WEBVTT 00:00:14.439 --> 00:00:17.019 Hello, and welcome to Administrative Remedies, 00:00:17.039 --> 00:00:19.019 because you can't fix what you don't understand. 00:00:19.820 --> 00:00:21.879 Brought to you in part by the University of Tulsa 00:00:21.879 --> 00:00:24.539 College of Law. I'm Gwendolyn Savitz, an associate 00:00:24.539 --> 00:00:26.600 professor here at TU and the associate dean of 00:00:26.600 --> 00:00:28.780 research and intellectual life. And I'm Mark 00:00:28.780 --> 00:00:31.300 Rourke. I'm the dean of the College of Law. We'll 00:00:31.300 --> 00:00:33.100 be breaking down complex doctrines with real 00:00:33.100 --> 00:00:35.380 -life analogies and examples to demystify the 00:00:35.380 --> 00:00:37.740 world of administrative law for everyone trying 00:00:37.740 --> 00:00:39.460 to understand how government actually works. 00:00:40.119 --> 00:00:41.859 Agencies are the main way the federal government 00:00:41.859 --> 00:00:44.460 gets things done. It's not through Congress for 00:00:44.460 --> 00:00:46.299 reasons we'll be addressing over the course of 00:00:46.299 --> 00:00:50.359 this series. All right, Mark, picture a job, 00:00:50.520 --> 00:00:53.200 pneumatic tube operator. You receive cylinders 00:00:53.200 --> 00:00:55.359 that arrive through a network of tubes running 00:00:55.359 --> 00:00:57.719 through a building. You unload whatever's inside, 00:00:58.000 --> 00:01:00.640 paperwork or small items. You route them to the 00:01:00.640 --> 00:01:02.479 right department and you send the tubes back. 00:01:02.820 --> 00:01:05.260 Sure, like what they use at pharmacies and banks 00:01:05.260 --> 00:01:07.739 at the drive -thru. Yeah, they still have a specific 00:01:07.739 --> 00:01:10.079 purpose, but what they don't really have are 00:01:10.079 --> 00:01:12.140 people dedicated solely to operating them all 00:01:12.140 --> 00:01:14.760 day long. In large department stores, hospitals, 00:01:14.840 --> 00:01:16.760 and office buildings, that used to be an actual 00:01:16.760 --> 00:01:19.980 job, doing that all day. All right. Let's talk 00:01:19.980 --> 00:01:22.219 about another one now. Nuts order. You sit at 00:01:22.219 --> 00:01:24.019 a conveyor belt and you separate the good nuts 00:01:24.019 --> 00:01:26.939 from the bad nuts as they go by, by hand. Yeah, 00:01:27.019 --> 00:01:28.900 the trained squirrels in Willy Wonka, right? 00:01:29.019 --> 00:01:33.000 So not really a thriving profession. Right. Except 00:01:33.000 --> 00:01:35.859 until June of 2024, the Social Security Administration 00:01:35.859 --> 00:01:39.060 was using both of these jobs and others like 00:01:39.060 --> 00:01:41.920 them as reasons to deny people disability benefits. 00:01:42.519 --> 00:01:45.159 OK, what do you mean using them? The last step 00:01:45.159 --> 00:01:47.379 of the disability evaluation, and we will get 00:01:47.379 --> 00:01:50.060 to this in detail, the agency has to show that 00:01:50.060 --> 00:01:52.379 even if you can't do your old job, there's some 00:01:52.379 --> 00:01:54.620 other job you could do. And when it's looking 00:01:54.620 --> 00:01:56.439 for that other job, it reaches for a federal 00:01:56.439 --> 00:01:59.400 catalog of occupations. And that catalog's last 00:01:59.400 --> 00:02:04.409 edition came out in 1991. 1991. Right. And for 00:02:04.409 --> 00:02:06.390 years, if you applied for disability and you 00:02:06.390 --> 00:02:08.590 couldn't go back to your old job, the agency 00:02:08.590 --> 00:02:10.629 might tell you, well, you could be a pneumatic 00:02:10.629 --> 00:02:13.449 tube operator or a nut sorter and deny your claim 00:02:13.449 --> 00:02:17.490 on that basis. So real people have actually been 00:02:17.490 --> 00:02:20.590 denied disability because a federal adjudicator 00:02:20.590 --> 00:02:24.009 said that they could potentially be a pneumatic 00:02:24.009 --> 00:02:27.569 tube operator. Thousands of times. The Washington 00:02:27.569 --> 00:02:29.810 Post did a piece on it. Congress looked at it. 00:02:29.870 --> 00:02:31.590 Courts have been complaining about this for years. 