Bubb, Timothy v. Saul, Andrew

IN THE UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF WISCONSIN TIMOTHY BUBB, Plaintiff, OPINION AND ORDER v. 20-cv-101-wmc KILOLO KIJAKAZI, ACTING COMMISSIONER OF SOCIAL SECURITY, Defendant. On January 17, 2019, Administrative Law Judge (“ALJ”) Trina Mengesha-Brown issued a decision denying plaintiff Timothy Bubb’s claim for supplemental security income. Bubb appealed this decision, arguing that the ALJ erred in two ways: (1) by failing to ensure that the vocational expert’s opinion testimony was based on a reliable method; and (2) by failing to justify the adopting of an off-task limitation of no more than 10%. For the reasons discussed below, the court agrees that the ALJ erred and will remand this case for rehearing.1 BACKGROUND On May 16, 2016, plaintiff Timothy Bubb filed an application for supplemental security income, alleging disability due to a variety of physical and mental limitations. In particular, Bubb claimed he suffers from: “Asperger’s, Autism, ADHD, ODD, Social Anxiety, Bad Knee, Asthma, Lyme Disease, Body pain, Anxiety Attacks, Panic, Attacks, Sensitivity to Noise, [and] Schmorl’s Node of Thoracic Spine.” (Pl.’s Br. (dkt. #12) 1.) 1 Plaintiff moved to file a sur-reply to further discuss two relevant cases published after the close of briefing. (Dkt. #18.) That motion will be granted, and the court has considered plaintiff’s surreply in the opinion below. Despite his limitations, Bubb’s application for supplemental security income was denied initially and on reconsideration. He then requested a hearing before an ALJ, which was held on December 6, 2018. Present at the hearing was ALJ Mengesha-Brown, Bubb, his representative Patrick Schamer, and impartial vocational expert (“VE”) Bernard Preston. During the hearing, the ALJ posed the following hypothetical to the VE: Assume an individual of the same age and education as the claimant who . . . can work at the medium exertional level, is limited to simple, routine one to two step[] tasks, not at a production pace, is limited to simple work related instructions and decisions, requires a work environment with no more than moderate noise, can have occasional superficial interaction with supervisors and coworkers, none with the general public, can tolerate only occasional changes to a routine work setting, and would be off task 10% of the work day. Is there work in the economy that can be performed? (AR at 72.) In response, the VE explained that there were 102,260 uniform attendant positions, 82,192 laundry worker positions, and 201,263 cleaner positions, each of which could be performed given the limitations outlined in the ALJ’s hypothetical. (AR at 72.) The ALJ then posed a second hypothetical, where the individual would be off-task for 20% of the work day, but otherwise had the same limitations as outlined in the first hypothetical. (AR at 73.) The VE testified that the 20% off-task limitation would eliminate all the work he had previously identified. (AR at 73.) Bubb’s representative then questioned the VE regarding his methodology, asking: [Schamer] . . . And what is the source of your numbers? [VE] They come from U.S. Publishing Employment Quarterly, sir. [Schamer] And do you adjust the numbers or you take them as they’re given in that publication? [VE] I don’t adjust the numbers unless they are – have more than DOT title scores associated with them. And then when I adjust the numbers I don’t adjust them, I go based on percentages of how the jobs -- how many percentages are within the job numbers. So for example, I’ll just use a general example, a janitor may have 1,000,000 jobs in the job scope but they may have 50 DOT titles. So in that particular situation if I was testifying on a job that I was testifying I would find the industry it is and then use a percentage of how many jobs within that 1,000,000 and give the job numbers. (AR at 76.) Schamer then informed the ALJ that he objected to the VE’s testimony on the grounds that he did not believe that the numbers were reliable, arguing that use of “the Occupational Employment Quarterly” alone is not “a reliable method.” (AR at 76.) The ALJ noted Schamer’s objection, but did not ask any follow-up questions as to the reliability of the method used by the VE. She did, however, ask the VE whether his testimony was consistent with the Dictionary of Occupational Titles. The VE responded that it was, except for as I noted my employment numbers come from