Friday, December 25, 2020

Updated Data Sets from Labor

The Department of Labor released two new data sets.  On November 17, 2020, the O*NET Center released the O*NET 25.1 database.  On December 22, 2020, BLS released the latest iteration of the Occupational Requirements Survey.  

What's New with the O*NET

The Occupational Information Network (O*NET) went through eight updates in 2020.  O*NET released updated data and website in August (version 25.0).  The most recent update incorporated the 2019 taxonomy  to bring the O*NET in line with the 2018 Standard Occupational Classification (SOC) system.  

The SOC has 867 detailed occupations coming out of 459 broad occupations, 98 minor groups, and 23 major groups.  The SOC uses a six-digit designation, xx-xxxx.  The O*NET uses an eight-digit designation, xx-xxxx.xx.  The O*NET lists 722 SOC detailed occupations without a detailed O*NET classification.  The O*NET lists 52 SOC detailed occupations with detailed O*NET classification resulting in 149 detailed O*NET-SOC designations.  

The O*NET does not provide data, other than job numbers, for 19 military SOC, 24 "all other" SOC that do have data for a detailed O*NET-SOC, 50 "all other" SOC have detailed O*NET-SOC.  The 867 SOC results in 1,016 O*NET-SOC occupations with detailed data in 947 O*NET-SOC.  

The 2010 SOC had 840 occupations.  The 2010 O*NET SOC had 1,110 occupations.  The number of occupations described shrank in the 2018 SOC and the 2019 O*NET-SOC.

What's New with the ORS

The Occupational Requirements Survey (ORS) is now on its fourth publication with the 2019 and 2020 datasets released this year.  The 2020 ORS provides data for all occupations, 22 major SOC groups, and 304 detailed SOC codes.  

The 2019 ORS provides data for all occupations, the 21 major SOC groups, and 204 detailed SOC.  

The 2020 dataset expanded the coverage to an array that exceeds the 2019 coverage.  That data presentation is not as broad as the 2018 dataset which had data for all occupations, 22 major SOC groups, and 397 detailed SOC.  The good news is that the data now conforms more closely to SSA's needs for the Occupational Information System, the eventual replacement for the DOT.  


Suggested Citation:

Lawrence Rohlfing, Updated Data Sets, California Social Security Attorney (December 25, 2020)

Tuesday, December 15, 2020

Fees for Representing Claimants -- Administrative -- How Much?

 Thirty years ago, Congress amended the representation section of the Social Security Act to establish a presumptive reasonable fee not to exceed 25% of the past due benefits or $4,000, whichever is less.  The US Inflation Calculator states that $4,000 in 1990 is worth $7,964.16 in 2020 dollars.  The current fee cap of $6,000 is behind by 30%.  Social Security informs the representative community that the average fee is less than $4,000 so there is no need for a fee cap increase.  The average will always be lower than $6,000 because that is the maximum and there are cases where very small fees are paid -- initial application paid with one or two months of past due benefits accrued, for example.  

Fees have fallen behind the cost of living by 30%.  And that is the good news.  The bad new starts with the recognition that the CPI includes consideration of Other Services.  Other Services in turn includes Legal Services.  The cost of legal services since 1990 have either lagged inflation, tracked inflation, or outpaced inflation.  Those are the three possible answers.  The correct answer is the third one:  the cost of legal services have outpaced inflation just as education and medical care have outpaced inflation.  I see a hand in the back, "by how much?"  Good question but you won't like the answer.  

Legal services has a base calculation of 100 for December 1986.  As of November 2020, the CPI-U for legal services had risen to 369.112.  We can compare that to the CPI-U for all goods with a base of 1982-84 at 100 to a current value in November 2020 of 260.817.  Using an earlier point in time, the all goods CPI represents 161% inflation.  The legal services inflation is 269% since 1986.  "Goodness gracious, are you kidding me?"  Sorry, I don't make it up, I just report the numbers.  The CPI data tool is here.  

The next question is simple, what does $4,000 worth of legal services in 1991 cost today?  The answer is disturbing:  $11,379.90.  See Historical Pricing for Legal Services.

Before the 1990 amendments, the presumptive fee was $3,000 maximum.  That was the extent of ALJ discretion.  Anything more than that required RCALJ approval.  A fee of $3,000 in 1989 would require $9,691.33 in todays dollars to buy the same quantity and quality of legal services.  That $3,000 fee ceiling was in effect in December 1985 when I got my license.  A 1986 fee requires $11,059 in November 2020 to buy the same quantity and quality of legal services.  

Are some representative overpaid for their potted plant posture during hearings?  Yes.  Should the fee caps get raised to attract the same caliber of legal talent as other areas of law?  Yes.  

The Commissioner should raise ALJ discretion to $15,000 on fee petitions.  The Commissioner should raise the fee agreement process ceiling to $9,000.  Claimants deserve that caliber of representation.  

"You get what you pay for."  


