Opinion
2:22-cv-00691-SB
05-09-2023
CATHERINE L.,[1] Plaintiff, v. KILOLO KIJAKAZI, Acting Commissioner of Social Security, Defendant.
OPINION AND ORDER
HON. STACIE F. BECKERMAN, UNITED STATES MAGISTRATE JUDGE.
Catherine L. (“Plaintiff”) brings this appeal challenging the Commissioner of Social Security's (“Commissioner”) partial denial of her applications for Disability Insurance Benefits (“DIB”) and Supplemental Security Income (“SSI”) under Titles II and XVI of the Social Security Act. The Court has jurisdiction pursuant to 42 U.S.C. §§ 405(g) and 1383(c)(3), and the parties have consented to the jurisdiction of a magistrate judge pursuant to 28 U.S.C. § 636(c). For the reasons explained below, the Court reverses the Commissioner's decision because it is based on harmful legal error and not supported by substantial evidence in the record.
Plaintiff filed her applications on May 8, 2019, and had a date last insured of December 31, 2017, and the Commissioner found that Plaintiff was not disabled before October 23, 2020, but that she became disabled on that date. (Tr. 34-35.) Thus, the Commissioner effectively denied Plaintiff's DIB application and granted in part Plaintiff's SSI application. See Truelsen v. Comm'r Soc. Sec., No. 15-2386, 2016 WL 4494471, at *1 n.4 (E.D. Cal. Aug. 26, 2016) (“To be entitled to DIB, [a] plaintiff must establish that he was disabled . . . on or before his date last insured.” (citing Tidwell v. Apfel, 161 F.3d 599, 601 (9th Cir. 1999))); Schiller v. Colvin, No. 12-771-AA, 2013 WL 3874044, at *1 n.1 (D. Or. July 23, 2013) (“[T]he earliest an SSI claimant can obtain benefits is the month after which he filed his application[.]”) (citation omitted).
STANDARD OF REVIEW
The district court may set aside a denial of benefits only if the Commissioner's findings are “not supported by substantial evidence or based on legal error.” Bray v. Comm'r of Soc. Sec. Admin., 554 F.3d 1219, 1222 (9th Cir. 2009) (quoting Robbins v. Soc. Sec. Admin., 466 F.3d 880, 882 (9th Cir. 2006)). Substantial evidence is defined as “more than a mere scintilla [of evidence] but less than a preponderance; it is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Id. (quoting Andrews v. Shalala, 53 F.3d 1035, 1039 (9th Cir. 1995)).
The district court “cannot affirm the Commissioner's decision ‘simply by isolating a specific quantum of supporting evidence.'” Holohan v. Massanari, 246 F.3d 1195, 1201 (9th Cir. 2001) (quoting Tackett v. Apfel, 180 F.3d 1094, 1098 (9th Cir. 1999)). Instead, the district court must consider the entire record, weighing the evidence that both supports and detracts from the Commissioner's conclusions. Id. Where the record as a whole can support either the grant or denial of Social Security benefits, the district court “may not substitute [its] judgment for the [Commissioner's].” Bray, 554 F.3d at 1222 (quoting Massachi v. Astrue, 486 F.3d 1149, 1152 (9th Cir. 2007)).
BACKGROUND
I. PLAINTIFF'S APPLICATIONS
Plaintiff was born in April 1971, making her forty-four years old on December 19, 2015, her alleged disability onset date. (Tr. 243, 279.) Plaintiff is a high school graduate who has completed some college coursework and past relevant work experience as a housecleaner or “day worker.” (Id. at 46-47, 63, 78, 450.) In her applications, Plaintiff alleges disability due to deep vein thrombosis (“DVT”), diabetes, peripheral neuropathy, degenerative disc disease, “[c]hronic arthritis,” osteoarthritis, panic disorder, agoraphobia, lower back pain, sciatica, depression, anxiety, posttraumatic stress disorder (“PTSD”), high blood pressure, and memory loss. (Id. at 244, 280.)