the Department of Labor Employment Quarterly. My testimony in regards to off task behaviors is based on publications on work productivity, and my testimony in regards to contact with the public and coworkers is based on my experience as a job developer. (AR at 76-77.) Following the hearing, ALJ Mengesha-Brown issued a written opinion denying Bubb’s application for benefits. In her opinion, the ALJ followed the well-established, fivestep sequential method for assessing Bubb’s disability. At step one, she concluded that Bubb was not engaged in substantial gainful activity. (AR at 15.) At steps two and three, the ALJ found that Bubb suffered from a variety of severe impairments but that none of his impairments (singly or in combination) met or equaled the criteria of a listing-level impairment. (AR at 17-19.) ALJ Mengesha-Brown then proceeded to step four, finding that Bubb had the residual functional capacity (“RFC”) to perform medium work as defined in 20 CFR 416.967(c) except he could perform simple, routine one to two step tasks but not at production pace. He could make simple, work instructions and decisions. He could work around no more than moderate noise. He could have occasional superficial interaction with supervisors and coworkers but no contact with the general public. He could tolerate occasional changes to a routine work setting. He would be off tasks ten percent of the workday. (AR at 19.) This RFC reflected the limitations included in the first hypothetical the ALJ posed to the VE during the hearing. (See AR at 72 as quoted above.) Finally, at step five, the ALJ considered whether jobs existed in sufficient numbers in the national economy that Bubb could perform. (AR at 26.) The ALJ explained that she found “the vocational expert testimony persuasive and reliable as he has several years of experience working in vocational and rehabilitation services.” (AR at 26.) Thus, the ALJ accepted the VE’s opinion that Bubb could perform the jobs of uniform attendant, laundry worker, and cleaner and that each of these occupations had between 80,000 and 200,000 jobs available nationwide, meaning Bubb was not disabled as defined by the Social Security Act. (AR at 26-27.) OPINION When a federal court reviews a final decision by the Commissioner of Social Security, findings of fact supported by “substantial evidence” are considered “conclusive.” 42 U.S.C. § 405 (g). “Evidence is considered substantial if a reasonable person would accept it as adequate to support a conclusion.” Young v. Barnhart, 362 F.3d 995, 1001 (7th Cir. 2004). When reviewing the Commissioner's findings under § 405(g), therefore, the court cannot reconsider facts a reasonable person would accept, re-weigh the evidence, decide questions of credibility or otherwise substitute its own judgment for that of the administrative law judge. Clifford v. Apfel, 227 F.3d 863, 869 (7th Cir. 2000). Instead, a federal court reviews an administrative disability determination with deference, upholding a denial of benefits unless the ALJ's decision is not supported by substantial evidence or is based on an error of law. 42 U.S.C. § 405 (g); see also Terry v. Astrue, 580 F.3d 471, 475 (7th Cir. 2009). At the same time, a district court may not simply “rubber-stamp” the Commissioner’s decision without a critical review of the evidence. See Ehrhart v. Sec'y of Health and Human Servs., 969 F.2d 534, 538 (7th Cir. 1992). A decision cannot stand if it lacks evidentiary support. Steele v. Barnhart, 290 F.3d 936, 940 (7th Cir. 2002). The ALJ must also explain his “analysis of the evidence with enough detail and clarity to permit meaningful appellate review.” Id.; Herron v. Shalala, 19 F.3d 329, 333-34 (7th Cir. 1994). Finally, in denying benefits, the administrative law judge must build a logical and accurate bridge from substantial evidence to his conclusion. Zurawski v. Halter, 245 F.3d 881, 887 (7th Cir. 2001). Applying this standard, the court turns to plaintiff’s principal criticisms of the ALJ’s decision. I. Reliability of VE Testimony on Available Jobs Bubb first contends that the ALJ erred by uncritically adopting the VE’s unreliable method of estimating jobs in the national economy. “In the context of job-number estimates,” the Seventh Circuit has previously explained that “the substantial evidence standard requires the ALJ to ensure that the approximation [of the availability of jobs] is the product of a reliable method.” Chavez v. Berryhill, 895 F.3d 