The Board's Next Liability Exposure: Did You Independently Assess Whether the CEO Could Carry What You Approved?

Caremark_Board_Liability_CEO_Capacity_Gaconnet_2026.md.pdf

Download PDF • 153KB

Caremark Oversight Duties Are Expanding. The Question of Independent Executive Capacity Measurement Has Not Been Litigated. It Will Be.

Don L. Gaconnet, CSE III

Founder & Principal Investigator, LifePillar Institute for Structural Identity Sciences

ORCID: 0009-0001-6174-8384 · SSRN Author ID: 7657314

June 2026

In 1996, Chancellor William Allen established in In re Caremark International Inc. Derivative Litigation that a board's failure to ensure a reasonable information and reporting system constitutes a breach of the duty of loyalty. The standard was deliberately narrow: directors would be liable only if they acted in bad faith — defined as a sustained or systematic failure to exercise oversight.

For nearly three decades, the standard held as one of the most difficult claims for plaintiffs to sustain. That is changing.

In January 2023, the Delaware Court of Chancery held in In re McDonald's Corporation Stockholder Derivative Litigation that corporate officers — not just directors — owe Caremark oversight duties. The Chief People Officer was held to a duty of oversight within his functional area of responsibility.

In September 2025, the Court denied dismissal of Caremark claims in Giuliano v. Grenfell-Gardner et al. (the Teligent case), permitting claims against both the CEO and Chief Scientific Officer to proceed. The basis: compliance fell within their respective areas of responsibility and they failed to report known regulatory issues to the board.

The trajectory is clear. Caremark duties are expanding in scope (from directors to officers), in reach (from financial compliance to operational risk), and in enforcement (from near-impossible plaintiff burden to a 30% survival rate on motions to dismiss, according to American Bar Association analysis).

One category of oversight risk has not yet been tested: the structural capacity of the key executive to carry the mandate the board approved.

When it is tested, the board that independently measured the executive's structural condition will have a documented defense in the governance file. The board that relied on the executive's self-report will have the same defense as every board that failed to implement a reasonable information system: none.

What Caremark Requires

The Caremark framework establishes two prongs of potential liability:

Prong one — the Information Systems Claim — asks whether the board utterly failed to implement any reporting or information system to monitor key corporate risks. The board must have a system. The system does not need to be perfect. It needs to exist and function at a reasonable level.

Prong two — the Red Flags Claim — asks whether the board, having implemented a system, failed to monitor it or respond to red flags the system produced. The board must not ignore what the system reveals.

The standard of care is not perfection. It is reasonable oversight — a good faith effort to implement information systems that identify risks fundamental to the organization's operations, and a good faith effort to respond to the information those systems produce.

Every mission-critical operational risk in a modern organization is monitored through independent information systems. Financial risk is monitored through audited financials and internal controls. Legal risk is monitored through compliance programs and outside counsel review. Cybersecurity risk is monitored through threat detection systems and third-party audits. Operational risk is monitored through performance management systems and process controls.

Key person risk — the structural capacity of the executive whose decisions determine whether the organization's strategy succeeds or fails — is monitored through the executive's self-report.

The board asks the CEO how they are doing. The CEO says they are doing well. The board records the response. That is the information system for the most consequential risk in the organization: whether the person carrying the strategy can sustain the weight of carrying it.

Why Self-Report Does Not Satisfy Caremark

The Caremark standard requires a reasonable information system. The question is whether accepting the CEO's self-report as the primary information system for executive capacity risk satisfies the reasonableness standard.

Six independent research programs confirm that self-assessment under structural load is systematically unreliable:

Davis et al. (2006) found minimal to no correlation between physician self-assessment and externally observed competence across 725 subjects. Eva and Regehr (2005) established that self-assessment accuracy does not improve with expertise. The ACE study (Felitti et al., 1998) documented superadditive dose-response compounding of stress effects across 17,000 subjects. The human factors workload assessment literature (Hart & Staveland, 1988; Webster et al., 2018) explicitly recommends external physiological measurement because self-report is unreliable under operational stress.