Suggested Citation:

Lawrence Rohlfing, Fees for Representing Claimants -- Administrative -- How Much?, California Social Security Attorney (December 15, 2020)

Thursday, December 3, 2020

The Ability to Stop and Stretch for One to Two Minutes

 Oh no, that would not impact the ability to perform the work identified.  You have heard the hypothetical question, seen the residual functional capacity assessment, and read the results in the ALJ decision.  The vocational expert is grossly uninformed, negligent, or making it up.  Pick 'em.  

According to the Handbook of Methods, the Occupational Requirements Survey collects data on the cognitive and mental requirements of work.  Included in that category of workplace functions, the ORS measures work pace and the ability to pause work.  See page 33.  The Collection Manual defines pause control:

Collect the presence (yes/no) of a worker’s ability, for a personal reason, to easily step away from work for short periods of time outside of scheduled breaks such as lunch or morning/afternoon break periods.

Page 83.  The Collection Manual explains how Labor accomplishes those data calculations and what those data points mean.  Employers or analysts code "yes" for the ability to pause work if either of two conditions are met:

  • Workers typically have the flexibility to choose when to take breaks throughout the day.
  • There is an overall time limit for breaks, but such breaks are allowed.
Id.  Employers or analysts code "no" for the ability to pause work any of three conditions are present:
  • If the worker would need to find someone to cover his or her responsibilities.
  • When breaks are usually allowed, but not during certain busy periods in the performance of a critical job task (for example, when work is exceedingly heavy, when a line of customers is building, etc.).
  • Workers are required to be present at a workstation for a defined period of time.
Id.  The Collection Manual gives three examples of when pause control exists:
  • Office workers can pause and take a quick walk down the hall when they need a mental break. Even when they have group meetings with coworkers or clients, they have the ability to step away quickly, if needed. A high school teacher teaches classes. In the middle of a lesson, the teacher can ask students to work quietly while she steps out into the hall for a few minutes when feeling overwhelmed. 
  • Landscapers mow lawns, prune trees and plants, and weed flower beds. Landscapers may briefly step away from their duties to take medication or attend to other medical needs without notifying a supervisor. 
  • An outside sales representative solicits business from clients via telephone, email, and in-person. Reps have the ability to schedule client appointments at their convenience. Even in the middle of a call or appointment, the rep has the ability to politely request to step away briefly
Page 84.  The Collection Manual gives examples of when pause control does not exist
  • A building security guard at a secure facility screens employees and visitors entering the facility; and walks standard patrols on a rotating basis with other guards. They are not able to leave their station without asking someone else to cover for them. 
  • A kindergarten teacher must call to the principal’s office to have another adult keep an eye on the students when he leaves the classroom. 
  • Cashiers are able to take a break when there are no customers but cannot leave their station when there is a line. 
  • A surgeon would not be able to step away easily while performing surgery. 

The academic questions are nice to know, but how the data plays out is where the rubber meets the road.  Selecting out from the ORS data set (2019) 13 of my favorites:

Without the Ability to Pause Work


all workers


office clerks, general


janitors and cleaners, except maids and housekeeping cleaners


workers in office and administrative support occupations


inspectors, testers, sorters, samplers, and weighers


maids and housekeeping cleaners


receptionists and information clerks


laborers and freight, stock, and material movers, hand


stockers and order fillers




workers in production occupations




Nearly half of all workers do not have the ability to pause work.  That is a surprising statistic in light of typical VE testimony.  Less than a quarter of general office clerks cannot self-regulate.  General office clerks is a huge unskilled body of work, generally sedentary, that becomes unavailable with either an education deficit (limited or less), a limitation to simple work, or a limitation in interacting with others.  

Cashiers is another large unskilled occupational group, this time light.  Less than 10% of cashiers have the ability to pause work activity.  Cashiers get eliminated in toto with a limitation on contact with the public or a limitation to simple work.  Check out the physical demands of cashiers in the ORS; hint: light is not the maximum exertion and standing is not limited to four or even six hours for full-time work.  

Workers in production occupations, including all supervisors and a large swath of skilled workers, have the ability to pause work in just over 60% of jobs.  Whether unskilled production workers in tandem with other workers could ever pause is highly doubtful.  

The ORS has and is developing data that answers many of the routine questions that we hear in ALJ examination of vocational experts.  The anecdotal experience of the VE without an accepted methodology for extrapolating that experience to the national economy, when nationally collected data disagrees, is not substantial evidence.  The representative, us, must submit rebuttal data to the agency. 


