In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 22-2831
JENNIFER L. HOHMAN,
Plaintiff-Appellant,
v.
KILOLO KIJAKAZI, Acting Commissioner of Social Security,
Defendant-Appellee.
____________________
Appeal from the United States District Court for the
Western District of Wisconsin.
No. 3:21-cv-59 — James D. Peterson, Chief Judge.
____________________
ARGUED MAY 24, 2023 — DECIDED JUNE 28, 2023
____________________
Before SCUDDER, ST. EVE, and KIRSCH, Circuit Judges.
SCUDDER, Circuit Judge. Jennifer Hohman applied for So-
cial Security benefits after she stopped working in April 2018.
She suffers from fibromyalgia, post-traumatic stress disorder,
depression, and anxiety, and she alleged that the combined
effects of these conditions left her unable to work. Relying on
testimony from a vocational expert, an administrative law
judge denied the claim for benefits, finding that despite her
limitations, Hohman still had the capacity to perform certain
2 No. 22-2831
jobs available in significant numbers in today’s economy. Be-
cause the record contains substantial evidence supporting the
ALJ’s determination, we affirm.
I
Since 1993 Hohman has regularly visited doctors and
taken prescription medications to treat her anxiety and de-
pression. She points to these impairments as the primary rea-
son she can no longer work as a medical records clerk or pa-
tient access representative. She says that stress, which she fre-
quently experiences upon interacting with others at work,
triggers disabling levels of anxiety and depression not fully
treatable with medication.
In 2013, when Hohman was 47, her doctors diagnosed her
with fibromyalgia, a complex medical condition character-
ized by persistent, widespread pain. Hohman acknowledges
that the treatment she has received for her fibromyalgia has
generally proven effective, but she says that social stress and
anxiety sometimes cause the pain to flare up, leaving her un-
able to work.
Hohman applied for disability benefits in May 2019. Fol-
lowing an evidentiary hearing in August 2020, the ALJ ap-
plied the five-step analysis outlined in 20 C.F.R. § 416.920 and
concluded that Hohman was not disabled. Two aspects of the
ALJ’s analysis feature in this appeal: the determination of
Hohman’s residual functional capacity or RFC, and the step-
five estimate of the number of jobs available that Hohman can
perform.
The ALJ determined Hohman’s RFC after considering the
entire medical record, Hohman’s daily activities, and her
hearing testimony. In the ALJ’s view, Hohman could perform
No. 22-2831 3
light work, which the Social Security Act’s implementing reg-
ulations define as work that does not require lifting more than
20 pounds but does require walking and standing. See 20
C.F.R. § 404.1567(b). The ALJ imposed additional physical
limitations (including occasional stooping and kneeling) and
social limitations (performing tasks independently and only
occasionally interacting with supervisors, coworkers, and the
public). Based on this RFC, the ALJ found that Hohman could
not perform her past work as a medical records clerk or pa-
tient access representative.
Proceeding to step five of the analysis, the ALJ considered
whether Hohman’s RFC allowed her to perform other jobs
and whether those jobs were significantly available in the na-
tional economy. On this point, and as is common practice, the
ALJ relied on the testimony of a vocational expert. The VE tes-
tified that someone with Hohman’s RFC could work as a pho-
tocopy machine operator, small products assembler, or mail
clerk. The VE then stated that these jobs exist in significant
numbers nationwide, with an estimated 50,000 photocopy
machine operator jobs, 40,000 small products assembler jobs,
and 40,000 mail clerk jobs.
During cross-examination by Hohman’s attorney, the VE
elaborated on his methodology to arrive at these estimates.
The VE first explained that he relied upon information pro-
vided by the Bureau of Labor Statistics in its Occupational
Employment Statistics database, a source frequently used as
a starting point by VEs in their application of the equal distri-
bution method to estimate jobs in the national economy. We
explained that method—and its shortcomings—at length in
Chavez v. Berryhill, 895 F.3d 962, 969 (7th Cir. 2018). The VE
here drew upon that source and from there modified the
4 No. 22-2831
results of the equal distribution approach to estimate the job
numbers Hohman could work.
