Stacy S. v. Kijakazi and the Occupational Requirements Survey

 Stacy S. v. Kijakazi addresses vocational expert testimony given in 2021 after a 2019 application for benefits.  The dates are important in this case.  The case stands for two propositions important to me: (1) the vocational witness did not have a reasonable explanation of a reliable methodology; and (2) the vocational witness used the Occupational Requirements Survey (ORS) to estimate job numbers at least in part.  The Court's summary of the vocational witness testimony is the typical smoke and mirrors that the VW give -- they really don't have a methodology.  Justice Gorsuch was right, they do pull it out of a hat.  

Stacy S. interests me because of its discussion of the ORS. The Court summarizes:

The VE averred that his testimony regarding the ability to perform Claimant's past work was based on the Dictionary of Occupational Titles ("DOT"); testimony regarding occasional handling and fingering and overhead reaching was based on the Occupational Requirements Survey ("ORS"); testimony regarding "the attendant work" was based on the Selected Characteristics of Occupations ("SCO"); and testimony regarding pushing, pulling, reaching, foot controls, absenteeism, and time off-task was based on "42 plus years of vocational counseling practice and placement experience of hundreds of people with disabilities." Id. at 60-61.

 The VW explained:

Well, I'm talking about—I think the ORS is based on SOC categories, which are general classifications of jobs. And within the [SOC] category, they're characterizing the number of jobs that are unskilled, semi-skilled, and so forth, and also breaking them down according to how often they are sedentary, light, and other physical—other requirements. And with an [SOC] code, all I'm doing is providing a representative DOT number of—a DOT number that falls within that SOC code.

 The Court held:

To the extent it appears the VE relied on the ORS for the estimated numbers, Claimant correctly argues that the ORS data shows "a percentage of total jobs within broad categories," but does not readily show how many jobs altogether are available within a given job category. [Dkt. 9 at 16.] Moreover, the currently available ORS data is only a preliminary estimate based on two years of a planned five-year sample period. As Claimant underscores, "any job numbers obtained from this source would necessarily be incomplete and unreliable." [Dkt. 9 at 17.]

Now we can take off our gloves.  The ORS does not answer DOT questions.  The ORS does answer whether any DOT code within the group has the characteristics that the VW wants to impute.  For instance, a VW says that a production job stands four hours per day and represents 50,000 jobs.  The ORS establishes that of the circa 200,000 production worker jobs, those workers are on their feet at least 75% of the day at the 10th percentile.  It is not possible for less than 10% of 200,000 to mathematically come up with 50,000 jobs.  The ORS can also establish that the incidence of sedentary work is negligible.  The ORS can establish that most of the jobs require medium exertion and most require skills.  To use the ORS to rebut the description that general office worker (Router, 222.587-038) or receptionist (Information Clerk, 237.367-018) do not require frequent manipulation when the SCO describes them that way is a misuse of the data.  Attendant (Children's Attendant, 349.677-018) is a part-time job in the same category as usher.  

The VW was also using stale data.  As of February 2021 when this hearing occurred, BLS had published the final first wave of the 2018 dataset, the 2019 dataset, and the 2020 dataset.  The 2018 dataset represented the culmination of the 2016-17-18 datasets.  The data that we now have in 2022 represents six sets of data representing responses from human resources (employers) about the requirements of work.  "Two years of a planned five-year sample" screams that the VW had not kept up with the data.  

The current data is in the 2021 dataset.  OccuCollect publishes the final first wave representing the 2018 dataset.  BLS publishes both but the presentation of the 2018 dataset is in an XLSX file that requires another file to understand the cells.  

A knife in the hands of a criminal is a deadly weapon.  A knife in the hands of a skilled surgeon saves lives.  It is apparent that data in the hands of vocational witnesses is akin to the former and not the latter.  

___________________________

Suggested Citation:

Lawrence Rohlfing, Stacy S. v. Kijakazi and the Occupational Requirements Survey, California Social Security Attorney (August 3, 2022) https://californiasocialsecurityattorney.blogspot.com

The author has been AV-rated since 2000 and listed in Super Lawyers since 2009.