Suggested Citation:

Lawrence Rohlfing, The Ability to Stop and Stretch for One to Two Minutes, California Social Security Attorney (December 3, 2020) 

Tuesday, October 6, 2020

Assembler, Production (DOT 706.687-010) and the JBP Industries

This is the first follow-up on the generic instruction to always check the NAICS codes when the VE uses JBP as a basis for job number evidence.  Production occupations are the bread the butter of vocational experts (VE).  The American economy has a large manufacturing sector.  It makes sense that production work exists.  The question is how many exist as light, unskilled work.  A VE could identify production assembler and cite Job Browser Pro (JBP) for the proposition that this single DOT represents over 67,000 jobs.  The question that the representative must ask is whether JBP is reliable as to this occupation.  We start with the DOT code:

706.687-010 ASSEMBLER, PRODUCTION (any industry)
Performs repetitive bench or line assembly operations to mass-produce products, such as automobile or tractor radiators, blower wheels, refrigerators, or gas stoves: Places parts in specified relationship to each other. Bolts, clips, screws, cements, or otherwise fastens parts together by hand, or using handtools or portable power tools. May tend machines, such as arbor presses or riveting machine, to perform force fitting or fastening operations on assembly line. May be assigned to different work stations as production needs require. May work on line where tasks vary as different model of same article moves along line. ay be designated according to part or product produced.
GOE: 06.04.22 STRENGTH: L GED: R2 M1 L1 SVP: 2 DLU: 80
The DOT designates any industry but the narrative specifies the mass production of radiators, blowers, wheels, refrigerators, and gas stoves. Any industry does not mean all industries, it means many industries and at least more than four. The narrative narrows the scope of any industry to those industries that make vehicle parts, refrigerators, and gas stoves. Anything outside of that scope requires an explanation under SSR 00-4p.

JBP starts the list of North American Industry Classification System (NAICS) list with food manufacturing subsector (NAICS 311000).  The manufacture of food does not involve the mass production of vehicles parts, refrigerators, or gas stoves.  It requires an explanation for that apparent conflict with the DOT.  

JBP lists eight industry groups (NAICS 311100 - 311800) after listing the food industry subsector.  That is double counting.  Each of the industry groups list between 4 and 12 DOT codes in each occupation-industry intersection.  The line for the industry subsector claims that production assembler is the only DOT code at that intersection.  The counting of the industry subsector fails to account for the other DOT codes in the industry groups.  

The CBP hyperlink in JBP lists no DOT codes.  The CBP page for other food manufacturing (NAICS 311900) does not list any production workers, all other (SOC 51-9199) occupations.  There are more jobs for production workers within NAICS 311000 than there are for NAICS 311100 - 311800.  

The next industry subsector listed by JBP is beverage and tobacco manufacturing (NAICS 312000).  JBP duplicates the number of jobs with industry group of beverage manufacturing (NAICS 312100).  The group has six DOT codes, the industry subsector just has production assembler.  Production assembler does not manufacture beverages or tobacco products and JBP has double counted.  

JBP lists textile product mills (NAICS 314000) and leather and allied product manufacturing (NAICS 316000).  JBP lists the printing and related support activities industry (NAICS 32300) and the petroleum and coal products manufacturing industry (NAICS 324000) subsectors.  These industries require an explanation in light of the narrative description of what production assemblers do on the job.  

JBP lists wood product manufacturing subsector (NAICS 321000) and three industry groups (NAICS 321100 - 321900).  JBP commits the same errors in listi9ng the paper manufacturing industry subsector (NAICS 322000)  and the pulp, paper, and paperboard mills manufacturing industry group (NAICS 322100).  JBP lists chemical manufacturing subsector (NAICS 325000) and seven industry groups (six of them in two composite groups) (NAICS 325100 - 325900).  JBP suggests the plastics and rubber products manufacturing subsector (NAICS 326000) as well as the plastics industry group and the rubber product manufacturing group (NAICS 326100 - 326200).  There is an apparent conflict with the DOT and double counting without accounting for all the DOT codes listed in the industry groups.  

The process of questioning the NAICS code assignment and the double counting problem continues.  JBP has not set forth a reliable methodology and any testimony based in whole or in part of JBP for this occupation is necessarily flawed.  

JBP confesses the unreliable methodology on the face of the Estimated DOT Employment Numbers tab.  JBP states that the Occupational Employment Statistics estimates that 1,590 DOT codes inside of production workers, all other represent 222,230 jobs in the nation.  The selected industries represent 274,076 jobs in version 1.70.2 of JBP.  The part cannot exceed the size of the whole.  JBP's job number estimate is not reliable for this occupation.  


Lawrence Rohlfing, Assembler, Production (DOT 706.687-010) and the JBP Industries, California Social Security Attorney (October 6, 2020),

Sunday, October 4, 2020

When the Vocational Expert Cites Job Browser Pro for Job Numbers; Check the NAICS Codes

 The title says it all and if you know exactly what that means, this piece tells you something that you already know.  For those that don't know exactly what the title of this piece means, this is a primer.  


The DOT contains a parenthetical statement of industry.  The industry statement is an integral and inseparable part of the DOT description of any occupation. The Introduction, Parts of the Occupational Definition to the DOT says exactly that.  

The body of the definition contains the lead statement that describes worker actions; objective or purposes of worker actions; machines, tools, equipment and work aids used; materials, products, subject matter dealt with or services rendered; and instructions/judgment involved.  This is the other clue to where the occupation belongs.  

Job Browser Pro

JBP uses the Department of Labor crosswalk to determine the SOC code for an occupation.  This used to be always true; it is now true most of the time.  See the prior post about Goode v. Commissioner.  