Even more specifically, the VE stated that he arrived at his
final numbers through a “weighted estimate based upon [ ]
professional experience, which contains over 25 years of both
placement and labor market survey work, [including] 2,000
labor market surveys.” The VE explained that many of those
job titles in the database had been merged into one role or had
been eliminated by technology, so he decided that each of the
job titles should not receive equal weight. Within the category
of jobs that includes mail clerks, for example, the VE ex-
plained that the database contained 14 total job titles that add
up to 85,800 total jobs. But the number produced by dividing
85,800 by 14—the equal distribution method—did not accu-
rately represent the number of mail clerk jobs in the national
economy. Drawing on his familiarity with jobs in the mailing
industry, the VE determined that mail clerks instead make up
a larger share of the jobs in that group. So he arrived at an
estimate of 40,000 mail clerk jobs. The VE also testified that he
regularly consulted his colleagues to ensure his views accu-
rately reflected labor market trends.
Hohman’s attorney referenced our 2018 decision in Chavez
and objected to the VE’s testimony as too vague. The ALJ then
asked the VE an additional question about his methodology,
and the VE provided further information about the content of
his labor-market surveys. Finding the answers satisfactory,
the ALJ accepted the VE’s testimony over Hohman’s objec-
tion. The ALJ stated that the VE’s expertise and reasoned ex-
planations provided a sufficient basis to conclude the testi-
mony was reliable, even though he did not provide an exact
formula for his estimates.
No. 22-2831 5
The district court affirmed, and Hohman now seeks our
review.
II
Hohman first contends that the ALJ, in finding she could
perform light work, failed to credit her testimony of pain and
limitations she experiences from her fibromyalgia. Hohman
believes the ALJ would have limited her to sedentary work
had he credited her testimony.
The question before us is not whether Hohman suffers
from pain related to fibromyalgia. Everyone agrees she does.
Our task is limited to whether the ALJ sufficiently canvassed
the record and appropriately weighed the evidence to deter-
mine the conditions, if any, that would allow Hohman to
work given her illness. Our review, in short, is for substantial
evidence—we reverse the ALJ’s decision “only if the record
compels a contrary result.” Deborah M. v. Saul, 994 F.3d 785,
788 (7th Cir. 2021) (quoting Borovsky v. Holder, 612 F.3d 917,
921 (7th Cir. 2010)). And we review credibility determinations
with even more deference—we reverse only those determina-
tions that are “patently wrong.” Id. at 789 (quoting Curvin v.
Colvin, 778 F.3d 645, 651 (7th Cir. 2015)).
We have previously observed that disability claims rooted
in the effects of fibromyalgia warrant careful attention given
the nature of the illness. See Gerstner v. Berryhill, 879 F.3d 257,
264 (7th Cir. 2018) (overturning an ALJ’s credibility determi-
nation where “the ALJ’s analysis reveal[ed] that he misunder-
stood the nature of [the claimant’s] fibromyalgia pain”). The
Social Security Administration also recognizes the complexity
of fibromyalgia and the importance of subjective evidence in
assessing a claimant’s ability to work. See SSR 12-2P, 2012 WL
6 No. 22-2831
3104869. For example, when objective medical evidence con-
flicts with a claimant’s description of the intensity or persis-
tence of pain, an ALJ must consider all evidence in the rec-
ord—such as the claimant’s daily activities, medications,
course of treatment, and statements by other people about the
claimant’s symptoms—before determining the credibility of a
claimant’s self-reported pain. See SSR 12-2P(IV)(B). ALJs
must also take care to examine the claimant’s treatment over
time because fibromyalgia pain waxes and wanes. See SSR 12-
2P(VI)(D) (explaining that ALJs must “consider a longitudinal
record whenever possible” to assess fibromyalgia pain).
The ALJ applied the guidance in SSR 12-2P to arrive at the
light-work RFC for Hohman. Rather than relying solely on
objective medical evidence, the ALJ reviewed the entire rec-
ord for information about Hohman’s pain and how she has
managed it over time. The ALJ pointed to evidence that since
her diagnosis in 2013, Hohman has regularly walked her dog
and performed other household activities without difficulty.