The Commissioner denied Plaintiff's applications initially and upon reconsideration, and on August 5, 2020, Plaintiff requested a hearing before an Administrative Law Judge (“ALJ”). (Id. at 34.) Plaintiff and a vocational expert (“VE”) appeared and testified at an administrative hearing held on April 19, 2021. (Id. at 60-89.) On May 13, 2021, the ALJ issued a written decision partially denying Plaintiff's applications. (Id. at 34-49.) On March 15, 2022, the Appeals Council denied Plaintiff's request for review, making the ALJ's written decision the final decision of the Commissioner. (Id. at 1-6.) Plaintiff now seeks judicial review of that decision.
II. THE SEQUENTIAL PROCESS
A claimant is considered disabled if he or she is unable to “engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which . . . has lasted or can be expected to last for a continuous period of not less than 12 months[.]” 42 U.S.C. § 423(d)(1)(A). “Social Security Regulations set out a five-step sequential process for determining whether an applicant is disabled within the meaning of the Social Security Act.” Keyser v. Comm'r Soc. Sec. Admin., 648 F.3d 721, 724 (9th Cir. 2011). Those five steps are: (1) whether the claimant is engaged in substantial gainful activity; (2) whether the claimant has a severe impairment; (3) whether the impairment meets or equals a listed impairment; (4) whether the claimant can return to any past relevant work; and (5) whether the claimant can perform other work that exists in significant numbers in the national economy. Id.at 724-25.
The claimant bears the burden of proof for the first four steps. Se eBustamante v. Massanari, 262 F.3d 949, 953-54 (9th Cir. 2001). If the claimant fails to meet the burden at any of those steps, the claimant is not disabled. Seeid. at 954. The Commissioner bears the burden of proof at step five, where the Commissioner must show the claimant can perform other work that exists in significant numbers in the national economy, “taking into consideration the claimant's residual functional capacity, age, education, and work experience.” Tackett, 180 F.3d at 1100. If the Commissioner fails to meet this burden, the claimant is disabled. SeeBustamante, 262 F.3d at 954.
III. THE ALJ'S DECISION
The ALJ applied the five-step sequential evaluation process to determine if Plaintiff is disabled. (Tr. 34-49.) At step one, the ALJ determined that Plaintiff had not engaged in substantial gainful activity since December 19, 2015, her alleged disability onset date. (Id. at 36.) At step two, the ALJ determined that Plaintiff suffered from the following severe impairments: “[O]besity, diabetes mellitus, lumbar degenerative disc disease, right knee arthritis, right femoral [DVT], bilateral hand numbness, Meniere's disease, depression, anxiety, and [PTSD].” (Id. at 36-37.)
At step three, the ALJ concluded that Plaintiff did not have an impairment that meets or medically equals a listed impairment. (Id. at 37.) The ALJ then concluded that Plaintiff had the residual functional capacity (“RFC”) to perform sedentary work, subject to these limitations: (1) Plaintiff can stand and/or walk for two hours and sit for six hours during an eight-hour workday, (2) Plaintiff can frequently handle and finger bilaterally, occasionally balance and stoop, and “rarely (no more than 15 percent of the time) climb ramps or stairs,” (3) Plaintiff cannot crouch, kneel, crawl, operate foot controls, “work at a production-rate, conveyor-belt-type pace,” or climb ladders, ropes, or scaffolds, (4) Plaintiff needs to “avoid all exposure to moving or dangerous machinery and unprotected heights” and “a predictable work environment with no more than occasional, simple workplace changes,” and (5) Plaintiff can have occasional contact with supervisors and “tolerate brief, superficial contact with the public and co-workers[.]” (Id. at 40.)
At step four, the ALJ determined that Plaintiff was unable to perform her past relevant work as a housecleaner or “day worker.” (Id. at 46-47.) At step five, the ALJ determined that Plaintiff was not disabled before October 23, 2020, the date her age category changed to an individual closely approaching advanced age, because a significant number of jobs existed in the national economy that she could perform, including work as a final assembler, bench hand, and semiconductor bonder. (Id. at 47-48.) However, given the change in Plaintiff's age category and the RFC limitation to less than a full range of sedentary work, the ALJ found that Rule 201.14 of the Medical-Vocational Guidelines dictated a finding of disabled as of October 23, 2020. (Id. at 48.)