JBP uses a combination of the industry designation and the narrative definition of the occupation to assign the industry according to the NAICS codes.  NAICS codes are the foundation of County Business Patterns and are used by both the Occupational Employment Statistics and Employment Projections to distribute jobs within a SOC code.  


NAICS codes in four flavors:  two-digit industry sectors; three-digit industry subsectors; four-digit industry groups; and five and six-digit specific industries.  Other than the six digit variety, the NAICS code is the same in the United States, Canada, and Mexico.  The key to understanding what is presented is easy -- ignore the zeroes, they are place holders, not digits for coding.  

The Problem

On many occupations, JBP cites NAICS codes using three and four-digit to describe the SOC-NAICS code intersections.  When JBP does that, it double counts the jobs and will not do so consistently between the other DOT codes involved.  The problem is the OES.  Labor does not always report four-digit codes for job numbers within the SOC or does not report all of the four-digit codes within the SOC.  Where the DOT code exists in more than one industry group but OES does not report all the industry groups, JBP will report the three-digit subsector and then also report the four digit subsectors that OES does report.  This is an invalid methodology.  We will look at production assembler as an egregious example of this problem later this week.  Production assembler has multiple industry subsector and industry group assignments that lead unreliable job numbers.   

The second problem is not obvious.  JBP reports an industry group for one occupation and the industry subsector for another occupation.  This is again an invalid methodology and wrong.  This is the problem with advertising distributor.  

The third problem is apparent upon inspection of the JBP for the occupation cited by the VE.  JBP assigns industries that are in conflict with either the DOT industry designation or the narrative lead statement of the work performed.  Next week, we will look at small products assembler (I and II) as an example of this problem.  As a tease, know that small product assemblers do not work in any food industry.  

When JBP double counts jobs, the method is facially unreliable.  When JBP assigns NAICS codes inconsistently between DOT codes in the same SOC-NAICS intersection, the method is facially unreliable.  When JBP assigns DOT codes to industries inconsistent with either the DOT industry designation or the narrative, JBP has an apparent conflict with the DOT without a reliable explanation to permit resolution of the conflict.  More about this in the posts to come.  


Lawrence Rohlfing, When the Vocational Expert Cites Job Browser Pro for Job Numbers; Check the NAICS Codes, California Social Security Attorney (October 4, 2020),  

Thursday, October 1, 2020

Advertising Distributor and Job Browser Pro -- Why is it Wrong?

 A vocational expert (VE) identifies advertising-material distributor (DOT 230.687-010) as a light, unskilled, simple 1-2 step occupation.  The VE says that the occupation represents about 30,000 jobs in the nation.  I think:  "That can't be right."  So I ask:  "How do you know that?"

The VE points to Job Browser Pro (JBP) as the basis for the testimony.  We are in the middle of the CV-19 crisis; the hearing is by telephone; I pull up JBP on the fly.  Damn, the VE is close, JBP says that advertising-material distributor represents 29,792 jobs in the nation.  Almost all of the jobs are in the employment services (NAICS 561300) industry.  Employment services is the industry group that contains temporary agencies.  "Don't temp agencies send people in all occupations?"  After some struggle back and forth, the VE affirms that that suspicion is true but that JBP is still a valid source for job numbers.  I now need to disassemble the occupation and its job numbers.  

JBP lists  advertising-material distributor and billposter (DOT 299.667-010) (medium unskilled) as helpers - production workers (SOC 51-9198) as existing in employment services.  Four digit industry designations represent industry groups that contain five and six-digit industries.  NAICS 561300 contains specific industries.  I need to find them and determine whether JBP has designated a specific industry and attributed all of those jobs to other DOT codes.  

JBP has a CBP hyperlink in the box describing the industries in which the DOT code exists.  CBP is the designation for County Business Patterns, #2 on the list of enumerated examples of administrative notice.  The CBP link takes me to the SkillTRAN Industry Analyzer.  I click on the Staffing Patterns button.   Clicking on the green "+" button for 51-9198, I find the DOT codes that SkillTran assigns to NAICS 561300.  Low and behold, there are not two but ten DOT codes:

SOC 51-9198 – Helpers – Production Worker








business services





any industry





any industry





any industry





oils & grease





machine shop





any industry





any industry





any industry





any industry



I change the 561310 to 561320.  This is the Temporary Help Services industry.  This has the green "+" button for 51-9198.  I click on it.  There is that same ten DOT code list that I saw for 561300.  Curiosity killed the cat so I meow and look at the specific five-digit industries.  I change the 561300 in the URL to 561310.  This is the industry designation for Employment Placement Agencies.  Sorting by SOC code, I see that SkillTRAN lists our two culprits, advertising-material distributor and billposter, but not the other eight.  

I change 561320 to 561330.  This is the Professional Employer Organizations industry.  This industry as a DOT list so I sort by SOC code.  The occupations from 51-9198 are advertising-material distributor and billposter.  

Back to JBP, I search randomly for store laborer (DOT 922.687-058).  JBP states that this occupation exists in industry 561320, that the industry employs 53,532 helpers - production worker, and that store laborer accounts for 6,691 of the jobs (which is almost all of the jobs attributed to store laborer by JBP).  JBP lists eight DOT codes:  the list of ten occupations less advertising-material distributor and billposter.  