She also continued working for five years until she quit her
job in 2018. The ALJ further observed that Hohman’s course
of treatment has been conservative. For her part, Hohman tes-
tified that her medications effectively alleviated her pain
symptoms, and the record lacked evidence that she had
sought more aggressive treatment. Weighing these pieces of
evidence against Hohman’s subjective testimony, the ALJ
found that Hohman could perform light work with certain so-
cial and physical limitations.
Hohman sees the ALJ’s light-work RFC determination as
“patently wrong.” Deborah M., 994 F.3d at 789. But we cannot
agree: Hohman’s testimony supports, and does not contra-
dict, the ALJ’s determination. Her emphasis on depression
No. 22-2831 7
and anxiety as the primary reasons she cannot work, and as
the root cause of her fibromyalgia-related pain, underscores
the ALJ’s reliance on the record. By imposing social limita-
tions—limiting her to jobs that do not require regular contact
with supervisors or the public—the ALJ faithfully adopted
Hohman’s testimony to reflect her realistic capacity to work
despite her impairments. This is substantial evidence sup-
porting the ALJ’s RFC determination, and Hohman cannot
point to evidence that compels a different result.
III
A
Hohman next challenges the ALJ’s reliance on the VE’s
testimony that there are a significant number of jobs available
for her to work given her RFC.
In reviewing a VE’s testimony, we ask whether substantial
evidence supports the ALJ’s reliance on that testimony, in-
cluding whether the VE used a reliable methodology. See
Chavez, 895 F.3d at 968. We consider “all features of the [VE’s]
testimony” to determine whether the testimony establishes
“‘more than a mere scintilla’ of evidence supporting the ALJ’s
conclusion,” keeping in mind that our determination is made
“case-by-case.” Biestek v. Berryhill, 139 S. Ct. 1148, 1156–57
(2019).
A methodology is reliable when it is based on “well-ac-
cepted” sources and its steps are “cogently and thoroughly”
explained. Id. at 1155. Establishing the reliability of a VE’s job-
number estimate “does not require meeting an overly exact-
ing standard,” as the “law recognizes and respects” the “real-
ities and limitations” of calculating the number of jobs avail-
able for a claimant to work. Chavez, 895 F.3d at 968. The VE’s
8 No. 22-2831
testimony need only “instill some confidence that the estimate
was not ‘conjured out of whole cloth.’” Brace v. Saul, 970 F.3d
818, 822 (7th Cir. 2020) (quoting Donahue v. Barnhart, 279 F.3d
441, 446 (7th Cir. 2002)).
B
The ALJ found the VE’s testimony to be supported by sub-
stantial evidence. So do we. The VE’s approach aligns with
the Supreme Court’s guidance in Biestek and our observations
in Chavez. We see no reason to disturb the ALJ’s job-number
findings.
We can readily trace the path of reasoning the VE used to
arrive at the estimate of 40,000 mail clerk jobs. Instead of rely-
ing on the equal distribution method—a method whose short-
comings are well known—the VE made new estimates based
on his “knowledge of labor market conditions and occupa-
tional trends, gleaned from reviewing relevant data sources
[and] placing workers in jobs” over the course of his 25-year
career. Chavez, 895 F.3d at 970. Just as we have encouraged,
the VE here brought “his extensive experience to bear on the
reality of those numbers,” rather than “mechanically rel[y] on
outdated sources” like the Bureau of Labor Statistics data-
base, which was last updated in 1991. Id. In short, we have
little difficulty following the VE’s process, and we have no
reason to question the reasonableness of the job numbers he
approximated.
Hohman urges a different approach. In her view, the VE
failed to show all his work, so his estimates cannot be sup-
ported by substantial evidence. On appeal, Hohman identi-
fies specific questions that the VE left unanswered. She is un-
sure about which jobs the VE believed had been eliminated
No. 22-2831 9
by technology, for instance, and she does not precisely know
the step-by-step calculations the VE performed to arrive at his
final job-number estimates. Hohman effectively challenges
the VE’s modified application of the equal distribution
method as too vague because she cannot recreate the job-
number estimates herself.