DISCUSSION
In this appeal, Plaintiff argues that the ALJ erred in, among other ways, concluding that Plaintiff could perform other jobs that exist in significant numbers in the national economy. (Pl.'s Opening Br. at 2-6.) Plaintiff's first and primary assignment of error concerns whether the Commissioner failed to carry her step-five burden. (See id.) As explained below, the Court concludes that the Commissioner did not carry her burden on this potentially dispositive issue, and therefore reverses the Commissioner's decision and remands for further administrative proceedings.
Plaintiff's opening brief is styled as a motion for summary judgment. (See Pl.'s Mot. Summ. J. at 1, ECF No. 14.) CM/ECF did not treat this brief as a “motion,” see generally Atkins v. Stivers, No. 21-5798, 2021 WL 7084872, at *1 (6th Cir. Dec. 16, 2021) (“‘CM/ECF' stands for Case Management/Electronic Case Files and is the federal judiciary's case management and electronic filing system.”), and the local rules require only an “opening brief” to trigger the judicial review process. See LR 4000 (addressing the deadlines and rules related to the briefing in a Social Security Administration appeal, and the “opening brief” necessary to present an action for judicial review). Consistent with these facts, the Court refers to Plaintiff's motion as an “opening brief” and cites to the motion as “Pl.'s Opening Br.” Court document citations refer to the CM/ECF-generated page numbers.
I. APPLICABLE LAW
When, as here, “a claimant establishes at steps one through four of the sequential evaluation process that she suffers from a severe impairment that prevents her from doing any work she has done in the past, . . . she has made out a prima facie case of a disability.” White v.Kijakazi, 44 F.4th 828, 833 (9th Cir. 2022) (citing Tackett, 180 F.3d at 1098). Given that Plaintiff made out such a case, the Commissioner bore the step-five burden of demonstrating that Plaintiff “can perform a ‘significant number[]' of jobs in the national economy given [her] . . . physical and mental limitations, age, education, and work experience.” White, 44 F.4th at 833 (quoting 20 C.F.R. § 416.960(c)(2) and citing Thomas v. Barnhart, 278 F.3d 947, 955 (9th Cir. 2002)).
At step five, the Commissioner can carry her burden by “rely[ing] on a VE's testimony to support a finding that the claimant can perform occupations that exist in significant numbers in the national economy.” Kilpatrick v. Kijakazi, 35 F.4th 1187, 1192 (9th Cir. 2022) (citing Ford v.Saul, 950 F.3d 1141, 1149 (9th Cir. 2020) and Tackett, 180 F.3d at 1100). “VEs may use a wide range of data sources and methodologies to generate job-number estimates.” White, 44 F.4th at 834 (citing, inter alia, Biestek v. Berryhill, 139 S.Ct. 1148, 1152-53 (2019)). A claimant may cross-examine a VE “during the proceedings about the data sources or methodologies underlying her job-number estimates, [but] there is no requirement that a VE disclose the primary data underlying the estimates prior to, during, or after the hearing.” Id. (citations omitted). In fact, “the Supreme Court recently held that VE job-numbers testimony may constitute substantial evidence supporting a finding of no disability even if the VE refuses to disclose the specific, non-public data sources on which the estimates were based.” Id. at 835 (citing Biestek, 139 S.Ct. at 1159).
The Ninth Circuit has recognized that despite its “inherent[] reliab[ility],” a VE's jobnumbers testimony is “not incontestable.” Kilpatrick, 35 F.4th at 1193 (quoting Buck v. Berryhill, 869 F.3d 1040, 1051 (9th Cir. 2017)). There may be cases where a VE “offer[s] testimony that is so feeble, or contradicted, that it would fail to clear the substantial-evidence bar.” White, 44 F.4th at 835 (quoting Biestek, 139 S.Ct. at 1155-56). A court's “substantial evidence inquiry . . . must proceed on a ‘case-by-case' basis, taking ‘into account all features of [the VE's] testimony, as well as the rest of the administrative record.'” Id. (quoting Biestek, 139 S.Ct. at 1157).