I need to complete the circle.  Resorting to the OES May 2019 National Industry-Specific Occupational Employment and Wage Estimates for NAICS 561300.  The OES confirms that "Industries within NAICS 561300 - Employment Services" includes 561320 - Temporary Help Services.

The OES reports that Helpers - Production Workers represents 78,690 jobs as of May 2019.  

The OES reports that Helpers - Production Workers represents 68,130 jobs as of May 2019.  

The difference is 10,560 jobs.  The OES does not report job numbers for NAICS 561310 or 561320.  CBP does but I have already completed the circle.  

JBP double counted the 53,532 jobs that it places in NAICS 561320 in accounting for them all again in NAICS 561300.  Including the part-time work, JBP estimates that NAICS 561320 has 84,971 jobs and NAICS 561300 has 94,492 jobs.  JBP is starting 16,000 jobs too high.  

The question is, how many  advertising-material distributor jobs are there?  JBP uses equal distribution as its methodology.  JBP does not list the extra eight occupations in NAICS 561310 or 561330 but lists all ten occupations in NAICS 561320.  Using JBP's intersectional equal distribution, half of the 10,560 jobs and one-tenth of the 68,130 jobs to arrive at 12,063 jobs, full and part-time for  advertising-material distributor.  

JBP says that 37% of the jobs are part-time.  The O*NET says that 6% of helpers - production workers have part-time employment.  The industry-specific full versus part-time is more applicable.  Advertising material distributor represents 7,600 full-time jobs that matter at step five of the sequential evaluation process.   

This is the danger of failing to fastidiously stick to either three, four, and five digit NAICS codes by JBP.  The data limitation exists because JBP relies on the OES and the OES only reports statistically significant occupation clusters.  When JBP reports job numbers, the careful user must travel up and down the NAICS code tree to determine whether the industry designation is accurate and that the jobs were not over or undercounted.  The mantra must remain, trust but verify.  When it comes to vocational experts, they do not get paid enough to take that journey.  We have a fiduciary obligation to take the trip.  Bon voyage.    


Suggested Citation:

Lawrence Rohlfing, Advertising Distributor and Job Browser Pro -- Why is it Wrong?, California Social Security Attorney (October 1, 2020),

Tuesday, September 29, 2020

Training Videos on YouTube has training videos.   Both are an hour.  The first focuses on marker and cashier II.  The second addresses egg processor, hand almond blancher, and stuffer.  

The written materials covering both presentations are at the download section of  HERE.  

The OccuCollect presentation is HERE.  This presentation addresses OccuCollect data for marker and stuffer.  

The Indiana bar presentation is HERE.  This presentation addresses OccuCollect data and Job Browser Pro data for egg processor, hand almond blancher, and stuffer.  

If you would like to see an occupation addressed in this manner, comment below.  If we get to three occupations of interest, we will put up another video.  

Thursday, August 27, 2020

Larson v. Saul and Skidmore Deference

 When you cannot possibly sleep at night and need a brain game as a complete distraction from the world, the deference doctrine is a good choice.  This is the Chevron watershed that stands at the heart of administrative law in the last 35 years. Deference is the doctrine that fills in the gaps where the law is ambiguous by a publication from the agency that is not law.  Deference to a statutory construct is the core of Chevron deference.  

If a statute is genuinely ambiguous, the path opens for the executive agency to interpret the statute.  Chevron, U.S.A., Inc v. Nat. Res. Def. Council, Inc..  The courts will defer to the agency's construction of the statute unless plainly erroneous or inconsistent with the statute.  

If a regulation is ambiguous, the agency can interpret is own lack of clarity.  Kisor v. Wilkie.  The courts will defer to the agency's construction of the regulation unless plainly erroneous or inconsistent with the statute or regulation.  

But when the statute is ambiguous and either the regulations do not interpret the statute or Congress did not delegate to the agency, full-blown deference under Chevron or Kisor does not attach.  Then the agency gets the power to persuade deference.   Skidmore v. Swift & Co.

Enter Larson v. Saul and the Windfall Elimination Provision of the Social Security Act.  42 U.S.C. § 415(a)(7).  The Eighth Circuit decided Petersen v. Astrue, finding that the WEP did not apply to dual status technicians (civilian workers that are members of the uniformed services).  The COSS responded to Petersen with Acquiescence Ruling 12-1(8)POMS 00605.380 regurgitates the ruling.  

After the publication of Peterson, the ruling, and POMS, three other circuits weighed in on the dual service technician issue.  In the Eleventh Circuit, Martin v. Social Security Administration Commissioner  relied at least in part on Skidmore deference to agree with the COSS.  The Sixth and Tenth Circuits did not get past the step one question (is the statute ambiguous) to require resort to a deference doctrine.  Babcock v. Soc. Sec. Comm'r; Kientz v. Comm'r, SSA

Larson rejected the non-ambiguity findings of Babcock and Kientz.  Larson found the WEP provisions truly ambiguous and that both the claimant and the COSS made reasonable arguments for the construction of the statute.  Based on that equipoise, Larson granted Skidmore deference to the ruling and POMS.  Larson's benefits were subject to the windfall elimination provision of the Act.  

