Biestek v. Berryhill
The Court ruled that a vocational expert's refusal to hand over her private job-market survey data, after a disability applicant asked for it, does not automatically disqualify her testimony from counting as enough evidence to support a benefits denial.
Instead of a fixed rule, judges must weigh the refusal along with everything else in the record on a case-by-case basis, deferring to the judge who actually saw the hearing.
“It means-and means only-"such relevant evidence as a reasonable mind might accept as adequate to support a conclusion."”
The majority's description of how low the substantial-evidence bar actually is.
How it got here: A federal trial court and the Sixth Circuit rejected Biestek's argument that the expert's testimony couldn't be substantial evidence; he asked the Supreme Court to adopt a categorical rule against it.
The Case in Depth
What happened
Michael Biestek, a former construction worker, developed disc disease, Hepatitis C, and depression and applied for Social Security disability benefits. At his hearing, a vocational expert testified that thousands of sedentary jobs existed that Biestek could still perform. When his lawyer asked to see the private survey data behind those numbers, the expert refused, calling it confidential client information, and the judge did not require her to produce it.
The question before the Court
If a disability applicant asks a government expert witness to hand over the job-market data behind her testimony and she refuses, does that automatically mean her testimony can't count as enough evidence to deny benefits?
The Court's answer
No — the Court held that an expert's refusal to produce her supporting data upon request does not automatically knock her testimony out of counting as substantial evidence. The parties agreed that an expert's testimony can qualify as substantial evidence even without any supporting data being offered at all, so the Court saw no reason why a refusal to produce that same data, once asked, should always change that result.
Instead, judges must look at everything in the record — the expert's credentials, how she answered questions, whether her refusal seems suspicious, and whether cross-examination was possible without the data — and decide case by case whether the evidence clears the low bar of being "more than a mere scintilla." Sometimes a refusal will sink the testimony; sometimes it won't. The Court did not decide whether Biestek's own case actually had enough evidence, because he had only asked for the categorical rule, not a ruling on his specific facts.
Curious how the Court got there? See the step-by-step legal reasoning →
Why it matters
Disability applicants who challenge the job-availability testimony of agency-hired experts will not get an automatic win just because the expert withheld her underlying survey data. Judges reviewing benefits denials must instead weigh the refusal against the rest of the record, meaning outcomes will vary case by case rather than following one clear rule.
What changes now
The Sixth Circuit's judgment affirming the denial of benefits for the earlier period stands, since Biestek only asked the Court to adopt a categorical rule and did not separately argue that his specific record lacked substantial evidence. Future disability cases involving experts who withhold data will be decided case by case by ALJs and reviewing courts, without a bright-line rule either way. This is a final merits decision, not a temporary order.
What this does not decide
The Court did not decide whether the actual evidence in Biestek's case — the expert's bare conclusion with no data or details provided — was itself substantial enough to support denying him benefits. It resolved only the broader, categorical legal question, leaving the fact-specific question open, a point both dissents stressed the Court should have reached.
Concurrences and dissents
Dissent — Justice Sotomayor
Justice Sotomayor agreed the categorical rule should be rejected but argued the Court should have gone further and found that, on these specific facts, the expert's conclusory testimony — offered with no detail about her surveys and cut off from cross-examination by the ALJ — did not amount to substantial evidence. She stressed the ALJ's duty to fully develop the record and faulted the judge for shutting down further questioning.
Dissent — Justice Gorsuch
“An expert who supplies nothing but a bottom line supplies nothing of value to the judicial process”Gorsuch's central objection to relying on the expert's unsupported conclusion.
Justice Gorsuch, joined by Justice Ginsburg, argued that a bare, unsupported bottom-line conclusion from an expert who refuses to back it up should never count as substantial evidence, drawing on cases holding that conclusory, speculative, or unverifiable testimony fails that standard. He argued the undisputed facts of Biestek's case show the expert's testimony was exactly this kind of insufficient evidence, and would have ruled for Biestek instead of ducking the fact-specific question.
How the Court got there
The legal reasoning, step by step
- The Court explained that 'substantial evidence' is a long-used legal standard for reviewing agency fact-finding that sets a low bar: it means only the kind of evidence a reasonable person might accept as adequate to support a conclusion, more than a bare scintilla but far short of a preponderance.
- The Court started from common ground between the parties: an expert's testimony about job numbers can count as substantial evidence even when she never produces any supporting data at all, as long as nothing else in the record undermines her testimony.
- Reasoning from that starting point, the Court asked why adding one more fact — the expert's refusal after being asked — should flip an otherwise-sufficient opinion into an automatically insufficient one, when the refusal itself changes nothing about the substance of what she said.
- The Court acknowledged that a refusal can support an 'adverse inference' that the expert is hiding something, and can sometimes hamper an applicant's ability to cross-examine the expert, but concluded that whether either problem actually undermines the evidence depends on the rest of the record, such as the expert's credentials, the plausibility of her reasons for confidentiality, and how she answered other questions.
- Because the effect of a refusal varies depending on case-specific facts, the Court concluded that no categorical rule could work; instead, judges reviewing an agency's benefits decision must weigh the refusal together with the whole record, deferring to the administrative judge who observed the hearing directly.
Doctrinal impact
Cases affected by this decision
Reaffirms Perales (402 U.S. 389)
The Court leans on this case's rejection of a categorical rule for medical evidence to reject one here too.