A fair CEO review when the family relationship is part of the decision
Family Office Private Wealth Journal · Editorial research desk
Family Office Private Wealth Journal · 2026-10-26
CC BY 4.0 · https://creativecommons.org/licenses/by/4.0/
Source: https://journal.research.university/journals/family-office-private-wealth/articles/expansion-10-1371-journal-pone-0285029-solutions

Abstract

AI-assisted editorial analysis of Yu and colleagues’ study of CEO dismissals in Chinese listed family firms. This companion proposes an auditable retention and succession process for family-controlled operating assets, with role-based standards, independent challenge and explicit limits on transferring the evidence to family offices. These procedures are editorial proposals, not interventions tested in the source.

The decision problem: family membership is not one category

A family office can carefully underwrite a business and still leave its most consequential governance decision informal: who remains in charge when performance deteriorates. Treating every relative as an equally protected insider conceals differences in influence, access and expectations. Yu, Sun, Cheng, Lin and Li examine a narrower empirical question: whether family CEOs connected through blood relationships and those connected through marriage face different dismissal patterns. Their analysis challenges the assumption that a business-owning family always acts as a single unit with aligned interests. For an office overseeing operating companies, the useful question is whether the board can show that its reasons for retaining or replacing a CEO would remain defensible if the person’s family relationship were different.

The study does not establish that any category of relative is more competent, trustworthy or deserving of office. Its evolutionary and socioemotional explanations are theoretical interpretations of observed patterns, not tests of an individual’s character or biological fitness. The practical response should therefore be consistent evidence and procedural fairness. Do not convert the paper into a hiring screen, a family hierarchy or a presumption against spouses and in-laws. The proposed controls apply to every executive, including founders, descendants, spouses, in-laws and professionals outside the family.

What the source actually measured

The source reports 1,942 observations across 455 Chinese listed family firms. The family-firm definition requires at least one additional family member working in the business beyond the chairman and family ownership of at least 20 percent. The study also excludes firms where the chairman is not a controlling family member. Financial and governance information comes from CSMAR, while kinship information is hand-collected from offering documents and subsequent public disclosures. The authors classify a CEO as blood-related when at least one other family member holding specified roles has a blood relationship with that CEO; a robustness analysis instead focuses on the relationship with the chairman. That distinction matters: the principal measure is a relationship within the observed management and governance structure, not a universal measure of closeness to an entire family.

The reported baseline fixed-effects coefficient for a blood-related CEO is negative (B = −0.064, p < 0.05). The authors interpret this as a lower likelihood of dismissal than for CEOs related through marriage. They also report stronger differentiation under poorer performance and greater family ownership. However, the interaction terms in the separate models are reported at p < 0.1, while the combined model reports p < 0.05. A board should not turn these coefficients into a personal dismissal probability, a monetary valuation adjustment or a threshold for acceptable ownership concentration. No proposed governance intervention is experimentally evaluated.

Dismissal is itself inferred partly through what happens to concurrent posts. When a departing CEO keeps other positions, the authors treat that pattern as consistent with dissatisfaction rather than an objective inability to continue working. This is a consequential measurement choice. A person can retain another role for reasons unrelated to dismissal, and private disputes may not appear in public documents. In a real board review, establish the reason from contemporaneous records and the executive’s account; do not classify an individual departure from the research proxy alone.

Set the standard before the difficult quarter

The board chair and company secretary should record an annual CEO mandate before performance is known. Separate controllable operating commitments from market exposure, capital allocation, risk controls, talent development and conduct. Identify which measures are leading indicators, which are lagging outcomes and which require qualitative judgment. For each measure record the baseline, reporting owner, assessment period and material limitations. The family office investment lead can reconcile the mandate with the original underwriting thesis, but should not quietly replace the board’s authority with an investor preference.

Document the same escalation logic for a family and a nonfamily CEO in comparable roles. An agreed financial covenant breach may call for immediate liquidity planning; a missed sales forecast may call for an explanation and revised operating plan. Neither automatically establishes incompetence. Conversely, a family relationship does not excuse concealed losses or a refusal to provide records. The mandate should state when a performance-improvement plan, external assessment, emergency interim appointment or formal succession review becomes appropriate. Counsel should check the process against actual contracts and applicable duties before implementation; this companion supplies no jurisdiction-specific legal conclusion.