00:02:32.199 --> 00:02:34.840 And in 2024, they fixed it, right? Well, they 00:02:34.840 --> 00:02:37.939 did something. They issued two emergency messages 00:02:37.939 --> 00:02:41.280 on the same day. The first one identified 114 00:02:41.280 --> 00:02:44.520 occupations that it called isolated. These were 00:02:44.520 --> 00:02:46.819 jobs that only existed in very limited numbers, 00:02:47.000 --> 00:02:50.199 like astronomer or oral and maxillofacial surgeon 00:02:50.199 --> 00:02:53.979 or director of a state historical society. Adjudicators 00:02:53.979 --> 00:02:56.120 can't cite to any of these to deny a claim anymore. 00:02:56.620 --> 00:02:59.120 Those don't sound like the jobs we were just 00:02:59.120 --> 00:03:01.840 talking about, though. No, they're not. And that's 00:03:01.840 --> 00:03:04.800 the point. So most of those on the first list 00:03:04.800 --> 00:03:06.800 weren't jobs that people were citing in the first 00:03:06.800 --> 00:03:10.379 place. Nobody was saying, we understand you can't 00:03:10.379 --> 00:03:12.280 go back to your old job, but that's fine. You 00:03:12.280 --> 00:03:15.099 can just become an astronomer. Yeah. So what 00:03:15.099 --> 00:03:17.479 about the pneumatic tube operator or the nut 00:03:17.479 --> 00:03:20.360 sorter or the document preparer or the addresser? 00:03:20.439 --> 00:03:22.560 All right. Those are on the second list. Those 00:03:22.560 --> 00:03:25.379 are 13 occupations that had been showing up in 00:03:25.379 --> 00:03:27.319 denials and that federal courts had been questioning. 00:03:27.960 --> 00:03:30.719 So the second list is where the reform actually 00:03:30.719 --> 00:03:33.259 bites. Right. It doesn't prohibit citing those 00:03:33.259 --> 00:03:35.979 jobs, but it requires the ALJ to get additional 00:03:35.979 --> 00:03:37.879 evidence from the vocational expert that the 00:03:37.879 --> 00:03:40.939 job, as it is currently performed, is consistent 00:03:40.939 --> 00:03:43.319 with the claimant's functional capacity and exists 00:03:43.319 --> 00:03:46.500 in significant numbers. Okay, hold on. That's 00:03:46.500 --> 00:03:48.919 not a fix. That's a paperwork requirement. The 00:03:48.919 --> 00:03:52.539 same V .E. who was citing pneumatic tube operator 00:03:52.539 --> 00:03:56.979 in 2023 can now cite it in 2026 as long as they 00:03:56.979 --> 00:04:00.139 say the magic words. Hundreds of cases per year 00:04:00.139 --> 00:04:02.919 per judge. How much friction does adding a sentence 00:04:02.919 --> 00:04:06.879 to the testimony actually create? Well, potentially 00:04:06.879 --> 00:04:09.180 not that much, but that's the question of the 00:04:09.180 --> 00:04:12.080 whole episode. Notice what they didn't do in 00:04:12.080 --> 00:04:14.240 either message. They didn't update the catalog. 00:04:14.500 --> 00:04:16.339 They didn't revise any of the job descriptions. 00:04:16.660 --> 00:04:19.360 They didn't fix how the agency understood modern 00:04:19.360 --> 00:04:21.680 work. They pulled a few entries off one list 00:04:21.680 --> 00:04:23.639 and they added a documentation step to another. 00:04:24.079 --> 00:04:26.600 Although certainly if someone were denied because 00:04:26.600 --> 00:04:28.740 it was claimed that there were more than a thousand 00:04:28.740 --> 00:04:31.120 local pneumatic tube operating jobs, that would 00:04:31.120 --> 00:04:33.279 be something they could challenge. OK, I feel 00:04:33.279 --> 00:04:35.139 like I should be angrier about this than I am. 00:04:35.819 --> 00:04:38.699 Why am I not that angry? It might be because 00:04:38.699 --> 00:04:41.160 you're not applying it to the 2 .5 million disability 00:04:41.160 --> 00:04:43.800 applications that Social Security receives every 00:04:43.800 --> 00:04:47.300 year. 2 .5 million? We've talked about it before, 00:04:47.420 --> 00:04:48.920 but really we're talking about the population 00:04:48.920 --> 00:04:51.480 of Chicago applying for disability every year. 00:04:51.620 --> 00:04:53.800 And every one of those applications has to get 00:04:53.800 --> 00:04:56.379 decided. Someone has to look at the medical evidence, 00:04:56.620 --> 00:04:58.459 figure out whether this person can still work, 00:04:58.579 --> 00:05:01.480 and say yes or no. Okay, I think I see where 00:05:01.480 --> 00:05:03.410 you're going. The stuff we've been talking about 00:05:03.410 --> 00:05:05.910 this season with due process, the inquisitorial 00:05:05.910 --> 00:05:08.750 model, the ALJ developing the record, credibility 00:05:08.750 --> 00:05:11.790 findings, internal appeals, those are designed 00:05:11.790 --> 00:05:14.209 for the individual case. We were talking about 00:05:14.209 --> 00:05:17.449 one claimant, one record, one hearing, one decision. 