The SEC's 2022–2023 Staff Bulletins on Care Obligations, Conflicts of Interest, and Account Recommendations articulate a rigorous standard of care for investment fiduciaries. This 105-page report documents, with evidence, the gap between those standards and what the industry actually practices — and proposes that the gap be closed using data science tools that exist today.
The three SEC Staff Bulletins published between March 2022 and April 2023 — addressing Account Recommendations, Conflicts of Interest, and Care Obligations — collectively establish four core obligations for investment fiduciaries:
The April 2023 Care Obligations Bulletin requires advisors to understand "the expected returns, expected payout rates, and potential losses" and "likely performance in a variety of market and economic conditions" before advising on or recommending any investment.
Both the March 2022 and April 2023 Bulletins require consideration of "reasonably available alternatives." The Commission has already pursued enforcement actions against advisors who recommended higher-cost products when lower-cost alternatives were available.
The 2023 Bulletin reinforces that "reasonable investigation will require continued analysis after purchase of the investment and over the course of the relationship."
The August 2022 Conflicts Bulletin states that conflict management "should not be merely a 'check-the-box' exercise, but a robust, ongoing process." Disclosure alone does not satisfy the obligation.
The question before the Commission is not whether the standard is adequate. It is whether the industry is complying with it — and whether the current examination rate is sufficient to determine whether compliance exists.
The full report documents the evidence across 10 parts and two appendices. Request access here.
Advisors are not analyzing potential losses before recommending securities. They are not documenting why equity exposure is superior to lower-risk alternatives for specific clients. They are not monitoring holdings with independent, quantitative tools. And they are not addressing the structural fee conflict in any manner more substantive than a compliance checkbox.
The AUM fee model compounds this gap. An advisor earning a percentage of assets under management has a direct financial incentive to keep clients fully invested in equities. Moving a retiree's savings to Treasury securities yielding 4% protects the client — but reduces the advisor's income. The Commission's own August 2022 Bulletin acknowledges this conflict as universal. The question is whether the industry has addressed it in any substantive way.
Meanwhile, the alternatives that might better serve many clients — laddered Treasury portfolios, FDIC-insured certificates of deposit, no-load fixed annuities — are seldom presented. Not because they don't exist. Because they don't generate fees.
Most RIAs do not have proprietary quantitative risk models. They do not employ forensic accountants to analyze quarterly filings. The notion that they independently analyze the thousands of line items in a large public company's 10-Q filing each quarter is, for the vast majority, a fiction. They rely on brokerage research — and brokerage research is produced by organizations whose economic incentives are governed by investment banking relationships, not by the RIA's monitoring obligations. The RIA's continuous duty to monitor requires knowing when to reduce or exit a position. The RIA's only practical source of that information is an entity whose economic structure ensures the critical warning will rarely arrive.
The report documents eight companies where the business performed and the investor lost money — because the valuation at the time of purchase was too high to sustain. In every case, the mechanism was valuation compression. And in every case, the probability and magnitude of loss could have been estimated in advance using publicly available data.
| Company | Period | Revenue Growth | Stock Return | Valuation Compression |
|---|---|---|---|---|
| IBM | 1967–1993 | +2,800% | Negative real return (25 yrs) | ~60x → ~10x P/E |
| Microsoft | 1999–2016 | +310% | −1% | 30.9x → 5.4x P/S |
| NVIDIA | 2002–2015 | +339% | +5% total | 9.5x → 2.3x P/S |
| Amazon | 1999–2010 | +3,413% | +4% total | 35.9x → 1.7x P/S |
| Pfizer | 1999–2011 | +415% | −65% | 12.6x → 1.9x P/S |
| Tesla | 2014–2019 | +1,022% | −26% | 15.4x → 1.5x P/S |
| Walmart | 1999–2017 | +214% | −1% | 2.0x → 0.4x P/S |
| PayPal | 2021–2026 | +39% | −87% | 15.1x → 1.1x P/S |
None of these losses were caused by fraud or business failure. Every company continued to grow revenue throughout the period shown. The losses were caused by a single, measurable factor: the price paid relative to the company's underlying financial reality.
Every modeled scenario projected losses of −38% to −79%. The stock returned 0% over 17 years.
Every scenario projected gains of +21% to +264%. The stock rose over 2,000%.
Every scenario again projected losses. The stock lost approximately $1 trillion in market cap within months.
Sensitivity tables, NPV analysis, and ERS platform output for all three dates — presented in Parts III and VI and in Appendix B.