Separate relationship conflicts from performance evidence

For each retention or succession decision, the company secretary should prepare a restricted conflict register. Record decision-makers’ management roles, economic interests and relevant personal relationships only to the extent needed for the decision. Do not collect genetic information or rank family members by ancestry. A declared conflict is a reason to design participation and challenge carefully, not proof of wrongdoing. The board should decide who presents evidence, who deliberates and who votes, with the reason for each arrangement recorded. Where all directors are closely involved, obtain a properly authorized independent assessment rather than claiming independence that does not exist.

Use two written cases: the strongest supported case for retention and the strongest supported case for replacement. Require comparable evidence quality and allow the CEO to respond to factual allegations. A facilitator can ask whether the board would attribute the same setback to the market if another candidate held the role, or whether it would demand a turnaround plan from one executive but accept reassurance from another. These counterfactual questions do not prove discrimination. They expose inconsistent explanations while there is still an opportunity to correct the process. Keep conclusions tied to role requirements and documented conduct.

Stress-test governance when ownership and performance change

The paper’s interaction findings make a downturn and a concentration of voting influence sensible occasions for a governance check, not automatic evidence of favoritism. At the next annual review, the board risk lead should run a tabletop exercise combining weaker cash flow with an ownership transition. Ask who can authorize a turnaround, who can remove the CEO, who would act temporarily and whether liquidity decisions can continue during a dispute. Record any mismatch between formal powers and how participants expect decisions to occur. Complete the exercise before a cash crisis makes deliberation impossible.

An independent director or authorized reviewer should inspect a small set of completed decisions for differences in evidence requirements, improvement periods and treatment of external shocks. In a small family business, dismissal counts will usually be too sparse for meaningful statistical comparisons. Use structured case review and denominators that describe what was inspected; do not present one or two cases as an estimated causal effect of kinship. Review the successor-selection process separately, since a procedurally sound dismissal does not establish that the replacement was chosen fairly or can perform the job.

Measures, stopping conditions and evidence limits

For a first implementation cycle, measure process completion: mandates agreed before assessment, decisions with documented alternatives, conflicts addressed before voting, executive responses retained, and interim authority assigned. Track completion time and material factual corrections. Treat a higher dismissal rate as neither success nor failure. The relevant early outcome is a decision that can be reconstructed and challenged; operating results and trust may take longer to assess and have many other causes. The suggested first-cycle review is an editorial implementation choice, not a duration tested by Yu and colleagues.

Stop a proposed decision process if evidence has been fabricated, a conflicted participant controls the supposedly independent assessment, the mandate is rewritten after results are known, or protected personal information is being gathered without a legitimate purpose. Escalate immediate threats to company continuity through the authorized board and counsel rather than delaying essential action for a research-inspired checklist. Suspend any claimed causal or quantitative inference if the records cannot distinguish voluntary departure from dismissal. If the paper is later corrected or retracted, flag the companion and reassess its empirical statements; general procedural controls still require their own practical justification.

This is adjacent evidence from listed family operating businesses, not a study of single-family offices, multifamily offices, investment committees or unlisted holding companies. The authors emphasize the Chinese context, often first-generation firms with relatively simple kinship structures, and the need to test other countries and unlisted firms. Observational associations can reflect omitted circumstances, selection and measurement choices. The study cannot establish that the proposed process improves returns, reduces conflict or causes fairer decisions. This AI-assisted editorial analysis has not received documented human acceptance and does not imply endorsement by the source authors.

Source, attribution and limits

[1] Xiaodong Yu; Shize Sun; Xirong Cheng; Yize Lin; Huan Li. Why do family firms dismiss their family CEOs? A perspective on kinship ties. PLOS ONE (2023-05-04). https://doi.org/10.1371/journal.pone.0285029. CC BY 4.0.

Observational panel; family CEO kinship and dismissal classifications rely on public records and inferred reasons for departure. Chinese listed, often first-generation family firms; unlisted firms, other cultures and family offices were not tested. Interaction significance varies across specifications. No causal estimate of favoritism or evaluation of the companion procedures.

Prepared by the Family Office Private Wealth Journal editorial research desk with AI assistance at the direction of Dr. Freedom Cheteni. Original research authors retain full credit and are not authors or endorsers of this companion. This proposed process has not been evaluated in a new empirical study; no independent peer-review acceptance is claimed. This is educational analysis, not individualized investment, legal or tax advice.

Rights: © 2026 The Research University.
© 2026 The Research University. This original editorial text is licensed Creative Commons Attribution 4.0 International. Attribute the title, editorial research desk and publisher, link to the license and indicate changes. Underlying research, talks and recordings retain their separate attribution and rights.
Edition changes: Original AI-assisted editorial synthesis. Research findings, transcript themes and proposed practices remain distinct. No new participant data or independent peer-review acceptance is claimed.