When the Social Security Administration (SSA) denies an application for Social Security Disability Insurance (SSDI) benefits, the applicant is entitled to a hearing. One of the issues that an administrative judge will decide is whether jobs are available for a person with the applicant’s disability, in light of the applicant’s education, work experience, and other characteristics.

Administrative judges often rely on expert testimony about the availability of work in the economy. In Biestek v. Berryhill, the expert based her opinion on private market-survey data. The applicant asked to review that data but the expert refused to provide it.

The judge relied on the expert’s opinion despite the expert’s refusal to provide the underlying data to the applicant. The United States Supreme Court was asked to decide whether an expert’s opinion can support a denial of SSDI benefits when the expert relies upon data that she keeps secret.

Facts of the Case

Michael Biestek was a carpenter and construction worker. A combination of degenerative disc disease, Hepatitis C, and depression prevented him from working in his occupation. He applied for SSDI benefits.

The question at Biestek’s SSDI hearing was whether he was capable of transitioning to less physically demanding work. The judge needed to determine what kind of work Biestek could do and whether those jobs existed in significant numbers.

The judge used the services of Erin O’Callaghan, a vocational expert who was under contract with the SSA. Like many vocational experts, she also worked in the private sector helping people with disabilities find employment.

O’Callaghan opined that Biestek could work as a bench assembler or sorter. She testified that there are 240,000 bench assembler jobs and 120,000 sorter jobs in the national economy.

Biestek’s lawyer asked O’Callaghan about the source of those numbers. She replied that they came from her own labor market surveys. (She also relied on government sources, but those sources do not provide the specific detail about which she testified.)

Biestek’s lawyer asked O’Callaghan to produce those surveys. She refused, claiming that they contained confidential information because they were in her client files. The lawyer suggested that O’Callaghan produce the surveys after removing the clients’ names. She again refused. The judge stepped in and ruled that she was not required to produce the information.

Judge’s Ruling Appealed

The judge awarded Biestek benefits that would start after four years, when his advancing age would make it difficult for him to find employment. Biestek appealed, arguing that there was no reliable evidence that any work existed in the economy that he would be qualified to perform during the four-year period during which benefits were denied. He argued that O’Callaghan’s opinion should not be considered because it was based on data that Biestek was not allowed to review or challenge.

The District Court and the Sixth Circuit ruled against Biestek. Since the Sixth Circuit’s ruling conflicted with a decision from the Seventh Circuit, the Supreme Court agreed to resolve the conflict.

Supreme Court Decision

The SSA rules allow vocational experts to rely on publicly available sources of data and on information they privately acquire from employers or that they develop from their own experience. In a federal lawsuit, data upon which experts rely is usually discoverable, and courts deciding Daubert motions must determine whether the data is sufficient to support the expert’s opinion.

The Supreme Court decided that administrative proceedings are different. Administrative rules require decisions to be based on “substantial evidence,” although court decisions tend to regard that phrase as meaning “any evidence at all.” As the Supreme Court noted, “substantial” in this context means “more than a mere scintilla.”

Biestek did not ask the Court to adopt a procedural rule requiring production of a vocational expert’s underlying data. As the Court noted, that rule exists in court procedures that apply to federal lawsuits but no comparable rule requires production of an expert’s data in SSA proceedings. Nor did the Court seem inclined to adopt such a rule, even if it had been asked to do so.

The Court declined to rule that a vocational expert’s opinion in an SSA hearing will never constitute “substantial evidence” of available employment if the expert refuses to disclose the data upon which the opinion is based. The Court concluded that a qualified expert who answers questions credibly, who uses a widely accepted methodology, and whose testimony is not contradicted by other evidence may be deemed to have provided “substantial evidence” upon which an SSA judge may rely. It is up to the judge in each case to decide whether the evidence is substantial, even if the expert refuses to produce underlying data upon which it is based.

It’s All Up to the SSA Judge

The Court noted that an SSA judge might decide not to credit an expert’s testimony if the expert “has no good reason to keep the data private and her testimony lacks other markers of reliability.” If the judge accepts the testimony despite the expert’s refusal to provide underlying data, reviewing courts must decide case-by-case whether the expert’s testimony created “more than a mere scintilla” of evidence. Given the low bar, the reviewing court will nearly always agree that the judge’s decision is based on substantial evidence.

The Court acknowledged that “the testimony would be even better—more reliable and probative—if she had produced supporting data; that would be a best practice for the SSA and its experts.” In fact, the SSA’s Vocational Expert Handbook instructs vocational experts to have their supporting data available if the judge asks to see it, but it does not require the judge to do so.

A judge who questions an expert’s credibility might therefore reject the testimony of an expert who refuses to produce the data. Since judges pick the expert, however, it is unlikely that a judge will question the expert’s credibility. Thus, when a judge lets the expert keep her underlying data secret, an SSDI applicant will often be out of luck.

Dissenting Opinions

The majority opinion did not decide whether substantial evidence supported the denial of Biestek’s benefits. The majority explained that it was not asked to decide that question.

Justice Gorsuch, joined by Justice Ginsburg, thought that the questions were inseparable. Justice Sotomayor wrote a separate dissenting opinion that agreed with most of Justice Gorsuch’s reasoning.

While Justice Gorsuch agreed with the majority that an expert’s opinion might constitute substantial evidence in some cases, even if underlying data is not produced, he based his analysis upon precedent that deems an expert’s “conclusory” opinion to be incapable of meeting the substantial evidence standard.

Courts generally agree that experts must do more than provide a “bottom line” opinion. They must support those opinions with sufficient data and reasoning. An expert who claims to have supporting data but refuses to produce it offers little more than a conclusory, bottom-line opinion.

Justice Gorsuch thought O’Callaghan’s opinion, supported only by her secret data, was conclusory. He concluded that “an agency expert’s bottom-line conclusion, supported only by a claim of readily available evidence that she refuses to produce on request, fails to satisfy the government’s statutory burden of producing substantial evidence of available other work.”

Underlying Justice Gorsuch’s approach is the understanding that SSA denies most claims for disability benefits. The SSA has the burden of proving that substantial jobs are available. When an SSA judge hires a vocational expert and then excuses the expert from producing data that justifies her opinion, the SSA judge puts a thumb on the scale by making it impossible for the disabled applicant to challenge the expert’s opinion. That approach reduces the number of claims the SSA must pay, but it isn’t fair to the disabled applicant.

The bottom line is that, in most cases, experts should not rely on data that they are unwilling to produce. In Social Security cases, and perhaps in some other administrative cases, experts may be able to shield their data, but it will be up to the administrative judge to decide whether the expert’s opinion should carry any weight when the opinion amounts to a “bottom line” conclusion that is unsupported by any data that the expert is willing to reveal.