II. ADMINISTRATIVE PROCEEDINGS
The ALJ denied Plaintiff benefits based on the VE's testimony that there were an estimated 30,000 semiconductor bonder, 24,000 final assembler, and 22,000 bench hand jobs available in the national economy. (Tr. 48, 79-80, 82-84.) During the telephonic hearing before the ALJ, Plaintiff's counsel asked the VE about the data sources he used to generate job number estimates. (Id. at 86-87.) The VE testified that he used: (1) “SkillTRAN,” (2) “resources developed by professionals in [his] industry including job analyses and labor market surveys,” (3) “Occupational Requirement Statistics information,” and (4) data from the “Bureau of Labor Statistics,” which the VE stated “is SkillTRAN also[.]” (Id.) The VE explained that he “make[s] estimates from all [of this] information,” but “most of [it] is . . . contained in the SkillTRAN.” (Id. at 87.)
The Court notes that the VE identified two data sources that were “inaudible” to the court reporter. (Tr. 87) (caps omitted). The Court also notes that SkillTRAN is computer software that includes a program called “Job Browser Pro.” See Purdy v. Berryhill, 887 F.3d 7, 14-15 (1st Cir. 2018) (discussing “SkillTRAN's software”); Perkins v. Kijakazi, No. 21-35730, 2022 WL 3585587, at *1 (9th Cir. Aug. 22, 2022) (describing job-number estimates from “a program called ‘Job Browser Pro'”). Job Browser Pro is “widely used by VEs [because it] collects and analyzes Bureau of Labor Statistics data,” and is “a source that has been recognized as a reliable source of job-number estimates.” Rashidi v. Kijakazi, No. 22-55121, 2022 WL 17974638, at *2 & n.2 (9th Cir. Dec. 28, 2022) (citing White, 44 F.4th at 831-33 and Purdy, 887 F.3d at 14-17).
In seeking review of the ALJ's decision, Plaintiff provided the ALJ and Appeals Council with additional evidence related to the VE's job number estimates. (See id. at 1, 4-5, 12, 23-24, 26-29, 399-401, 505-06.) Specifically, Plaintiff submitted evidence from SkillTRAN/Job Browser Pro showing that in 2021 (the year of the hearing), only 87 full-time jobs existed in the national economy in the three occupations the ALJ identified as suitable for a worker with Plaintiff's RFC (19 semiconductor bonder, 27 final assembler, and 41 bench hand jobs), as compared to the VE's testimony that 76,000 total jobs existed. (See Tr. 12, 23-34, 26-29, 48, 399-401, 505-06.)
The Appeals Council incorporated Plaintiff's new job-numbers evidence into the administrative record. (See id. at 1, 4-5, see also 23-24, 26-29, 399-401, 505-06.) The Appeals Council, however, found that Plaintiff's evidence did “not provide a basis for changing the [ALJ's] decision.” (Id. at 1.) The Appeals Council therefore denied Plaintiff's request for review. (Id.)
“A disability claimant has the right to submit evidence to the Appeals Council that was not submitted to the ALJ, provided there is ‘good cause' for not having submitted the evidence earlier.” White, 44 F.4th at 836 (citing 20 C.F.R. § 404.970(b)). Plaintiff submitted her evidence to the ALJ (see Tr. 12-29), but even if Plaintiff had not done so, the Court interprets that Appeals Council's statement that it “considered” and “exhibited” Plaintiff's evidence as indicating that she had demonstrated good cause under § 404.970(b). See White, 44 F.4th at 836-37 (“We interpret the statement in the Appeals Council's Order that it had ‘consider[ed]' White's reasons and had made his additional evidence ‘part of the record' as indicating that White had shown ‘good cause' under § 404.970(b) in not having introduced it in the hearing before the ALJ. We have previously recognized that ‘a claimant will rarely, if ever, be in a position to anticipate the particular occupations a VE might list and the corresponding job numbers to which a VE might testify at a hearing.'” (quoting Shaibi v. Berryhill, 883 F.3d 1102, 1110 (9th Cir. 2017))).
III. ANALYSIS
In disputing whether the Commissioner (and the ALJ) failed to meet the step-five burden of demonstrating that Plaintiff can perform other occupations that exist in significant numbers in the national economy, Plaintiff and the Commissioner focus primarily on the Ninth Circuit's recent decision in White. (See Pl.'s Opening Br. at 2-6; Def.'s Br. at 2-4, ECF No. 16; Pl.'s Reply Br. at 2-3, ECF No. 17.)