There are two problems with the grant of deference by the Ninth and Eleventh Circuits.  The administrative state makes a decision about whether dual service technicians without input from the stakeholders and weighing of policy considerations in public.  Without notice and comment, the legislation by administrative fiat should not form part of the social contract.  Second, Congress punted.  Congress either did not consider in the statute whether dual service technicians would get unreduced Social Security benefits or did not want to make that decision public for fear of retribution by voters impacted by the WEP.  Members of Congress have plausible deniability in the effectuation of a statute on constituents.  

The four circuits and the COSS create another problem.  Dual service technicians should move to geographic boundaries of the Eighth Circuit before applying for Social Security retirement or disability benefits.  A national program lacks uniform application across the country.  But that is the nature of every Acquiescence Ruling that cabins a decision to a circuit -- a lack of uniform application of a national standard to residents of different states.  Nor is this the kind of problem that will ever get before the Supreme Court to resolve the split in the circuits.  It is not likely that any of the remaining circuits will ever disagree with the duo of decisions disagreeing with Petersen or the duo of decisions yielding under the deference doctrine to cabin Petersen to the Eighth Circuit.  

The practice pointer is simple.  A claimant for benefits in the dual service technician role should move to the Eighth Circuit before applying for benefits.  A concern for uniform application of a statute should prompt the agency to rescind the Acquiescence Ruling  and POMS with a regulation that addresses the problem and move forward with a request for Brand X deference.    


Suggested Citation:

Lawrence Rohlfing, Larson v. Saul and Skidmore Deference, California Social Security Attorney (August 27, 2020)

Tuesday, August 25, 2020

Dissecting the VE Testimony in Brace v. (Berryhill) Saul

 Earlier, we discussed applying and differentiating the Seventh Circuit decision in Brace v. Saul from the Supreme Court decision in Biestek v. Berryhill and the Ninth Circuit decision in Ford v. Saul.  To get a better handle on the depravity of the vocational expert testimony, we examine the District Court decision in Brace v. Berryhill.  The court summarized the residual functional capacity:

After careful consideration of the entire record, the undersigned finds that the claimant has the residual functional capacity to perform light work as defined in 20 CFR 404.1567(b), as follows: lift, carry, push, and pull 20 pounds occasionally and ten pounds frequently; sit, stand, and walk, each, for six hours in an eight-hour workday; can remain in one position for at least 30 minutes, but would need to change positions occasionally throughout the day; no reaching overhead; no foot pedals; occasional fingering, handling, and feeling with the nondominant hand, but no restrictions with the dominant hand; no ramps or stairs; no slippery or uneven surfaces; no dangerous machinery or unprotected heights; and no operating a motor vehicle as a condition of employment.
The critical limitations are light work, stand/walk six hours; changing positions every 30 minutes; occasional manipulation with the nondominant hand.  The other limitations will rarely impact light and sedentary work.  

The VE identified call out operator (DOT 237.367-014), semi-conductor bonder (DOT 726.685-066), and counter clerk (DOT 249.366-010).  The Court of Appeals adds registration clerk and the District Court refers to the four examples as representing 140,000 jobs.  

Call-Out Operator

Call-out operator is a sedentary, unskilled occupation requiring reasoning level 3.  It does require occasional handling and fingering.  Call-out operator belongs to Credit Authorizers, Checkers, and Clerks (SOC 43-4041).  Credit authorizers represent 30,300 jobs and are typically semi-skilled or skilled per the 2018 OOH.  The O*NET suggests that 25% of those jobs qualify as unskilled within the sub-group of Credit Checkers (O*NET 43-4041.02).  Job Browser Pro estimates fewer than 3,000 jobs.  

Semi-Conductor Bonder

Semi-conductor bonder is a sedentary, unskilled occupation requiring reasoning level 2.  It requires frequent handling and fingering.  Semi-conductor bonder has an apparent conflict with the SCO.  

Semi-conductor bonder belongs to Metal Workers and Plastic Workers, All Other (SOC  51-4199).  Metal and plastic workers represent 25,900 jobs in the nation and are typically semi-skilled or skilled per the 2018 OOH.  Neither the O*NET nor the ORS address this very small occupational group.  Job Browser Pro estimates less than 20 jobs.  

Counter Clerk

Counter clerk is a light, unskilled occupation requiring reasoning level 2.  It requires occasional handling and fingering.  Counter clerk belongs to the occupational group of Counter and Rental Clerks (SOC 41-2021).  Counter and rental clerks represent 436,100 jobs and are typically unskilled per the 2018 OOH.  

The O*NET says that 40% of the jobs are part-time.  Part-time work does not satisfy the step five burden on the Commissioner.  The O*NET says that almost 14% of these jobs are unskilled.  