A 10,000-case Monte Carlo simulation (Gaconnet, 2026; SSRN 7657314; DOI: 10.17605/OSF.IO/MVYZT) quantified the error rates in the near-capacity executive population: 81.4% misidentify the domain where their structural failure lives. 73.0% minimize the depth. 61.1% are simultaneously wrong about both domain and depth.

On May 28, 2026, Duncan et al. published in JAMA Network Open that the standardized diagnostic interview — the clinical assessment instrument most analogous to the structured executive interview — demonstrates only moderate test-retest reliability (κ = 0.69) across 57 studies and 8,146 participants. The instrument itself does not produce consistent results. The data source feeding the instrument — the subject's self-report — is structurally unreliable under load.

The forensic accounting analogy is direct. A board that accepted the CFO's verbal assurance that "the books are clean" as its financial monitoring system would not satisfy Caremark's reasonableness standard. The board retains independent auditors who read the books directly, produce documented findings, and report to the audit committee. The auditor does not ask the CFO to describe the books. The auditor reads the books.

A board that accepts the CEO's verbal assurance that "I'm handling it well" as its executive capacity monitoring system is relying on a data source that is wrong about domain 81.4% of the time, wrong about depth 73.0% of the time, and most wrong in exactly the cases where the structural failure carries the highest organizational consequence.

Whether that constitutes a reasonable information system under Caremark is a question Delaware courts have not yet answered. The question will be asked.

The Precedent Path

The expansion of Caremark from directors to officers (McDonald's, 2023) and the increasing survival of Caremark claims past motions to dismiss (Teligent, 2025) establish a directional trajectory. Courts are examining more carefully whether boards have implemented meaningful oversight systems for risks fundamental to the organization's operations.

The question of CEO structural capacity is fundamental to the organization's operations by any reasonable definition. The CEO is the single individual whose decisions, judgment, strategic direction, and operational execution determine whether the organization's strategy succeeds. If the CEO's structural capacity to carry the mandate degrades — if the system sustaining the CEO's performance fails — the organizational consequence is documented: 65% of PE firms replace the CEO during the hold period, 83% say unplanned turnover extends the hold, and nearly half say it erodes returns (AlixPartners, 2026).

The board approved the strategy. The board approved the CEO to execute the strategy. The board has a Caremark duty to implement a reasonable information system to monitor the risks fundamental to the strategy's execution. The CEO's structural capacity to carry the strategy is a risk fundamental to the strategy's execution.

The question that will be litigated is not whether the board should have predicted the CEO's failure. The question is whether the board implemented a reasonable system to monitor the CEO's capacity to sustain the mandate — or whether the board relied entirely on the CEO's self-report, knowing (or now being put on notice) that self-report under structural load is systematically unreliable at published, quantified rates.

What a Reasonable Information System Looks Like

An information system for executive capacity risk that satisfies the Caremark reasonableness standard would need to meet the same structural criteria as every other oversight system the board relies on:

Independence. The measurement must be independent of the subject being measured. The financial audit is independent of the CFO. The compliance review is independent of the general counsel. The executive capacity assessment must be independent of the CEO's self-report.

Documentation. The measurement must produce a documented finding that enters the governance record. The audit produces a written opinion. The compliance review produces a written report. The executive capacity assessment must produce a written structural finding that goes in the board's governance file.

Periodicity. The measurement must occur at regular intervals, not only in response to observable crisis. The audit occurs annually. The compliance review occurs on a defined schedule. The executive capacity assessment must occur at intervals that allow the board to monitor the trajectory before a crisis materializes — not after.

Responsiveness. The board must review the findings and respond to what the system reveals. If the financial audit identifies a material weakness, the board cannot ignore it. If the executive capacity assessment identifies structural degradation, the board cannot ignore it.

These are not novel standards. They are the same standards applied to every other oversight system in the Caremark framework. The only difference is the substrate being monitored: not financial accounts, not legal compliance, not cybersecurity posture — the structural condition of the person whose decisions determine everything else.

The Standard of Care That Does Not Yet Exist

No published standard of care currently requires boards to independently assess executive structural capacity. The standard does not exist because the assessment category does not exist at scale. Boards cannot implement a system that the market does not offer.

This is structurally identical to the position forensic accounting occupied before it became a standard of care in the early 1990s. Before forensic accounting was established, boards relied on management representations about financial condition. When forensic accounting became available as a professional services category, the standard of care shifted: boards that continued to rely solely on management representations when independent forensic assessment was available faced increased liability exposure.