Request the ReportThe standard this report proposes — identify known risks, estimate their probability and magnitude, and act before those risks produce irreversible consequences — is the minimum standard enforced in every comparable profession. Only in investment management is the professional permitted to commit a client's life savings without performing any equivalent risk calculation.
An engineer who certifies a structure without calculating load capacity loses their license. The defense that "consensus supported the design" is an admission of negligence, not a defense against it.
A drug that causes adverse effects in 3% of patients must disclose that risk — even though 97% are unharmed. Approval requires dose-response analysis, not optimism.
A physician cannot prescribe a high-risk treatment without disclosing that a safer alternative exists. The professional with superior knowledge bears the duty to disclose it.
The legal scholarship is equally clear. The Least Cost Information Gatherer principle holds that the party who can obtain information most cheaply has the duty to obtain and disclose it. The investment advisor possesses — or has the professional obligation to possess — detailed knowledge of the risks, costs, and probable outcomes of the securities they recommend. The client does not. That asymmetry is the entire basis for the advisory relationship and the entire basis for the fiduciary standard.
"Caveat emptor is dead in medicine. It is dead in engineering. It is dead in pharmaceutical regulation. It is dead in real estate. It remains, as a practical matter, alive in the one profession where the consequences of information asymmetry are measured in trillions of dollars of preventable losses." — Quantifying the Standard of Care, §1.3
The report does not ask the Commission to change the standard. It asks the Commission to require that the standard be met — with evidence, with documentation, and with consequences for non-compliance. Each recommendation below is anchored in the SEC's own published guidance.
When an RIA recommends a security trading above historical valuation medians, require documented analysis of the probability and magnitude of loss under valuation reversion scenarios.
Advisors should disclose the statistical probability of loss based on historical mean reversion — not merely that "past performance is not indicative of future results."
Broker-dealer analysts are not fiduciaries and owe no duty of care to the RIA's clients. Delegating the monitoring function to non-fiduciary sources does not fulfill the continuous duty of care.
Every equity recommendation should be explicitly compared to the risk-free rate. For clients whose primary need is income and capital preservation, the advisor must document why equity exposure is superior to Treasuries, FDIC-insured deposits, or fixed annuities for that specific client.
Require RIAs managing individual securities to evaluate, at minimum quarterly, three indicators: declining free cash flow, deteriorating balance sheet health, and valuation divergence from fundamentals — using the company's own published financial statements.
Analyst target prices should be accompanied by a probability panel — analogous to a nutrition label — showing the probability of reaching the target, the probability of significant decline, and the assumptions used.
The full regulatory recommendations, including proposed disclosure standards and monitoring frameworks, are presented in Part IX of the report.
Request the ReportUsing publicly available financial data, documented historical base rates, and standard statistical methods, it is possible today to calculate the probability that a stock at a given valuation will deliver negative returns over a specified horizon; to compare that probability against the guaranteed yield of a Treasury security; to generate that comparison for every holding in a client's portfolio; to update it quarterly; and to document it in a format that is auditable by examiners.
As in actuarial science, the goal is not certainty — it is probability. The fiduciary standard does not require predicting the future. It requires identifying conditions under which harm becomes materially probable and acting before those conditions produce irreversible consequences.
Approximately 15% of the registered investment adviser population is examined in any given year. The remaining 85% operate with no mechanism to verify that the obligations articulated in the Commission's own published Bulletins are being fulfilled. The gap between obligation and practice persists not because the standard is unclear, but because compliance is neither measured nor enforced at scale.
A person with $500,000 in retirement savings cannot afford a 50% drawdown. When that investor is told that broad index exposure is "safe" because it is diversified — without a clear explanation of valuation risk, drawdown history, and probability ranges — the result is not an informed choice. It is a decision made with material information missing.
The history of consumer protection in America is instructive. The automobile industry did not voluntarily install seatbelts. Pharmaceutical companies did not voluntarily disclose side effects. In every case, the industries that manage the public's safety were reformed by the institutions entrusted with that responsibility — not by the industries themselves.
We are committed to making ourselves available to regulators and policymakers to explain these methods, share evidence, and help implement standards that are practical, testable, and focused on the minimum disclosure the public deserves.
"Sunlight is said to be the best of disinfectants; electric light the most efficient policeman." — Justice Louis Brandeis, Other People's Money (1914)
Quantifying the Standard of Care — 105 pages documenting the gap between the standard the SEC has articulated and the standard the industry practices, with a framework to close it.