But a VE’s failure to provide exact data or calculations
does not itself make the testimony unreliable. See Biestek, 139
S. Ct. at 1157. A VE’s job-number estimate is “just that—an
estimate.” Chavez, 895 F.3d at 968. VEs, we have emphasized,
“are neither required nor expected to administer their own
surveys of employers to obtain a precise count of the number
of positions that exist.” Id. The VE here did not err in explain-
ing his methodology in more general, non-precise terms.
Compare the VE’s testimony in Hohman’s case to that of
the VEs in two cases we see as bookends on the substantial-
evidence standard. At one end is Fetting v. Kijakazi, where a
VE testified that his labor-market expertise from 30 years of
experience in job placement informed the adjustments he
made to numbers derived from the same Bureau of Labor Sta-
tistics database the VE used here. 62 F.4th 332 (7th Cir. 2019).
We found those estimates supported by substantial evidence.
See id. at 339.
On the other end is Brace v. Saul, where a VE testified only
that he made an “allocation based upon weighting or re-
weighting” to arrive at his estimates. 970 F.3d at 822. We held
this testimony lacked evidentiary support because that VE
disclaimed reliance on commonly used software and meth-
ods, and then failed to explain how his expertise informed his
alternative “weighting” and “allocation” method. Id. We ex-
plained that a VE’s experience and qualifications alone cannot
10 No. 22-2831
establish substantial evidence when he does not explain how
that experience informs his methodology. See id.
The VE’s testimony here is far closer to Fetting than it is to
Brace. No doubt the VE could have explained his methodol-
ogy with more clarity. But the VE “gave enough detail for us
to understand the sources of his data and the general process
he adopted.” Fetting, 62 F.4th at 339. That is all the substantial
evidence standard requires.
Hohman goes a bridge too far in suggesting that a VE must
provide exact calculations to establish this modicum of confi-
dence. We have never required VEs to meet such a standard,
and we decline to do so today. A VE who, as here, provides a
reasoned explanation based on hands-on experience working
in the field, market surveys, or conversations with employers,
can establish sufficient confidence in his job-number esti-
mates even though the claimant may not be able to exactly
duplicate them. See Chavez, 895 F.3d at 968 (recognizing that
the “agency’s regulations do not mandate a precise count of
job numbers”). The law does not require the evidentiary pre-
cision Hohman seeks.
Notice what we are not saying. We are not saying that a
VE’s testimony passes the substantial-evidence threshold
based on a VE’s credentials or expertise alone. A federal
court’s review of an ALJ’s decision “is case-by-case,” and
“takes into account all features of the vocational expert’s tes-
timony, as well as the rest of the administrative record.”
Biestek, 139 S. Ct. at 1157. We see this holistic review reflected
in the case law. Many VEs may testify to some type of
“weighting” approach to their job-number estimates. In cases
where a VE merely utters (or repeats) the word “weighting”
without further explanation, such a reference may doom the
No. 22-2831 11
reliability of their testimony. See, e.g., Brace, 970 F.3d at 822.
But where a VE reasonably connects their “weighting” to their
knowledge and experience, such as here and in Fetting, an ALJ
can more readily see the logical bridge underlying the esti-
mate.
A broader observation warrants emphasis. We have little
doubt the source of much of Hohman’s frustration is the equal
distribution method. While we share many concerns about
the method’s reliability, we have never categorically barred
VEs from using that method in social security disability hear-
ings. It is not, by itself, reversible error for an ALJ to rely on
the equal distribution method—in its purest form or modified
to reflect a VE’s experience—to make a step-five job-number
determination.
* * *
The party best positioned to devise a reliable approach to
estimating job numbers at step five is the one that first prom-
ised to do so in 2008—the Commissioner of Social Security.
See Chavez, 895 F.3d at 965. Fifteen years later, the agency still
has not put a new system in place. What makes this problem
frustrating for judges is that it is easy to imagine in today’s
data-driven world, detailed information about the national la-
bor market exists and can feasibly be collected and used by
ALJs in social security hearings. Until the agency follows
through with its promise, ALJs will continue to rely on VE
job-number estimates like the one that Jennifer Hohman chal-
lenges here to determine whether a claimant should receive,
or be denied, disability benefits. We—and social security
claimants nationwide—eagerly await an improved system.
With these closing observations, we AFFIRM.