00:05:17.750 --> 00:05:20.750 And today we're asking what happens to all of 00:05:20.750 --> 00:05:22.689 that when you run it two and a half million times 00:05:22.689 --> 00:05:26.329 a year? And the 1991 catalog is what it looks 00:05:26.329 --> 00:05:29.410 like. Right. The catalog isn't the problem. The 00:05:29.410 --> 00:05:32.000 catalog is the symptom. The problem is a system 00:05:32.000 --> 00:05:34.040 that has to decide all these cases and can't 00:05:34.040 --> 00:05:36.019 afford to look too closely at any one of them, 00:05:36.160 --> 00:05:38.199 including the question of whether the job the 00:05:38.199 --> 00:05:40.259 agency says you could still do exists in the 00:05:40.259 --> 00:05:43.540 form it's described. So this isn't a new doctrine 00:05:43.540 --> 00:05:46.019 episode. Right. It is a stress test episode. 00:05:46.379 --> 00:05:48.579 We've spent the season building a model of how 00:05:48.579 --> 00:05:50.899 adjudication is supposed to work. And Social 00:05:50.899 --> 00:05:53.060 Security is what happens when you run it at industrial 00:05:53.060 --> 00:05:56.639 volume. And this 1991 catalog is one of the places 00:05:56.639 --> 00:05:59.129 where the strain shows. Plus, it's also just 00:05:59.129 --> 00:06:01.689 generally interesting. So there are roughly two 00:06:01.689 --> 00:06:04.029 and a half million applications every year, and 00:06:04.029 --> 00:06:06.389 most of them get decided on paper at the initial 00:06:06.389 --> 00:06:08.610 stage. And most of those initial decisions are 00:06:08.610 --> 00:06:11.230 denials. So how many denials of that two and 00:06:11.230 --> 00:06:12.649 a half million are we talking about roughly? 00:06:12.850 --> 00:06:16.129 So they historically run 60 to 70 percent. So 00:06:16.129 --> 00:06:19.639 out of two and a half million applicants. A million 00:06:19.639 --> 00:06:22.480 and a half -ish are being told no in the first 00:06:22.480 --> 00:06:25.160 round before any hearing, before any judge, just 00:06:25.160 --> 00:06:27.120 based on the paper review by a state disability 00:06:27.120 --> 00:06:30.019 examiner working with the medical file. And then 00:06:30.019 --> 00:06:32.019 what happens to those denials? They could ask 00:06:32.019 --> 00:06:33.939 for a reconsideration, which is another paper 00:06:33.939 --> 00:06:35.839 review by a different examiner in the same state 00:06:35.839 --> 00:06:38.800 agency. Most reconsiderations affirm the denial. 00:06:39.319 --> 00:06:41.120 Only then do you get to ask for your hearing 00:06:41.120 --> 00:06:43.199 in front of the administrative law judge. So 00:06:43.199 --> 00:06:45.459 by the time you actually see a judge, you've 00:06:45.459 --> 00:06:48.300 already been told no twice. Right. And statistically, 00:06:48.600 --> 00:06:50.439 the people who may get to the hearing level are 00:06:50.439 --> 00:06:52.360 much more likely to win than people at the earlier 00:06:52.360 --> 00:06:54.379 stages, which is kind of telling on its own. 00:06:55.319 --> 00:06:58.160 That's a strange signal. The initial determinations 00:06:58.160 --> 00:07:00.819 are getting things wrong often enough that a 00:07:00.819 --> 00:07:02.839 meaningful percentage of them flip when someone 00:07:02.839 --> 00:07:04.980 actually listens. That would certainly be one 00:07:04.980 --> 00:07:06.899 reading of it. The agency would tell you that 00:07:06.899 --> 00:07:08.600 the cases look different by the hearing stage 00:07:08.600 --> 00:07:11.300 because by that point, more medical evidence 00:07:11.300 --> 00:07:13.389 has been developed. Both things are probably 00:07:13.389 --> 00:07:16.009 true, but yes, the paper review process at the 00:07:16.009 --> 00:07:17.569 front end is doing something the hearing keeps 00:07:17.569 --> 00:07:20.110 correcting. And how many hearings is that now? 00:07:20.449 --> 00:07:23.029 Hundreds of thousands a year. And these are conducted 00:07:23.029 --> 00:07:26.089 out of the more than 160 hearing offices scattered 00:07:26.089 --> 00:07:28.649 across the country. And their own performance 00:07:28.649 --> 00:07:33.149 dashboard put the backlog at about 330 ,000 hearings 00:07:33.149 --> 00:07:36.089 as of January 26, with an average wait time of 00:07:36.089 --> 00:07:38.550 about nine months, which is actually a big improvement. 