In White, the VE's job numbers testimony was based on data from SkillTRAN and the ALJ denied the plaintiff benefits based on the VE's testimony that that were 72,000 table worker, 65,000 assembler, and 32,000 film touch up inspector jobs (169,000 total jobs) in the national economy that the plaintiff could perform. 44 F.4th at 830-32. The plaintiff asked the Appeals Council to review the ALJ's decision and submitted evidence in the form of “screenshots of SkillTRAN Job Browser Pro,” which reflected that “there were only 2,957 table worker jobs, 0 assembler jobs, and 1,333 film touch-up inspector jobs in the national economy” (4,290 total jobs). Id. at 832. The Appeals Council considered and exhibited the evidence but found that it did not provide a basis for changing the ALJ's decision, and thus denied the plaintiff's request for review. Id. at 833.
On appeal, the Ninth Circuit “h[e]ld that remand [was] appropriate to allow the ALJ to address the [plaintiff's] evidence and to resolve the inconsistency between the job-number estimates provided by [the plaintiff] and by the VE.” Id. at 837. In support of its holding, the Ninth Circuit emphasized that the plaintiff's evidence was produced using SkillTRAN, the SSA frequently relies on SkillTRAN, and the estimate discrepancy was comparable to one of its past decisions:
If the Appeals Council had credited White's estimates that there were only a total of 4,290 jobs (counting all three jobs) in the national economy that he could perform, there is a reasonable probability that the outcome of his proceeding may have been different[, as we have previously held that 1,680 jobs nationally was not significant and 25,000 jobs nationally was significant but still a ‘close call.'] . . . Furthermore, White's conflicting evidence was ‘significant' and ‘probative.' . . . White's evidence was produced using [SkillTRAN,] a data source and methodology frequently relied on by the SSA.... Indeed, according to White, his job estimates were produced using the same methodology as that used by the VE.
In sum, White's job estimates differed substantially from those of the VE. White estimated-using SkillTRAN Job Browser Pro and the same DOT codes the VE had used-that there were 2,957 table worker, 0 assembler, and 1,333 film touch-up inspector jobs in the national economy. The VE had testified that there were 72,000, 65,000, and 32,000 jobs, respectively, in these same occupations. The discrepancy between these estimates is comparable to the discrepancy in Buck....
We recognize that the claimant in Buck submitted his estimated job numbers to the ALJ, and that White submitted his estimated job numbers to the Appeals Council. This distinction is not fatal. Under SSA regulations, a claimant has a right to submit additional evidence to the Appeals Council, which the agency may choose to make part of the administrative record where it finds that certain conditions are met, as it appears to have done here.... Applying Buck, we hold that remand is appropriate to allow the ALJ to address the evidence and to resolve the inconsistency between the job-number estimates provided by White and by the VE.Id. (citing Beltran v. Astrue, 700 F.3d 386, 390 (9th Cir. 2012) and quoting Gutierrez v. Comm'r of Soc. Sec., 740 F.3d 519, 528-29 (9th Cir. 2014) and Kilpatrick, 35 F.4th at 1193); see also Buck, 869 F.3d at 1047 (reflecting that the VE estimated that there were 600,000 bottling-line attendant jobs, 8,800 bottle-packer jobs, and 235,000 conveyor-belt maker jobs in the national economy, and the plaintiff claimed that he used the same software as the VE in estimating that there were only 231 bottling-line attendant jobs, 2,039 bottle-packer jobs, and 26 conveyor-belt maker jobs nationally).
Here, Plaintiff submitted her estimated job numbers to the ALJ and Appeals Council, and the discrepancy between Plaintiff's and the VE's estimates (87 total jobs versus 76,000 total jobs) is comparable to the discrepancy in White (4,290 total jobs versus 169,000 total jobs). The Commissioner nevertheless argues that she met her burden at step five. (See Def.'s Br. at 4.) The Court disagrees.
