Dec 6, 2023

Social Security Says Don't Rely On SkillTRAN For Numbers Of Jobs

     From Emergency Message EM-21065 REV, Guidelines for Using Occupational Information in Electronic Tools issued yesterday (emphasis added):

In making disability determinations, SSA relies primarily on the Dictionary of Occupational Titles (DOT) (including its companion publication, the Selected Characteristics of Occupations (SCO)) for information about the requirements of work in the national economy. The SSA Digital Library hosts three searchable databases developed by SkillTRAN for the DOT. Adjudicators may use these tools to help make an assessment at step four (see 20 CFR 404.1560(b)(2) and 416.960(b)(2)) or step five (see 20 CFR 404.1566(d), 404.1568(d)(2), 416.966(d), and 416.968(d)(2)) of the sequential evaluation process. ...

     [The three searchable databases] contain DOT and SCO occupational information developed by the United States Department of Labor (DOL). This information is useful, but it does not replace SSA policy or adjudicative judgment and decision-making. While the DOT and SCO are acceptable sources of occupational information for adjudicating disability claims, they also contain (1) information that must not be used in disability adjudication because our rules and subregulatory guidance do not permit it and (2) information of which we do not take administrative notice. ...
    Federal agencies now publish labor market information by the Standard Occupational Classification (SOC) code. Those with vocational expertise use various approaches to arrive at informed estimates of numbers of jobs that exist within a DOT occupation. Results may differ given the method used for the estimate. We have not reviewed and do not specifically endorse the SkillTRAN proprietary algorithm. ...


    To explain, in responding to hypothetical questions, vocational experts testifying at hearings on Social Security disability claims give lists of jobs that a hypothetical claimant could perform. They also testify to the numbers of such jobs in the economy. The numbers of jobs figures come from a propriety program, SkillTRAN. I think that everyone who has looked at this issue, including Administrative Law Judges, knows that those SkillTRAN numbers are bull. Social Security is finally acknowledging this. 
    Will this affect cases where these numbers have already been cited in decisions? We'll see.                                                                            

14 comments:

Anonymous said...

The problem that Social Security has is that when a Vocational Witness says there are 50,000 lens inserters and then when they are crossed and it is pointed out that JobBrowser Pro, the SkilTran application says there are actually there are actually only 76 such jobs, they want the ALJ to only consider the original 50,000 based on supposed expertise and not the latter number based on JobBrowser Pro's analysis. And worse, at least some witnesses, will back down and say that's right, there are only a very limited number of such jobs.

I don't know how many such jobs there are but I know that there are not 50,000 lens inserters in the United States. The issue of significant numbers of jobs has been a mess since the sequential evaluation process was adopted in 1979 . And willfully closing their eyes to questions about reasoning level of supposed jobs or general learning ability of supposed jobs or job incidence figures conjured from wholecloth is simply SSA's way of ignoring the issue that in today's world, the number of unskilled, sedentary jobs is vanishingly small and not a significant number for someone so impaired.

Anonymous said...

SSA has also historically overestimated the cognitive impact of 99% of claimants’ non substance abuse related mental health conditions. So on balance, I’d wager the claimants have had an advantage.

Anonymous said...

This should not be a welcome change. SkillTRAN DOT specific job estimates are the lowest compared to any other source I have seen VEs use. I have routinely caught VEs using the OES group job number total, instead of the DOT specific estimate. Pending a total replacement of the DOT, SkillTRAN is valuable resource for attorneys. The alternative is just asking VEs to make up numbers.

Anonymous said...

The EM is a mess. GED should not be considered? Really? So, lets say you have a brain injured person who as a result of their injury functions at a GED math level of 1 (simple 1-2 digit addition and subtraction only). SSA won't consider that limitation against the math requirements of a job like Physicist, GED math level 6, "Advanced calculus: Work with limits, continuity, real number systems, mean value theorems, and implicit functions theorems..." That's what is implied, and it makes no sense.