00:07:39.240 --> 00:07:41.660 In 2023, the average was more like 15 months. 00:07:41.879 --> 00:07:45.819 And it was worse in fiscal year 2018 when it 00:07:45.819 --> 00:07:50.839 was 595 days. Wow. People wait over a year, sometimes 00:07:50.839 --> 00:07:54.199 years, multiple years for their hearing. Right. 00:07:54.259 --> 00:07:57.120 And so before we go further, nothing we're about 00:07:57.120 --> 00:07:59.560 to talk about is an accusation of misconduct. 00:07:59.819 --> 00:08:02.480 The ALJs, the disability examiners, the support 00:08:02.480 --> 00:08:04.779 staff, they're doing their jobs. The system itself 00:08:04.779 --> 00:08:07.930 is what's under stress. So when I hear a judge 00:08:07.930 --> 00:08:10.290 is deciding hundreds of these cases a year... 00:08:10.290 --> 00:08:12.769 This is not a criticism of the judge. That's 00:08:12.769 --> 00:08:15.490 their job. We talked about this in the acquisitorial 00:08:15.490 --> 00:08:18.069 model episode. The ALJ isn't a neutral referee 00:08:18.069 --> 00:08:20.750 listening to two sides. The ALJ is supposed to 00:08:20.750 --> 00:08:22.949 be developing the record, questioning witnesses, 00:08:23.170 --> 00:08:25.850 and deciding the case. So the judge is theoretically 00:08:25.850 --> 00:08:28.569 the prosecutor's job, the defense attorney's 00:08:28.569 --> 00:08:31.189 job, and the judge's job all at once. Right. 00:08:31.290 --> 00:08:33.090 And they get to do that several hundred times 00:08:33.090 --> 00:08:37.259 a year. So you can ask how deep the record realistically 00:08:37.259 --> 00:08:39.259 is going to be developed in any single case. 00:08:39.759 --> 00:08:42.580 The structural problem isn't that anyone is lazy 00:08:42.580 --> 00:08:45.179 or careless. It's that the model assumes an amount 00:08:45.179 --> 00:08:47.779 of attention per case that's just mathematically 00:08:47.779 --> 00:08:50.320 impossible at this volume. Yeah, that's part 00:08:50.320 --> 00:08:52.240 of what we're talking about this episode. All 00:08:52.240 --> 00:08:54.340 of this is in one way or another a response to 00:08:54.340 --> 00:08:57.059 that math. Okay, how do you actually decide in 00:08:57.059 --> 00:08:59.820 two and a half million cases? What's the framework? 00:09:00.320 --> 00:09:02.539 This is the five -step sequential evaluation. 00:09:02.779 --> 00:09:05.340 It is the doctrinal heart of how disability gets 00:09:05.340 --> 00:09:07.179 decided, and we haven't really talked about it 00:09:07.179 --> 00:09:12.840 before. Let's look at 20 CFR 404 .1520. The sequential 00:09:12.840 --> 00:09:15.779 evaluation process is a series of five steps 00:09:15.779 --> 00:09:19.399 that we follow in set order. If we can find you 00:09:19.399 --> 00:09:22.200 are disabled or not disabled at a step, we can 00:09:22.200 --> 00:09:24.860 make our determination or decision, and we don't 00:09:24.860 --> 00:09:27.539 go to the next step. If we cannot find that you 00:09:27.539 --> 00:09:30.179 are disabled or not disabled at a step, we go 00:09:30.179 --> 00:09:32.519 on to the next step. All right. So there was 00:09:32.519 --> 00:09:34.460 a previous episode where we talked about Kathleen. 00:09:34.659 --> 00:09:37.899 She was 54 living in Oklahoma. She'd worked 22 00:09:37.899 --> 00:09:40.740 years as a warehouse supervisor. She had degenerative 00:09:40.740 --> 00:09:42.799 disc disease and she'd gotten denied at the initial 00:09:42.799 --> 00:09:45.419 stage by a state DDS analyst she'd never met. 00:09:45.679 --> 00:09:47.779 Then she got denied again at reconsideration 00:09:47.779 --> 00:09:50.460 by a different analyst at the same agency. And 00:09:50.460 --> 00:09:52.960 we sat with her in her hearing. But what we weren't 00:09:52.960 --> 00:09:54.940 walking through is exactly what the determination 00:09:54.940 --> 00:09:57.960 was based on, what those five steps ask. That's 00:09:57.960 --> 00:10:01.379 what we do today. Okay, step one. Substantial 00:10:01.379 --> 00:10:03.580 gainful activity. Are you working right now above 00:10:03.580 --> 00:10:06.340 a certain income threshold? If yes, you're automatically 00:10:06.340 --> 00:10:09.039 denied. It doesn't matter how sick you are. What 00:10:09.039 --> 00:10:12.320 amount do you think that is? Oh, I would say 00:10:12.320 --> 00:10:16.080 it's got to be above the poverty rate. I'm going 00:10:16.080 --> 00:10:19.779 to say $40 ,000. That's per month. Oh, per month. 