In arguing that substantial evidence supports her step-five findings, the Commissioner emphasizes that the Ninth Circuit has declined to “provide a bright-line rule for what constitutes a significant number of jobs existing in the national economy.” (Def.'s Br. at 2-3, citing Gutierrez, 740 F.3d at 528-29.) In both White and Gutierrez, however, the Ninth Circuit recognized that 1,680 jobs nationally, which is far more than Plaintiff's estimate of 87 jobs nationally, is not significant. See White, 44 F.4th at 837 (recognizing that Beltran “h[e]ld[] that 1,680 jobs nationally was not significant” (citing Beltran, 700 F.3d at 390)); Gutierrez, 740 F.3d at 529 (noting that Beltran reflects that “1,680 nationwide jobs [is] insignificant” (citing Beltran, 700 F.3d at 390)). The Commissioner does not address this fact, nor does she address the comparable discrepancy between the plaintiff's and VE's estimates in White and the case before this Court.
Instead, the Commissioner emphasizes that Plaintiff's estimates are based on “her own lay interpretations of vocational resources, including those not recognized in regulations,” that in some cases, a VE's testimony can constitute substantial evidence even without supporting data, and that unlike White, the VE relied on a “variety of sources,” not just SkillTRAN. (Def.'s Br. at 3-4.) The Court finds unpersuasive the Commissioner's arguments and attempts to distinguish White.
The cases cited herein reflect that VEs often use SkillTRAN because it is a reliable source of job-number estimates, and the Commissioner fails adequately to explain why Plaintiff's “lay interpretations” of SkillTRAN are problematic under these circumstances or materially different than the plaintiff's evidence/interpretation of SkillTRAN in White. See 44 F.4th at 832 (“[White's Appeals Council] evidence gave job estimates generated using SkillTRAN's ‘flagship program,' Job Browser Pro. According to this evidence, there were only 2,957 table worker jobs, 0 assembler jobs, and 1,333 film touch-up inspector jobs in the national economy. White attached screenshots of SkillTRAN Job Browser Pro showing these job numbers.”).
The Commissioner is correct that the VE testimony discussed in White appears to differ in certain respects from the VE's testimony in this case. In White, the VE “testified that the [estimated job] numbers came indirectly from the ‘Bureau of Labor Statistics' and that she used an automated program, ‘SkillTRAN,' to get the actual numbers.” 44 F.4th at 831. By contrast, the VE here testified that he used two sources that were inaudible to the court reporter, SkillTRAN, “resources developed by professionals in [his] industry including job analyses and labor market surveys,” “Occupational Requirement Statistics information,” and “Bureau of Labor Statistics [data],” which “is SkillTRAN also[.]” (Tr. 86-87.) Notably, however, the VE stated that he “make[s] estimates from all [of this] information,” but “most of [it] is contained . . . in the SkillTRAN.” (Id. at 87.)
In assessing whether the VE's testimony “‘clear[s] the substantial-evidence bar[,]' . . . [the Court's] inquiry . . . must proceed on a ‘case-by-case' basis, taking ‘into account all features of [VE's] testimony, as well as the rest of the administrative record.'” White, 44 F.4th at 835 (quoting Biestek, 139 S.Ct. at 1157). The VE testified that SkillTRAN includes “most” of the data underlying his estimate (76,000 jobs nationally), and yet, Plaintiff's SkillTRAN-based estimate differs substantially from the VE's. In fact, Plaintiff estimates that that are only 87 total jobs nationally, which is nearly one-tenth of one percent (0.10%) of the VE's estimate and far less than the 1,680 jobs that Beltran deemed not significant. See Beltran, 700 F.3d at 390 (“Considering all of her limitations, [such as an inability to stand and walk for more than two hours,] it would be unconscionable to expect Beltran to find even one of 135 jobs as a surveillance system monitor in her region or one of 1,680 jobs scattered across several regions.”).
It is difficult to see how any other sources, which cover the remainder (but not “most”) of the data underlying the VE's estimate, could possibly account for a discrepancy of 75,913 jobs nationally, or nearly 99.9% of the VE's estimate. It is also difficult to see how Plaintiff's and the VE's methodologies differed in any material way. Cf.Kilpatrick, 35 F.4th at 1194 (“Unlike in Buck, Kilpatrick's attorney did not replicate the VE's same methodology [and instead used his own]. It is thus not surprising that Kilpatrick's different approach led to different results. And there is no basis to conclude that these results qualified as significant probative evidence .... Kilpatrick's attorney . . . had no identified expertise in calculating job figures in the national economy.”).