The EM statement that "aptitudes do not represent functional requirements for work," is dubious. The aptitude job ratings that the EM refers to include U.S. Dept. of Labor surveys documenting job requirements for categories like manual dexterity, motor coordination, verbal aptitude, ability to learn, and others. Of course they represent functional employer work expectations and requirements. Work typically has speed and accuracy requirements for both mental and physical tasks. This EM encourages adjudicators to ignore those important categories (which aren't rated well in the DOT or SCO), to the detriment of people with disabilities.

Anonymous said...

Misleading excerpt there. The EM says adjudicators need VE evidence to rely on SkillTRAN numbers. It does NOT say SSA can’t rely on them. If the VE gives the SkillTRAN numbers, they’re proper evidence.

Basically, all this says is SSA doesn’t take administrative notice of SkillTRAN numbers.

Anonymous said...

@2:54

I've actually gotten AC remands specifically due to the GED ratings. It's wild that SSA is now suggesting they are not considered. I suspect what happened is, the instruction was intended to just state the GED ratings do not correspond to unskilled, semiskilled, or skilled. That's an accurate statement.

Drew C said...

@3:46 I think you have it right. I misread this on first review. This is the key sentence: "We have not reviewed and do not specifically endorse the SkillTRAN proprietary algorithm." So SSA neither rejects nor endorses SkillTran job estimates.

But why is this an "Emergency Message"? Very strange that after 10+ years of VEs, attorneys, and Fed Courts citing SkillTran numbers, SSA is issuing a poorly drafted EM, as opposed to an SSR or HALLEX. Why now?

Are EMs afforded much legal weight in Federal Court? I am sure I have seen case law where the Commissioner argues against the legal weight of POMs, asserting they are "non-binding".

Anonymous said...

This is not new , but rather an update of an AM that was first published last year. The changes made are not substantive. I don't know why any claimant's representative would want to argue Skilltran's numbers are no good. When it comes to unskilled sedentary work, Skilltran posits very few occupations. It was the source of the numbers in the Washington Post article that came out last year arguing there were essentially no unskilled occupations in the national economy anymore and all y'all were singing the praises of Skilltran at that time.

Anonymous said...

@ 10:10

I agree. It's bizarre that SSA's policy and investigative people have not caught on to the problem. Or maybe they have and just don't care. Anybody who sits in more than just a few hearings knows that different vocational experts will often give, and ALJs will often accept, extremely different national job numbers for the exact same DOT title.

If SSA is going to write Emergency Messages about vocational testimony, then how about this instead. We need to investigate why these vocational experts keep testifying to such wildly different numbers for the same DOT job titles, figure out which of them are using unreliable methods, and then do something to stop it.

Anonymous said...

@ 3:46 is correct. The post is misleading. SkillTRAN is not prohibited.

Annie42 said...

This gives ALJs cover for ignoring the reasoning levels of jobs (i.e., the GED ratings). They are trying to undo, via "emergency message" the case law in virtually every circuit. (In the 10th, it is Hackett).

This is yet another attempt to undo our work in the federal courts (see also the elimination of the court-created treating physician rule via regulation). This is sub-regulatory, which means it is not subject to notice and comment. SSA loves to issue sub-regulatory guidance that says "it is our policy that the sun rises in the West," and all the ALJs dutifully get in line.

This is also contrary to the VE Handbook, p. 39, which says the VE must explain how someone limited to only simple work can do an R3 job.

Anonymous said...

@9:49

In the 9th, it's Zavalin v. Colvin and Rounds v Comm. of SSA. I've always suspected, given the VE handbook literally identifies them as example of an apparent conflict, the reason there hadn't been any ARs issued in response to the numerous circuit court opinions is that it is NOT contrary to agency policy, and in fact IS agency policy to consider the GED ratings.

This ER is weird.

Anonymous said...

The EM directly conflicts with the statute which specifically requires consideration of education in disability determination at step five. Telling adjudicators not to consider GED in those claims is telling the to disregard the statute’s requirements.

Anonymous said...

The problem with using GED as defined in the DOT and by SSA is that even some of the examples even in levels one and two are beyond the abilities of even many college graduates. That does not mean educational level is not considered as it is a consideration in the grids