00:10:20.190 --> 00:10:24.870 So per month, 40 ,000 divided by 12 would be 00:10:24.870 --> 00:10:28.490 3 ,000. So I'm going to say $3 ,500 a month. 00:10:28.730 --> 00:10:32.190 All right. It is actually $1 ,690 per month. 00:10:32.350 --> 00:10:35.629 Wow. So if you make more than $1 ,690 per month, 00:10:35.690 --> 00:10:40.070 you're automatically denied. Yes. Wow. So here 00:10:40.070 --> 00:10:42.169 Kathleen lost her job. She used it for Family 00:10:42.169 --> 00:10:44.230 Medical Leave Act. She couldn't perform the essential 00:10:44.230 --> 00:10:47.110 functions anymore. So she's not working. She's 00:10:47.110 --> 00:10:50.379 earning nothing. So she passes step one. Okay, 00:10:50.440 --> 00:10:53.820 what's step two? Step two, we ask, is there a 00:10:53.820 --> 00:10:56.139 severe impairment? Does she have a medically 00:10:56.139 --> 00:10:58.460 determinable impairment that significantly limits 00:10:58.460 --> 00:11:01.220 basic work activities? Degenerative disc disease 00:11:01.220 --> 00:11:03.980 with imaging to back it up? Sure. That counts 00:11:03.980 --> 00:11:06.299 as severe. She passes step two. This isn't a 00:11:06.299 --> 00:11:09.159 really hard bar to pass. And step two is mostly 00:11:09.159 --> 00:11:11.740 a filter for people with really minor conditions. 00:11:12.139 --> 00:11:15.740 Right. So mild allergy, a sprained ankle, these 00:11:15.740 --> 00:11:18.240 won't make it past step two. But most serious 00:11:18.240 --> 00:11:20.559 medical conditions do. Step three is where things 00:11:20.559 --> 00:11:22.919 start to get interesting. OK. This is where we 00:11:22.919 --> 00:11:25.399 say, does your condition meet or equal one of 00:11:25.399 --> 00:11:28.240 the listings? The listings are Social Security's 00:11:28.240 --> 00:11:30.200 catalog of conditions that are presumptively 00:11:30.200 --> 00:11:33.100 disabling. If you can document that your impairment 00:11:33.100 --> 00:11:35.360 matches the specific medical criteria of one 00:11:35.360 --> 00:11:37.480 of these listed impairments, you're approved 00:11:37.480 --> 00:11:40.340 without having to go any further. About 10 to 00:11:40.340 --> 00:11:42.419 15 percent of approvals happen at step three. 00:11:42.840 --> 00:11:44.639 This would be like you've been diagnosed with 00:11:44.639 --> 00:11:46.919 Huntington's disease or ALS. They're not going 00:11:46.919 --> 00:11:48.960 to need a lot of time figuring out exactly what 00:11:48.960 --> 00:11:51.799 you can do. Okay. So most people don't clear 00:11:51.799 --> 00:11:53.960 step three. Right. Most people don't get approved 00:11:53.960 --> 00:11:56.340 at step three. We still have to keep going. So 00:11:56.340 --> 00:11:58.519 here for Kathleen, her degenerative disc disease 00:11:58.519 --> 00:12:01.179 is real, but it probably doesn't meet the specific 00:12:01.179 --> 00:12:03.720 listing criteria for spinal disorders. They require 00:12:03.720 --> 00:12:06.139 things like specific neurological findings, particular 00:12:06.139 --> 00:12:09.779 MRI evidence, a lot of really detailed specific 00:12:09.779 --> 00:12:12.590 stuff. She's not going to get approved at step 00:12:12.590 --> 00:12:14.629 three, but she's also not going to get denied, 00:12:14.809 --> 00:12:18.230 so we move on. Okay, what about step four? Can 00:12:18.230 --> 00:12:20.850 you do your past relevant work? Here, the agency 00:12:20.850 --> 00:12:23.230 is making what it calls a residual function capacity 00:12:23.230 --> 00:12:26.830 finding, an RFC. What can you still do given 00:12:26.830 --> 00:12:29.570 your limitations? Can you lift stuff? How much 00:12:29.570 --> 00:12:32.070 can you lift? How long can you stand? Can you 00:12:32.070 --> 00:12:34.570 bend and reach? Okay, what about for Kathleen? 