The ORS disagrees with the O*NET about skills level:  almost 69% are unskilled.  Counter and rental clerks use both hands for gross manipulation in 76.5% (frequently in 57.2%) of jobs and fine manipulation in 65.6% of jobs.  As a separate description involving fingering, counter and rental clerks keyboard in almost all jobs and occasionally in 67.3% of jobs.  Counter and rental clerks stand 100% of the day at the 75th and 90th percentiles.  Counter clerks sit 95% of the day at the 90th percentile.  

Job Browser Pro estimates fewer than 1,800 jobs.  

Registration Clerk

The DOT lists two registration clerks.  One is SVP 5 and skilled.  The second is SVP 3 and semi-skilled.  The ALJ did not find transferable skills.  This occupation has an apparent conflict with the DOT and the absence of transferable skills.  

Registration clerk (DOT 205.367-042) is a sedentary, semi-skilled occupation requiring reasoning level 3.  It requires frequent handling. Registration clerk has an apparent conflict with the handling limitations in the RFC.  

Registration clerk belongs to Interviewers, Except Eligibility and Loan (SOC 43-4111).  Interviewers represent 204,600 jobs and are typically unskilled per the 2018 OOH.  

The O*NET states that interviewers work part-time in 55% of jobs.  Part-time work does not satisfy the step five burden on the Commissioner.  The O*NET states that almost 76% of jobs are unskilled.  

The ORS states that 42.1% of interviewer jobs are unskilled.  Interviewers engage in sedentary work in 86.3% of jobs.  Interviewers sit 87.5% of the day at the median.  Interviewers can arrange their workday to sit or stand/walk at will in about half the jobs.  Interviewers engage in fine manipulation occasionally in 88.4% of jobs and keyboard in 97.9% of jobs.  

Job Browser Pro estimates the number of registration clerk jobs at less than 1,500.  


A full-court press is mixed.  The DOT/SCO knocks out two of the four occupations.  Submitting the JBP data shows the absence of a significant number of jobs, except in the Sixth Circuit.  The ORS requires careful analysis and familiarity with the data by the vocational expert for unskilled counter clerks and interviewers.  Every occupation represents a different problem amplified by the peculiarities of the specific residual functional capacity finding.  The Employment Projections, Occupational Employment Statistics, O*NET OnLine, Occupational Requirements Survey and private resources such as Job Browser Pro and provide greater insight into the mostly opaque testimony of a vocational expert.  


Suggested Citation:

Lawrence Rohlfing, Dissecting the VE Testimony in Brace v. (Berryhill) Saul, California Social Security Attorney (August 25, 2020)

Saturday, August 22, 2020

Brace Yourself for an Emerging Paradigm

 The Seventh Circuit decided Brace v. Saul on August 14, 2020.  The opening paragraph says it all:  the vocational expert's testimony was inscrutable and that does not satisfy the lenient substantial evidence standard.  The problem of the black box approach to vocational expert testimony is recurrent.  

In Biestek v. Berryhill, the vocational expert relied on in-house and confidential labor market surveys, which the ALJ refused to order that the VE turn over.

Biestek's attorney asked O'Callaghan "where [she was] getting those [numbers] from." [...].  O'Callaghan replied that they came from the Bureau of Labor Statistics and her "own individual labor market surveys." Ibid. The lawyer then requested that O'Callaghan turn over the private surveys so he could review them. Ibid. O'Callaghan responded that she wished to keep the surveys confidential because they were "part of [her] client files."

Without the BLS data or some description of methodology, Biestek affirmed.  

In Ford v. Saul, the VE refused to provide information about the sources establishing the number of jobs testimony.  

The vocational expert testified that 130,000 addresser and 9,800 ink-printing jobs existed nationwide and that Ford's RFC allowed her to perform these jobs. In response to cross-examination about how he had derived those estimates, the vocational expert stated "[m]y numbers come from a variety of sources which include the Department of Labor and the U.S. Chamber of Commerce and actually Social Security, itself, the Census Bureau, through the [International Trade Administration], supply really good numbers and, believe it or not, the state of Alaska has good national numbers." Probing the expert's conclusion regarding the number of addresser jobs, Ford's counsel asked "[w]hich publication indicated that there were that number of jobs?" The expert responded, "I don't have that information in my notes. I typically average all my sources." The counsel then asked, "What were the numbers that you averaged together to get 130,000?" Again, the expert responded, "I don't have that information in my notes, either." The expert explained that he averages the numbers from his various sources once a year, and then puts those numbers in his notes. Ford's counsel then stated he had no further questions.

Without sources form the DOL, Social Security, and Census Bureau, Ford affirmed.  

The biggest difference between Brace and the decisions in Biestek and Ford is the question of methodology.  Biestek and Ford asked for sources, not methods.  Brace asked the "how" question.  How do the sources inform your opinion about the call-out operator, semiconductor bonder, registration clerk, and counter clerk aggregated to total 140,000 jobs in the national economy.  "Brace's attorney asked the vocational expert ("VE") to explain his methodology for estimating that 140,000 jobs are available in the national economy."  

The ALJ reasoned that Brace did not object to the admissibility of the VE testimony.  Imagine how long the hearings would last if the claimant had to object to every piece of evidence's admissibility in order to later argue that the evidence was not worthy of weight.  Brace quickly rejected that vaporous statement.  

