes impact claims defensible. The link between fiduciary duty and [Impact Washing](impact-washing.md) is direct: impact-washing begins when the public claim outruns what the governance file can support. A fiduciary record built to satisfy prudence, loyalty, and the relevant statute is the same record that substantiates an impact claim, because both demand the objective, the counterfactual, the evidence, and the review date. The office that governs its duties well rarely has to walk back a claim. The liability is that the analysis is genuinely hard and genuinely jurisdictional, and the office cannot do it from a template. The same investment is prudent in the foundation endowment, defensible as a PRI, and a possible breach in the dynasty trust, and only counsel reading the specific instruments and the specific state law can sort it. A family that treats fiduciary duty as a one-time legal opinion rather than as a per-pool, per-decision discipline will eventually approve something in the wrong pool, and won't notice until someone examines the file. The second-order effect is institutional maturity. An office that runs every duty-sensitive commitment through "what duty, to whom, under which pool, with what record" builds a governance muscle that compounds. The mission gets pursued, the fiduciaries are protected, and the family's purpose moves through the structure under governance rather than around it. ## Sources - UNEP FI, PRI, and Generation Foundation, [*Fiduciary Duty in the 21st Century*](https://www.unepfi.org/investment/history/fiduciary-duty/), final report 2019 — the field's foundational argument that integrating material sustainability factors is part of, not contrary to, fiduciary duty for long-horizon investors. - IRS, [Notice 2015-62](https://www.irs.gov/pub/irs-drop/n-15-62.pdf), 2015 — confirms that private-foundation managers may consider the relationship between an investment and the foundation's charitable purpose when exercising ordinary business care and prudence under IRC section 4944. - MacArthur Foundation, [*Fiduciary Duties in Investment Matters*](https://www.macfound.org/media/article_pdfs/fiduciary-duties-in-investment-matters.pdf), updated 2023 — a foundation-board framework for when ESG and impact factors may be weighed, with the objective, return implications, mission link, monitoring, and documentation discipline the prudent process requires. - Northern Trust Institute, [*Trusts and Sustainable Investing: Building the Bridge*](https://www.northerntrust.com/united-states/institute/articles/trusts-and-sustainable-investing-building-the-bridge) — names prudence, loyalty, and impartiality as the trust-duty triad and explains how a family-values statement can inform a prudent process without overriding the standard. - Freshfields, PRI, UNEP FI, and Generation Foundation, [*A Legal Framework for Impact*](https://www.freshfields.com/globalassets/our-thinking/campaigns/a-legal-framework-for-impact/a-legal-framework-for-impact.pdf), 2021 — legal analysis of when investors across major markets may, or should, pursue sustainability-impact goals within their duties. --- *This entry describes a legal and governance concept and is not legal, tax, or investment advice. Consult qualified counsel and tax advisors licensed in your jurisdiction before relying on fiduciary-duty analysis in trust, foundation, investment, reporting, or family-office governance documents.* --- - [Next: Family Employment Policy](family-employment-policy.md) - [Previous: Decision Rights Charter](decision-rights-charter.md)