00:12:36.139 --> 00:12:38.519 This can find that she can do light work, like 00:12:38.519 --> 00:12:41.179 maybe lifting up to 20 pounds occasionally, 10 00:12:41.179 --> 00:12:43.620 pounds frequently, standing and walking about 00:12:43.620 --> 00:12:46.200 six hours a day. She clearly cannot go back to 00:12:46.200 --> 00:12:48.000 warehouse supervision at that level. That job 00:12:48.000 --> 00:12:50.740 required serious lifting, prolonged standing, 00:12:50.940 --> 00:12:54.399 climbing. So she passes step four. And now we're 00:12:54.399 --> 00:12:56.519 at step five, which I'm guessing is the one that 00:12:56.519 --> 00:12:58.860 actually matters in most cases. Right. So this 00:12:58.860 --> 00:13:01.480 is where most claims are won or lost. Given this 00:13:01.480 --> 00:13:04.179 person's residual functional capacity, plus their 00:13:04.179 --> 00:13:06.279 age, their education, and their work history, 00:13:06.519 --> 00:13:09.100 is there other work in significant numbers in 00:13:09.100 --> 00:13:11.580 the national economy that they could do? Any 00:13:11.580 --> 00:13:14.279 other work at all? Right. In the entire national 00:13:14.279 --> 00:13:16.320 economy. This is where the vocational expert 00:13:16.320 --> 00:13:18.840 comes in. They testify at the hearing, and the 00:13:18.840 --> 00:13:21.600 ALJ will describe a hypothetical claimant. a 00:13:21.600 --> 00:13:25.100 54 -year -old with limited education and a residual 00:13:25.100 --> 00:13:27.620 functional capacity for light work, and ask the 00:13:27.620 --> 00:13:29.659 vocational expert what jobs someone like them 00:13:29.659 --> 00:13:32.960 could do. And the VE says, what exactly? So in 00:13:32.960 --> 00:13:34.740 a lot of these cases, they might say something 00:13:34.740 --> 00:13:37.299 like surveillance system monitor or document 00:13:37.299 --> 00:13:41.379 preparer or addresser. But surveillance system 00:13:41.379 --> 00:13:45.159 monitor, like watching security cameras? Yes. 00:13:45.600 --> 00:13:48.480 Where are these jobs coming from? Like, who actually 00:13:48.480 --> 00:13:51.399 has this list? This is the Dictionary of Occupational 00:13:51.399 --> 00:13:54.419 Titles, the 1991 catalog. The one we opened with. 00:13:54.539 --> 00:13:56.899 Yep. Now you can see where it actually operates 00:13:56.899 --> 00:14:00.120 in the framework. It is step five. The vocational 00:14:00.120 --> 00:14:02.659 expert at Kathleen's 2026 hearing can look at 00:14:02.659 --> 00:14:05.519 her file and say she could be a document preparer. 00:14:05.559 --> 00:14:08.139 Document preparer is in the catalog. The description 00:14:08.139 --> 00:14:11.519 is from 1991. This was before widespread internet, 00:14:11.700 --> 00:14:15.149 before most people had email. And the ALJ handling 00:14:15.149 --> 00:14:17.470 hundreds of cases generally isn't going to go 00:14:17.470 --> 00:14:20.289 behind the VE's testimony and litigate whether 00:14:20.289 --> 00:14:23.009 the document preparer is performed the same way 00:14:23.009 --> 00:14:26.929 it was in 1991. Right. Not at this volume. So 00:14:26.929 --> 00:14:29.169 the vocational expert cites the job, the ALJ 00:14:29.169 --> 00:14:31.370 accepts it, and the claim is denied at step five. 00:14:31.590 --> 00:14:34.649 Even after the 2024 cleanup. Right. The cleanup 00:14:34.649 --> 00:14:37.740 wouldn't reach this case. All it did was remove 00:14:37.740 --> 00:14:40.159 jobs that nobody was citing anyway. There's not 00:14:40.159 --> 00:14:43.139 really a lot of National Historical Society operators 00:14:43.139 --> 00:14:45.940 at the national level. The list that actually 00:14:45.940 --> 00:14:48.320 mattered was the 13 occupation one. And for those, 00:14:48.480 --> 00:14:51.019 the agency didn't even prohibit citing it. It 00:14:51.019 --> 00:14:53.379 just added a documentation requirement. At the 00:14:53.379 --> 00:14:55.519 volume we're talking about, a documentation requirement 00:14:55.519 --> 00:14:59.379 can get satisfied. And so the VE says yes, as 00:14:59.379 --> 00:15:01.860 currently performed. And here's my supplementary 00:15:01.860 --> 00:15:04.240 explanation. And the ALJ accepts it and writes 00:15:04.240 --> 00:15:07.279 the denial. So the underlying problem, the catalog 00:15:07.279 --> 00:15:11.100 that dates from the year AOL was born and