The ALJ reasoned that even if the VE has off by an unknown factor, the number of jobs would still be available.  Brace affirmed that substantial evidence does not find satisfaction in substantial speculation.  

The lesson from Brace is simple.  Representatives must ask the VE not only for the sources but also for the methodology.  Then the claimant can argue that the VE testimony is not substantial on its face.  Of course, submitting rebuttal evidence remains an equally mandatory approach to undermining VE testimony.  


Suggested Citation:

Lawrence Rohlfing, Brace Yourself for an Emerging Paradigm, California Social Security Attorney (August 22, 2020)

Thursday, August 13, 2020

Larkin v. Saul -- POMS Supports the Majority and the Dissent

  An unpublished decision containing a dissent provides some level of concern.  Larkin v. Saul provides the example of the day.  Judges Collins and VanDyke affirmed the final decision of the Commissioner to deny Larkin's claim for benefits.  Judge Berzon would have reversed and remanded.  The case centers around the state agency psych workup, juxtaposing the section I worksheet with the section III residual functional capacity assessment.  

The case centers around application of Stubbs-Danielson v. Astrue.  The majority Stubbs-Danielson as the "most applicable precedent."  Larkin's description of the case informs the application of precedent:

Not unlike this case, the ALJ in Stubbs-Danielson was faced with physicians' statements like: "claimant has several mental limitations" and "borderline intellectual functioning" and "slow pace in thought and action" and is "moderately limited in her ability to perform at a consistent pace without an unreasonable number and length of rest periods," but also that she "showed good persistence" and "retained the ability to carry out simple tasks" and "could perform simple work without public contact." Id. at 1171, 1173 (internal quotation marks omitted). The ALJ ultimately "determined that Stubbs-Danielson `retained the residual functional capacity to perform simple, routine, repetitive sedentary work, requiring no interaction with the public.'" Id. at 1171 (alteration marks omitted).

Hoopai v. Astrue stands for the same basic proposition.  The doctor said two things about persistence that reside in conflict.  Moderate limitations in performing at a consistent pace conflicts with the statement that the claimant had good persistence and still had the ability to perform simple tasks without public contact.  

POMS DI 24510.060 describes the process of the mental residual functional capacity assessment.  The purpose of section I is to document the presence and degree of specific functional limitations and hte adequacy of the documentation.  Section B.2.a states that section I is a mere worksheet and an aid that does not constitute the RFC assessment.  Section B.2.c describes moderately limited as greater than not significantly limited and less than markedly limited.  A moderate limitation in any of the 20 categories represents a significant impairment but it is not a RFC assessment.  That structure supports the treatment of the tension between moderate limitations in moderate limitation in performing at a consistent pace and having good persistence for simple non-public work.  The section III findings are the RFC and the section I categories are a worksheet.  That supports the majority opinion.  

Judge Berzon makes the observation:

The ability to complete a regular workweek is a separate employment qualification from performing simple, routine tasks. A person may be able to perform simple, routine tasks on one day and still be limited in her ability to appear for work the following day because she is suffering from an episode of depression or anxiety. Dr. Comrie's summary sentence on which the majority focuses—"[Larkin] will have some limitation with keeping a regular workweek due to her mood and anxiety [symptoms] but she retains the ability to follow through with routine tasks with reasonable [concentration, persistence, and pace]"—firmly distinguishes between the ability to show up to work regularly and the ability, when at work, to complete tasks. If the ALJ had questions about what the sentence meant, then he should have inquired further.

 POMS DI 24510.060, first bullet, describes an endorsement of a moderate limitation in performing at a consistent pace as present and a specific functional limitation.  POMS DI 25020.010 sec. B.3 lists 14 functional requirements that are critical to performing unskilled work.  Section B.3.i lists as one of the critical functions:

complete a normal workday and workweek without interruptions from psychologically based symptoms and perform at a consistent pace without an unreasonable number and length of rest periods. (These requirements are usually strict.)

It is not persistence but a consistent pace that is both critical to performing unskilled work and represents a usually strict requirement.  Persistence is an attribute of semi-skilled and skilled work per sec. B.4.b.  Judge Berzon poses the question of whether having the persistence to complete simple work represents a different functional problem compared to getting to work regularly and completing tasks on a timely basis.  POMS answers that question, yes.  More importantly, the need to complete tasks in a timely fashion is strictly enforced.  

My question is whether that question was posed by the representative at the hearing.  The lack of statement to the contrary tells me, "no."  The order adopting the report and recommendation is available on google scholar but the R&R is not.  

Nor was POMS centered in the available materials.  The Ninth Circuit is scattered on this issue.  POMS is entitled to at least respect.  As a construction of the RFC regulation that is facially incomplete (404.1545(c)), POMS warrants deference.  That is a subject for another day.  


Suggested Citation:

Lawrence Rohlfing, Larkin v. Saul -- POMS Supports the Majority and the Dissent, California Social Security Attorney (August 13, 2020)