The structural assessment of executive capacity is in the same position forensic accounting was before it became standard. The assessment category exists. It is available through professional services engagement. It produces the independent, documented, engineering-grade finding that the Caremark framework would require. The question is how quickly the standard of care shifts from "CEO self-report is sufficient" to "independent structural measurement is expected."

The board that adopts independent structural measurement before the standard shifts will have the documented defense when the question is litigated. The board that waits for the standard to be established by judicial ruling will be the board that established the precedent — as the defendant.

What the Board Should Do Now

The practical recommendation is not that every board must immediately implement executive structural measurement. The recommendation is that every board should be aware that the following conditions exist simultaneously:

Caremark oversight duties are expanding — from directors to officers, from financial compliance to operational risk, from near-impossible plaintiff burden to a meaningful survival rate on motions to dismiss.

Self-report under structural load is documented as systematically unreliable — across six independent research programs spanning decades, confirmed by a JAMA Network Open meta-analysis published May 2026, and quantified at specific error rates in the near-capacity executive population.

CEO structural failure produces documented organizational losses — 65% replacement during hold, 83% extended hold periods, year-two spike, returns erosion, all quantified in the industry's own surveys.

An assessment category that addresses this gap exists — independent, instrument-based structural measurement that does not depend on the CEO's self-report, produces a documented finding, and enters the governance file alongside the financial audit and the compliance review.

The board that is aware of these conditions and chooses to implement independent structural measurement has a documented, defensible position. The board that is aware of these conditions and chooses not to act has made a documented decision not to act. Both positions are recorded in the governance file.

The question is not whether the board will face this question. The question is when — and whether the governance file contains an answer when it arrives.

References

AlixPartners. (2026). 11th Annual PE Leadership Survey. March 2026.

American Bar Association. (2024). Boards' duty of oversight: From Caremark to the continuing travails of Boeing. Business Law Today. May 2024.

Bryan Cave Leighton Paisner. (2023). Recent Delaware cases clarify Caremark oversight duties. April 2023.

Davis, D. A., et al. (2006). Accuracy of physician self-assessment. JAMA, 296(9), 1094–1102.

Duncan, L. J., Xie, W., et al. (2026). Test-retest reliability of standardized diagnostic interviews. JAMA Network Open. DOI: 10.1001/jamanetworkopen.2026.15039.

Eva, K. W., & Regehr, G. (2005). Self-assessment in the health professions. Academic Medicine, 80(10), S46–S54.

Gaconnet, D. L. (2026). The Recursive Reliability Effect. LifePillar Institute. SSRN 7657314. DOI: 10.17605/OSF.IO/MVYZT.

Hart, S. G., & Staveland, L. E. (1988). Development of NASA-TLX. Advances in Psychology, 52, 139–183.

In re Caremark International Inc. Derivative Litigation, 698 A.2d 959 (Del. Ch. 1996).

In re McDonald's Corporation Stockholder Derivative Litigation (Del. Ch. Jan. 25, 2023).

Giuliano v. Grenfell-Gardner et al. (Del. Ch. Sept. 2, 2025).

Jenner & Block. (2026). Delaware Court denies dismissal of Caremark oversight claims in Teligent case. February 2026.

Taft Law. (2023). Corporate officers are subject to Caremark claims for breach of duty of oversight. January 2023.

Webster, C. S., et al. (2018). Psychophysiological measurement of cognitive load. British Journal of Anaesthesia, 120(6), 1145–1158.

Don L. Gaconnet, CSE III

Cognitive Systems Engineer III

Founder & Principal Investigator, LifePillar Institute for Structural Identity Sciences

ORCID: 0009-0001-6174-8384 · SSRN: 7657314

Institute: lifepillarinstitute.org/research · Practice: dongaconnet.com/writings

Lake Geneva, Wisconsin · don@lifepillar.org

Copyright © Don L. Gaconnet, June 2026. All rights reserved. The assessment instrument, its operational architecture, scoring methodology, and all associated protocols are proprietary trade secrets of Don L. Gaconnet and the LifePillar Institute for Structural Identity Sciences.