years 00:15:11.100 --> 00:15:14.200 before most of our students were born. So this 00:15:14.200 --> 00:15:16.299 is being used to tell present day claimants what 00:15:16.299 --> 00:15:18.980 jobs they could do. It's still there. The cleanup 00:15:18.980 --> 00:15:20.779 changed the words that get spoken at the hearing, 00:15:20.899 --> 00:15:23.340 but it didn't change a lot of the outcome. And 00:15:23.340 --> 00:15:25.700 that is because of the grids. They are officially 00:15:25.700 --> 00:15:27.980 called the medical vocational guidelines, but 00:15:27.980 --> 00:15:30.799 everyone calls them the grids. And what are the 00:15:30.799 --> 00:15:34.320 grids? These are literal rows and columns. They 00:15:34.320 --> 00:15:36.779 have four inputs, your residual functional capacity 00:15:36.779 --> 00:15:39.919 level, your age, your education, and your work 00:15:39.919 --> 00:15:42.659 experience. You plug those four values in, you 00:15:42.659 --> 00:15:44.519 read across the row, and the table tells you 00:15:44.519 --> 00:15:47.019 whether you're disabled or not. So the table 00:15:47.019 --> 00:15:49.460 decides the case. For a huge category of step 00:15:49.460 --> 00:15:53.299 five cases, yes, the table decides. So that's 00:15:53.299 --> 00:15:55.700 a core tension from Goldberg, isn't it? The whole 00:15:55.700 --> 00:15:57.679 point of an individualized hearing is that the 00:15:57.679 --> 00:16:00.720 human being considers your particular circumstances. 00:16:01.519 --> 00:16:04.679 But what the grid does is it basically amateurizes 00:16:04.679 --> 00:16:07.179 your disability. It puts a matrix in the middle 00:16:07.179 --> 00:16:10.440 of that. Right. This is before we've had any 00:16:10.440 --> 00:16:12.299 sort of property interest attached. We're deciding 00:16:12.299 --> 00:16:14.500 whether somebody gets something. The due process 00:16:14.500 --> 00:16:17.539 protections don't happen. And this mattered when 00:16:17.539 --> 00:16:20.789 it was challenged. In Heckler v. Campbell, this 00:16:20.789 --> 00:16:24.210 is 1983, Carmen Campbell had a bad back and hypertension. 00:16:24.529 --> 00:16:26.610 She couldn't do her job as a hotel maid anymore. 00:16:26.850 --> 00:16:29.470 The ALJ applied the grids and denied her claim. 00:16:29.690 --> 00:16:31.789 She argued the grids violated her right to an 00:16:31.789 --> 00:16:34.529 individualized determination. Okay, and so what 00:16:34.529 --> 00:16:36.690 happened? The Supreme Court upheld the grids 00:16:36.690 --> 00:16:40.509 here. Ah, that's shocking. I know. The court 00:16:40.509 --> 00:16:43.070 said the inquiry requires the secretary to determine 00:16:43.070 --> 00:16:46.110 an issue that is not unique to each claimant. 00:16:46.409 --> 00:16:48.590 the types and numbers of jobs that exist in the 00:16:48.590 --> 00:16:50.970 national economy. This type of general factual 00:16:50.970 --> 00:16:54.210 issue may be resolved as fairly through rulemaking 00:16:54.210 --> 00:16:57.049 as by introducing the testimony of vocational 00:16:57.049 --> 00:16:59.950 experts at each disability hearing. As the Secretary 00:16:59.950 --> 00:17:02.909 has argued, the use of published guidelines brings 00:17:02.909 --> 00:17:05.250 with it a uniformity that previously had been 00:17:05.250 --> 00:17:07.589 perceived as lacking. To require the Secretary 00:17:07.589 --> 00:17:10.450 to re -litigate the existence of jobs in the 00:17:10.450 --> 00:17:12.730 national economy at each hearing would hinder 00:17:12.730 --> 00:17:15.950 needlessly an already overburdened agency. A 00:17:15.950 --> 00:17:18.750 lot of people will have really standard residual 00:17:18.750 --> 00:17:22.130 functional capacities, RFCs. And so the grid 00:17:22.130 --> 00:17:24.710 works well for those. It's when you're deviating 00:17:24.710 --> 00:17:26.430 from what they define it. That's when we need 00:17:26.430 --> 00:17:30.349 the vocational experts. All right. So that last 00:17:30.349 --> 00:17:32.769 line there, that it would hinder needlessly an 00:17:32.769 --> 00:17:37.109 already overburdened agency. That was 1983. So 00:17:37.109 --> 00:17:39.990 the court is openly saying, we're letting you 00:17:39.990 --> 00:17:42.210 do this because doing it the other way would 00:17:42.210 --> 