962, 968 (7th Cir. 2018) (citing Donahue v. Barnhart, 279 F.3d 441, 446 (7th Cir. 2002)). Thus, “[a] finding based on unreliable VE testimony is equivalent to a finding that is not supported by substantial evidence and must be vacated.” Britton v. Astrue, 521 F.3d 799, 803 (7th Cir. 2008). Still, the Seventh Circuit has recognized that a VE’s job estimate number is necessarily an approximation and acknowledged that there is “no way to avoid uncertainty” in arriving at such a figure. Chavez, 895 F.3d at 968 . As a result, a VE’s testimony as to job availability need not meet an “overly exacting standard.” Id. More recently, in Brace v. Saul, 970 F.3d 818 (7th Cir. 2020), however, the court reemphasized that a VE’s “[t]estimony that incants unelaborated words and phrases such as ‘weighting’ and allocation’ and ‘my information that I have’ cannot possible satisfy the substantial- evidence standard.” Id. at 822 . Specifically, at issue in this case is the VE’s use of a so-called “equal distribution method” to calculate the job estimate number. In describing this method in the past, this court has explained that: the expert uses the job titles and job requirements listed in the Dictionary of Occupational Titles (DOT). But because the DOT does not include job-number estimates, the expert seeks that information from an outside source. Unfortunately, it appears that the only reliable statistics are census data for broad categories of jobs, rather than for the job titles used in the DOT. So to estimate the number of positions available for an individual job listed in the DOT, the vocational expert simply divides the total number of jobs in a broad category by the number of job titles that fall within that category. Courtney v. Berryhill, 385 F. Supp. 3d 761, 764 (W.D. Wis. 2018) (internal footnote and citations omitted). Worse for the Commissioner here, the Seventh Circuit has on numerous occasions questioned the reliability of this method. See Chavez, 895 F.3d at 966 - 69 (“We have seen the method applied in other Social Security cases and . . . questioned its use in at least four opinions”). The basis for both the Seventh Circuit’s and this court’s skepticism is a concern that “the method rests on an assumption about the relative distribution of jobs within a broader grouping that lacks any empirical footing.” Id. Indeed, this court has previously held that such a method is “almost certainly inaccurate because it relies on the unfounded assumption that all job titles within a category exist in equal numbers.” Courtney, 385 F.3d at 764 (citing Chavez, 895 F.3d at 966 ). However, neither the Seventh Circuit nor this court had adopted a per se rule barring the use of the equal distribution method. See Courtney, 385 F.3d at 764 (recognizing that it was not the court’s place to “enjoin use of the equal distribution method”); see also Coyier v. Saul, No. 19-cv-393-bbc (W.D. Apr. 2, 2020) (“the fact that the vocational expert used a method that has been criticized by the court of appeals in other cases is not sufficient reason to remand this case”). Rather, where a plaintiff objects to a VE’s use of the equal distribution method, courts have found that an ALJ must affirmatively ensure the reliability of the methods used and the conclusions reached. See Chavez, 895 F.3d 966 (remanding where plaintiff challenged the VE’s use of the equal distribution method and the ALJ failed to elicit a reasoned and principled explanation from the VE to support his job estimate); Courtney, 385 F. Supp. 3d at 765 (remanding where VE used the equal distribution method and the ALJ failed to ensure that the job estimates where the product of a reliable method); Westendorf v. Saul, No. 19-CV-1019-JDP, 2020 WL 4381991 , at *4 (W.D. Wis. July 31, 2020) (same). In this case, there is no dispute that the VE in fact used the equal distribution method to arrive at his job number estimate. (AR at 76; Def.’s Opp’n (dkt. #13) 9.) Moreover, plaintiff’s representative expressly objected to VE’s use of this method, asserting that it was not reliable. (AR at 76.) While the ALJ noted this objection, she failed to inquire into the reliability of the VE’s method during the hearing, nor did she discuss her reasons for relying on this method in her written decision. (AR at 76-77, 16-27.) In fact, the only question the ALJ posed to the VE regarding the reliability of his testimony was whether his conclusions were consistent with the DOT. (See AR at 76.) But