Aaron Brace v. Andrew M. SaulAaron Brace v. Andrew M. Saul
Brace, now age 40, was injured on the job in 2013 and can no longer perform his past work in health service, food service, and construction. He applied for disability benefits, claiming persistent back and neck pain and a host of other conditions. An administrative law judge received evidence of his medical and work history and applied the five-step analysis found in
The fifth and final question was whether Brace can nonetheless perform a significant number of other jobs in the national economy. The vocational expert testified that Brace could perform three jobs: a callout operator, semiconductor bonder, or registration clerk. The expert added a fourth—a counter clerk—subject to certain lifting, standing, and walking restrictions. He further testified that a significant number of jobs exist across the four job categories—an estimated 140,000 in total.
The expert’s jobs estimate is critical to this appeal. If a claimant cannot perform his previous work, the Social Security Administration bears the burden of showing that a significant number of other jobs are available to the claimant. Young v. Barnhart, 362 F.3d 995, 1000 (7th Cir. 2004); see also
Vocational experts often rely on the Dictionary of Occupational Titles, a database of job titles that has not been updated in almost 30 years. Occupational Information System Project, SOC. SEC. ADMIN., https://www.ssa.gov/disabilityresearch/occupational_info_systems.html (last visited Aug. 12, 2020). The database does not list the number of jobs associated with each job title, so the vocational expert must perform an estimate. Because the database of job titles is so outdated, an expert’s methodology for connecting job titles to reliable estimates of the number of jobs for each title is especially important. The Social Security Administration has itself acknowledged this issue and expressed an intent to update the database, but the new version has not yet arrived. Id.
Brace’s attorney asked the vocational expert (“VE”) to explain his methodology for estimating that 140,000 jobs are available in the national economy. The expert answered that no study (and no software available to him) lists the number of jobs associated with any job category in the database, including the categories that he assigned to Brace. Instead, he linked the job categories from the Dictionary of Occupational Titles to a database called the Occupational Employment Statistics:
We’re looking at the total number of DOT titles … that are in that OES category, and then based upon an understanding
of how those jobs are performed, how they exist, there’s a weighting that [is] done to those categories and from that we get an estimate of the amount of jobs in the specific categories.
Brace’s lawyer then asked the expert to explain his methodology for this “weighting” process:
Well, it’s—it’s that combination of, one, you are looking at the number of titles that are in that category[,] and then based upon the—my information that I have as far as how the frequency of those jobs are performed, then we do an allocation based upon weighting or reweighting those allocations to get the estimates of the numbers. So you look at that particular job title and how it is weighted to the total number of jobs that [are] in that OES category to make an estimate. And all these are estimates because … there is no specific calculation.
Brace’s attorney objected to this testimony. Citing Alaura v. Colvin, 797 F.3d 503 (7th Cir. 2015), he argued that the VE’s job-number estimate lacked sufficient foundation and methodological rationality. In Alaura we criticized the use of the so-called “equal distribution method,” which estimates jobs in one job title by assuming that all job titles in a large category share the category’s number of jobs equally. Id. at 507–08. We explained that this approach “assum[es] that each narrow category has the same number of jobs as each other narrow category—which is preposterous.” Id. at 508. We expressed concern that a job-number estimate based on this method is likely to be nothing more than “a fabrication.” Id.
The ALJ conceded that Brace’s objection had “some merit.” But he nonetheless accepted the VE’s testimony as “satisfactory,” reasoning that an estimate of this size—140,000 jobs—would leave a significant number of jobs available to Brace even if the VE’s approach had a considerable margin of error. Accordingly, the ALJ ruled against Brace at step five and rejected his application for benefits.
Brace sought review in district court. Among other claims of error, he challenged the VE’s job-number estimate as unreliable. The judge disagreed, concluding that the ALJ adequately “investigated” the foundation for the VE’s opinion and explained why he accepted it. Brace appealed, reprising his argument about the unreliability of the VE’s job-number estimate.1
II. Discussion
An ALJ’s factual findings in a disability proceeding are conclusive if supported by “substantial evidence.”