The Commissioner attempts to distinguish White based on the VE's testimony about using various estimate sources. However, in addressing the pertinent issues of whether there was a “reasonable probability that the outcome [of] the proceeding may have been different” if the Appeals Council had “credited [the plaintiff's] estimates,” and whether the plaintiff's “conflicting evidence was ‘significant' and ‘probative,'” White focused largely on the reliability of the plaintiff's estimate source (SkillTRAN), the discrepancy between the plaintiff's and VE's estimates and a comparator case, and the total number of jobs that the plaintiff's estimate produced. See 44 F.4th at 836-37 (citing Gutierrez, 740 F.3d at 528-29 and quoting Kilpatrick, 35 F.4th at 1193).
Cases applying White have done the same. For example, in Perkins, the plaintiff provided the Appeals Council (but not the ALJ) with new jobs-number evidence from Job Browser Pro, which demonstrated that “in 2018, only 4,827 jobs existed in the national economy in the three occupations noted by the 2011 VE (2,896 table worker jobs, 111 dowel inspector jobs, and 1,820 polisher jobs), as compared to the 579,000 total jobs to which the VE testified in 2011.” 2022 WL 3585587, at *1. The Appeals Council incorporated the evidence into the record but denied the plaintiff's request for review. Id. On appeal, the Ninth Circuit “agree[d]” with the plaintiff that the “the Appeals Council's cursory statement that [the plaintiff's] newly submitted jobnumbers evidence ‘d[id] not show a reasonable probability that it would change the outcome of the [ALJ's] decision' to deny [the plaintiff] benefits was insufficient under [Ninth Circuit] precedent.” Id. at *2. The Ninth Circuit explained that White, which held that the agency was “obligated” to resolve the inconsistency between the plaintiff's and VE's job number estimates, involved “a nearly identical scenario.” Id. The Ninth Circuit thus concluded that “the same result [was] appropriate” in Perkins, noting that the plaintiff's “new job-number evidence came from the same source as White's evidence-Job Browser Pro-so was equally ‘significant' and ‘probative,'” and that the estimate discrepancy was “even larger” than the one in White. Id.(citations omitted).
Perkins is instructive and supports the Court's determination that a remand for further proceedings is necessary here. SeeId. (“Because the Appeals Council chose to make this new evidence part of the administrative record, . . . the agency was bound to address and resolve the conflict between Perkins' job-numbers estimates and the . . . VE's estimates. It failed to do so, and so we reverse the district court's order affirming the denial and direct that the case be remanded to the agency for further proceedings consistent with this decision.”).
See also Grimm v. City of Portland, 971 F.3d 1060, 1062-67 (9th Cir. 2020) (explaining that unpublished decisions “can be cited, and may prove useful, as examples of the applications of settled legal principles when a district court or litigant is interested in demonstrating how a given principle operates in practice”).
For all of these reasons, the Commissioner failed to carry her step-five burden, and thus the Court reverses the Commissioner's decision. See Est. of Lisa H. v. Comm'r, Soc. Sec. Admin., No. 6:21-cv-01197-MC, 2023 WL 2707454, at *1-3 (D. Or. Mar. 30, 2023) (remanding for further proceedings based on White and the plaintiff's Job Browser Pro-based estimates and noting that the plaintiff “did not replicate the VE's exact methodology” but the “data sources significantly overlapped”). The Court offers no opinion as to whether Plaintiff's estimate is more or less than reliable than the VE's estimate, and notes that on remand, the parties should be permitted to submit additional evidence regarding the number of jobs available that Plaintiff could perform. See Perkins, 2022 WL 3585587, at *2 & n.3 (permitting the parties to submit additional evidence as to the number of jobs available and noting that “for the purposes of remand, [the Ninth Circuit] offer[ed] no opinion about whether Perkins' estimates are more or less reliable than the VE's earlier estimates; [the Ninth Circuit] h[e]ld only that the discrepancy between them needed to be directly addressed by the agency once it made Perkins' evidence part of the record”).
Given the need for further proceedings on an issue that is potentially dispositive with respect to the contested period of disability, the Court declines to address Plaintiff's remaining assignments of error. To the extent necessary, Plaintiff and the agency can address these issues on remand.
CONCLUSION
For the reasons stated, the Court REVERSES the Commissioner's decision and REMANDS for further proceedings consistent with this opinion.
IT IS SO ORDERED.