Practical Tips · The Standard · June 2026
The standard for proving an insurance expert’s bias isn’t new, and it isn’t a stretch — it’s the oldest rule in adjudication. The insurer is counting on you to argue it as if it were neither.
Most of the conversation about biased insurance experts jumps straight to the evidence — how much the expert was paid, how often the insurer hired them, how many times they found against the claimant. Those are the factors. They are where the work gets done. But factors only matter if there is a standard they serve, and the standard is the part most people skip. So before any tip in this series takes up the factors, this one takes up the rule the factors exist to satisfy.
Here is the rule, and here is why it is so durable: no one whose judgment determines how a claim comes out should have a financial stake in the answer.
Notice what that does not say. It does not say “decision-maker.” The insurer will be quick to tell you its retained expert decides nothing — they render an opinion; the insurer makes the call. That is true on the org chart and beside the point. The law has never pinned the neutrality requirement to a title; it pins it to a function. When the engineer’s report concludes the damage was pre-existing, the claim is denied. When the doctor’s review concludes the disability is not supported, the claim is denied. The denial letter ratifies a determination the expert has already made. Whoever supplies the judgment that drives the outcome is the person this rule is about — whatever the title on the door.
The instinct, when you argue that an insurer’s retained expert is biased, is that you are making a novel claim. You are not. You are invoking a line of authority that runs back a century. In Tumey v. Ohio (1927) the Supreme Court held that a judge with a financial interest in the outcome of a case is disqualified — not because the judge was shown to be corrupt, but because the financial interest itself creates a temptation the law will not tolerate. Forty years later, in Commonwealth Coatings (1968), the Court extended the same logic to private arbitrators, with a twist that turns out to be decisive for insurance experts: because an arbitrator’s decision is shielded from the appellate review that polices judges, arbitrators need more impartiality scrutiny, not less. Fewer safeguards, higher standard.
Hold onto that inversion. The insurer-retained expert sits at the far end of that spectrum — selected and paid by one party, under no disclosure obligation, unchallengeable by the claimant before the evaluation, and subject to no appellate review of the expert’s neutrality at all. Under the Commonwealth Coatings logic, that absence of safeguards is the strongest case for scrutiny, not the weakest.
The second thing the standard gives you is that bias is not binary. The law has never asked “biased: yes or no.” It asks how much, along a continuum:
| Appearance of bias | Facts that would make a reasonable person doubt impartiality — the objective “reasonably entertain a doubt” test of Haworth v. Superior Court (2010). |
|---|---|
| Substantial likelihood | A serious risk of actual bias, not mere appearance — the register of Caperton v. A.T. Massey Coal Co. (2009). |
| Constitutionally intolerable | A risk so high that due process forbids the decision-maker from acting at all — per se disqualification, no proof of a corrupt motive required. |
You do not need actual proof of a corrupt motive to land somewhere on that scale. You need facts about the structure. And the more the structure tilts — more money, more repeat business, more one-sided outcomes — the further up the scale the inference climbs.
If the gradation idea sounds abstract, federal benefits law has been operationalizing it for years. Under the ERISA line of cases, a conflicted reviewer’s conflict is not a yes/no switch either — it is a factor that gets weighed, and the weight rises or falls with the evidence. A little evidence of financial entanglement earns “modest” skepticism; a lot earns enhanced skepticism. That is the same gradation in different clothes: the strength of the inference of bias determines how heavily the conflict counts. State courts evaluating insurer experts and federal courts evaluating ERISA reviewers are running the same dial.
The skeptic’s question is fair: if this standard is so well-settled, why isn’t it everywhere in insurance bias law? The answer is that it is — it has just been hiding in plain sight. Courts have adopted the inference-of-bias standard for insurer experts both directly and indirectly. And there is a quieter, more powerful affirmation: the copious decisions across the country granting claimants discovery of an expert’s compensation and their outcomes in other claimants’ files. Courts do not order that discovery as a courtesy. They order it so the claimant can infer bias from the pattern. You do not grant discovery to prove something the law refuses to recognize. Every one of those orders is a court affirming, by its conduct, that inferable bias is a real and cognizable standard.
Read the deep-dive → See the bias-evaluation service →
With the standard in place, the series turns to the factors that satisfy it — financial dependence first, in The Two Metrics That Trigger the Burden-Shift. The full paradigm — the inference-of-bias standard, the four factors, and the rebuttable presumption — is set out in Demer’s Paradigm for Assessing Biased Insurance Experts (Advocate Magazine, 2024).
Distilled from the project’s own analysis of the judicial- and arbitral-bias authorities (Tumey, Commonwealth Coatings, Caperton, Haworth) and the ERISA conflict-weighing line. Case law is cited for its principles; the inference-of-bias standard is borrowed, not invented here. Educational and informational only; not legal advice.