As applied to an expert’s estimate of available jobs in the national economy, “the substantial evidence standard requires the ALJ to ensure that the approximation is the product of a reliable method.” Chavez, 895 F.3d at 968. This does not mean that the VE’s opinion must satisfy the standard for admission of expert
Brace’s attorney asked the VE to explain the method he used to arrive at his job-number estimate of 140,000. It’s worth repeating the key part of his response:
Well, it’s—it’s that combination of, one, you are looking at the number of titles that are in that [OES] category[,] and then based upon the—my information that I have as far as how the frequency of those jobs are performed, then we do an allocation based upon weighting or reweighting those allocations to get the estimates of the numbers.
This answer is entirely unilluminating. Testimony that incants unelaborated words and phrases such as “weighting” and “allocation” and “my information that I have” cannot possibly satisfy the substantial-evidence standard. What allocations? How is the weighting and re-weighting performed? According to what criteria? And what is the unidentified “information” in the expert’s possession?
The Supreme Court’s recent decision in Biestek explained that a vocational expert’s job-number testimony will survive review under the substantial-evidence standard as long as it rests on a well-accepted methodology and the expert describes the methodology “cogently and thoroughly.” 139 S. Ct. at 1155. The Court gave this illustrative example:
Suppose an expert … testifies about the approximate number of various sedentary jobs an applicant for benefits could perform. She explains that she arrived at her figures by surveying a range of representative employers; amassing specific information about their labor needs and employment of people with disabilities; and extrapolating those findings to the national economy by means of a well-accepted methodology. She answers cogently and thor-
oughly all questions put to her by the ALJ and the applicant’s lawyer. And nothing in the rest of the record conflicts with anything she says. … [H]er testimony would be the kind of evidence—far “more than a mere scintilla”—that “a reasonable mind might accept as adequate to support” a finding about job availability.
Id. (citation omitted).
None of these indicia of reliability are present here. The VE’s jargon about his weighting methodology was neither cogent nor thorough—indeed, it was unintelligible. And he never claimed that his method for estimating job numbers is a well-accepted one, much less explained why that is so.
The parties disagree on whether the expert impermissibly relied on the “equal distribution method,” the flawed job-estimate methodology at issue in both Alaura and Chavez. In Brace’s view the VE implicitly used this method because he apparently divided the number of jobs from a job group in the Occupational Employment Statistics by the number of categories in that group as reflected in the Dictionary of Occupational Titles. The Commissioner responds
The very existence of this debate confirms our conclusion that the VE’s testimony does not satisfy the substantial-evidence standard. See Chavez, 895 F.3d at 970. Unlike the equal-distribution method, which at least has a defined meaning, no one knows what the VE meant by “weighting.” This strikes us as the agency taking a “trust me” approach
rather than—as required by the statute and regulations—carrying its burden to demonstrate that significant employment exists in the national economy for a person with this claimant’s restrictions.
The ALJ gave two reasons for accepting the expert’s methodology but neither persuades us. First, the ALJ reasoned that Brace’s counsel should have objected to the expert’s qualifications before he testified. But a claimant need not object to an expert’s qualifications in order to object to the expert’s methodology. Chavez, 895 F.3d at 964. In Chavez the parties stipulated to the expert’s qualifications, id., but the claimant nonetheless successfully challenged the expert’s methodology, id. at 971.
Second, the ALJ reasoned that the jobs number cited by the expert was so large that “[e]ven if the methodology used create[d] a significant margin of error[,] … a significant number of jobs exist that can be performed by the claimant.” But “[e]vidence is not ‘substantial’ if vital testimony has been conjured out of whole cloth.” Donahue, 279 F.3d at 446. An unreliable job-number estimate cannot be considered reliable merely because it is large. Chavez, 895 F.3d at 970.
As in Chavez, “it may be that the evidentiary gap [can be] filled through expanded testimony from the VE about his estimates or through some other showing that there are a significant number of jobs in the economy” available to Brace given his limitations. Id. Accordingly, a remand for a new hearing is necessary to determine his entitlement to
benefits. We VACATE and REMAND for further proceedings consistent with this opinion.