this question does not directly relate to the reliability of the VE’s chosen method, and as discussed below, the VE’s superficial response to the ALJ’s question fails to support the VE’s use of the equal distribution method. And in addressing the methodology in her opinion, the ALJ vaguely referred to his VE’s “several years of experience in vocational and rehabilitation services.” (AR at 26.) By failing to meaningfully inquire into the basis for the VE’s confidence in this method or provide an affirmative explanation for the reliability job number estimate at step five in response to Bubb’s reasonable objection, the ALJ erred. The Commissioner points out that, while the Seventh Circuit criticized the equal distribution method in Chavez, the court also explained that the VE could have rehabilitated his conclusions by “draw[ing] on his past experience with the equal distribution method, knowledge of national or local job markets, or practical learning from assisting people with locating jobs throughout the region, to offer an informed view on the reasonableness of his estimates.” (Def.’s Opp’n (dkt. #13) 11 (citing Chavez, 895 F.3d at 969 ).) Moreover, according to the Commissioner, the VE did just that here, having “testified he relied on publications on work productivity and his own experience as other sources for his job numbers,” thereby satisfying the requirements set forth in Chavez. (Def.’s Opp’n (dkt. #13) 11.) However, as already alluded to, this misrepresents the VE’s actual testimony. During the hearing, in response to the ALJ’s only inquiry into whether his testimony was consistent with the DOT, the VE explained that it was except for as I noted my employment numbers come from the Department of Labor Employment Quarterly. My testimony in regards to off task behaviors is based on publications on work productivity, and my testimony in regards to contact with the public and coworkers is based on my experience as a job developer. (AR at 76-77.) At most, this testimony suggests the VE drew on “publications” and his own professional experience to determine whether Bubb could perform the jobs identified in the DOT, but does not provide any support for his use of the equal distribution method to arrive at the job number estimate. See Rennaker v. Saul, No. 20-1042, 2020 WL 4814120 , at *4 (7th Cir. Aug. 19, 2020) (“Although the VE pointed to his own education, research, training, and experience in job placement and vocational rehabilitation to explain the kind of work Rennaker could perform, the VE did not explicitly tie this background to his estimate of nationwide job numbers.”). To the contrary, the VE’s testimony suggested that his only adjustment to the Department of Labor’s quarterly employment numbers was to apply an equal distribution method, rather than actually adjust the numbers based on experience or publications. The Commissioner also attempts to fault plaintiff for failing to offer specific reasons for why the VE’s testimony was not reliable. (Def.’s Opp’n (dkt. #13) 12.). But this misstates plaintiff’s burden at step five, which shifts to the Commissioner to show that sufficient jobs exist in the national economy that the plaintiff can perform. See Clifford, 227 F.3d at 868 ; Rennaker, No. 20-1042, 2020 WL 4814120 , at *3 (“At the final step of its five-step analysis, the agency bore the burden of demonstrating the existence of significant numbers of jobs in the national economy that Rennaker could perform.”). More specifically still, after a reasonable question is raised as to the reliability of a VE’s job number estimate as plaintiff’s representative did expressly in this case, the burden is on the ALJ to ensure that the method offered by the VE is reliable. See Chavez, 895 F.3d at 969 (“By accepting the VE’s estimates at step five because they were ‘not contradicted,’ the ALJ effectively and impermissibly shifted the burden to Chavez.”). And by showing that the ALJ filed to meet this burden, plaintiff has demonstrated on appeal that the ALJ’s decision was not supported by substantial evidence. See Westendorf, 2020 WL 4381991 , at *4 (explaining that it was not the plaintiff’s burden to show that the VE’s numbers were unreliable and concluding that “[b]ecause the ALJ did not ensure that the VE’s job estimates were the product of a reliable method, his decision was not based on substantial evidence”) Finally, in fairness, this court has in the past summarily affirmed an ALJ’s reliance on VE testimony that job numbers were generally derived from the OEQ, as the VE did here. See Arms v. Berryhill, No. 18-CV-476-JDP, 2019 WL 1352809 , at *6 (W.D. Wis. Mar. 26, 2019); Coyier v. Saul, No. 19-cv-393-bbc (W.D. Apr. 2, 2020); Luzar v. Saul, No. 19-CV-1018-JDP, 2020 WL 5249225 , at *4 (W.D. Wis. Sept. 3, 2020); Hartman v. Saul, No. 20-CV-207-JDP (W.D. Wis. Jan. 12, 2021); Boardman v. Saul, No. 19-CV-546-JDP, 2021 WL 822558 , at *4 (W.D. Wis. Mar. 4, 2021). However, in light of Chavez and Brace, the court finds that vague testimony based on aggregated numbers from the OEQ alone do not ensure the reliability of VE job estimate numbers. See Jones v. Saul, No. 1:19-CV-494PPS, 2021 WL 100357 , at *4 (N.D. Ind. Jan. 12, 2021) (VE who based answers off of the OEQ must still be able to cogently describe his or her method at arriving at job estimate numbers). Accordingly, a remand is required on this ground alone. II. 10% Off Task Limitation Plaintiff also argues that the ALJ erred by failing to adequately support her conclusion that Bubb would not be off-task for more than 10% of his workday. (Pl.’s Br. (dkt. #12) 12.) As a general matter, an ALJ must again build a logical and accurate bridge from the evidence to this conclusion as well. Zurawski, 245 F.3d at 887 . In Lanigan v. Berryhill, 865 F.3d 558 (7th Cir. 2017), the Seventh Circuit in particular criticized an ALJ for failing to build an accurate and logical bridge between her “no more than 10% limitation” and relevant evidence in the record, including unrebutted testimony that the claimant was taking unscheduled breaks three to five times during his five hour shifts; had moderate limitations in concentration, persistence, and pace; and had moderate difficulties in performing activities within a schedule. Id. at 563 . Plaintiff also notes a variety of other cases finding error where the ALJ included a percent off-task limitation yet failed to explain adequately the evidentiary basis for this finding. (See Pl.’s Br. (dkt. #12) 17 (citing cases).) Here, to support her percent off-task finding, the ALJ states only that she “limited the claimant in off tasks ten percent of the workday due to his alleged ongoing symptoms,” citing nothing more than the hearing transcript generally in support. (AR at 22.) On the one hand, plaintiff is correct that this vague explanation arguably does not build a logical and accurate bridge from the evidence to her RFC conclusion; but on the other hand, as the Commissioner points out, plaintiff fails to point to any evidence showing that a greater off-task limitation was warranted. Unlike under step 5 where the burden shifts to the Commissioner, however, absent some showing that the evidence of record supports more off-task time, plaintiff has not shown that the ALJ erred. E.g., Ball v. Saul, No. 18-CV-888JDP, 2019 WL 3334658 , at *3 (W.D. Wis. July 25, 2019) (finding no reversible error where plaintiff cited to “no evidence or even identifie[d] a reason why she would be offtask more than 10 percent of the work day”); Madden v. Saul, No. 18-cv-605-bbc, 2019 WL 3059904 , at *5 (W.D. Wis. July 12, 2019) (declining to reverse on this ground when the plaintiff did “not identify any evidence that the administrative law judge failed to consider or account for in choosing the 10-percent limit”); Jackson v. Saul, No. 19-CV-290, 2020 WL 620091 , at *3 (E.D. Wis. Feb. 10, 2020) (“Although the ALJ failed to explain how he determined; that Jackson would be off task only 15% of the workday, this error does not require remand because Jackson has not shown any resulting harm.”). Accordingly, the court concludes that remand to reassess Bubb’s off-task limitation is not warranted, although this would not preclude the ALJ from exercising her discretion to reconsider the sufficiency of the evidence on this issue as well. ORDER IT IS ORDERED that: 1) The decision of defendant Kilolo Kijakazi, Acting Commissioner of the Social Security Administration, denying plaintiff Timothy Bubb’s application for supplemental security income is REVERSED AND REMANDED under sentence four of 42 U.S.C. § 405 (g) for further proceedings consistent with this opinion. 2) Plaintiff’s motion for leave to file a sur-reply brief (dkt. #18) is GRANTED. 3) The clerk’s office is directed to enter judgment in plaintiff’s favor and close this case. Entered this 21st day of December, 2021. BY THE COURT: /s/ __________________________________ WILLIAM M. CONLEY District Judge