00:17:45.160 crush the system. Yes. These grids are a deliberate 00:17:45.160 --> 00:17:47.440 trait. Congress and the agency looked at the 00:17:47.440 --> 00:17:50.119 scale of disability claims and decided that some 00:17:50.119 --> 00:17:52.079 part of the individualized determination had 00:17:52.079 --> 00:17:54.359 to give way to a rule -based table or the system 00:17:54.359 --> 00:17:57.880 wouldn't function. So the grids are admin law's 00:17:57.880 --> 00:18:01.200 clearest example of agencies replacing individualized 00:18:01.200 --> 00:18:03.720 judgment with systemized rules for the explicit 00:18:03.720 --> 00:18:06.539 reason of keeping up with volume. Yes. And Campbell 00:18:06.539 --> 00:18:09.519 gave that trait a constitutional blessing. Who 00:18:09.519 --> 00:18:12.200 gets hurt by that trade -off? It's the people 00:18:12.200 --> 00:18:14.740 who don't fit into the table cleanly. And the 00:18:14.740 --> 00:18:17.059 cleanest example of fitting is probably age. 00:18:17.539 --> 00:18:20.779 Kathleen is 54. There is a reason for that. That 00:18:20.779 --> 00:18:23.019 puts her at what the grids call closely approaching 00:18:23.019 --> 00:18:26.460 advanced age, the 50 to 54 bracket. If she were 00:18:26.460 --> 00:18:30.059 55, she'd be in advanced age. So on her exact 00:18:30.059 --> 00:18:33.259 facts, limited education, semi -skilled past 00:18:33.259 --> 00:18:35.980 work with limited transferable skills, light 00:18:35.980 --> 00:18:38.200 work residual functional capacity, the grids 00:18:38.200 --> 00:18:40.000 direct different outcomes for the two brackets. 00:18:40.650 --> 00:18:44.130 How is that different? So at 55, the grids direct 00:18:44.130 --> 00:18:47.329 a finding of disabled. At 54, it directs a finding 00:18:47.329 --> 00:18:50.430 of not disabled. So one year of age is the difference 00:18:50.430 --> 00:18:53.450 between a yes and a no. Yes. Same medical condition, 00:18:53.690 --> 00:18:56.269 same residual functional capacity, same education, 00:18:56.549 --> 00:18:59.250 same work history. The grid treats the move from 00:18:59.250 --> 00:19:02.250 54 to 55 as a structural break in the labor market. 00:19:02.680 --> 00:19:04.799 The working theory is someone closer to retirement 00:19:04.799 --> 00:19:07.059 has a meaningful, harder time learning new work. 00:19:07.160 --> 00:19:09.400 And the regulations bake that theory in as a 00:19:09.400 --> 00:19:12.420 categorical line. That's the grids doing what 00:19:12.420 --> 00:19:16.240 they were designed to do. Predictable, uniform, 00:19:16.480 --> 00:19:20.660 and quite honestly, brutal. Yes. The whole point 00:19:20.660 --> 00:19:23.039 is to remove judgment from this question. The 00:19:23.039 --> 00:19:25.319 cost is a cliff at the category boundary, which 00:19:25.319 --> 00:19:27.140 the regulations themselves noticed eventually 00:19:27.140 --> 00:19:31.309 and created a rule for it. Here, this is 20 CFR 00:19:31.309 --> 00:19:35.450 404 -1563. We will not apply the age categories 00:19:35.450 --> 00:19:38.450 mechanically in a borderline situation. If you 00:19:38.450 --> 00:19:41.329 are within a few days to a few months of reaching 00:19:41.329 --> 00:19:44.220 an older age category. and using the older age 00:19:44.220 --> 00:19:47.220 category would result in a determination or decision 00:19:47.220 --> 00:19:49.880 that you are disabled, we will consider whether 00:19:49.880 --> 00:19:52.279 to use the older age category after evaluating 00:19:52.279 --> 00:19:55.319 the overall impact of all the factors in your 00:19:55.319 --> 00:19:58.920 case. So if Kathleen is 54 and 10 months, the 00:19:58.920 --> 00:20:02.960 ALJ can then just round up her age to 55. The 00:20:02.960 --> 00:20:05.880 ALJ can consider rounding up. This rule is permissive. 00:20:05.880 --> 00:20:08.859 It is not automatic. And the agency reads a few 00:20:08.859 --> 00:20:11.509 days to a few months really narrowly. Howeck's 00:20:11.509 --> 00:20:13.529 guidance treats six months as the outer edge. 00:20:13.670 --> 00:20:16.650 And in 2025, the Fifth Circuit affirmed a denial 00:20:16.650 --> 00:20:19.069 where the claimant was five months shy at 55. 00:20:19.609 --> 00:20:22.690 Five months shy of 55 at his onset date. Five 00:20:22.690 --> 00:20:25.670 months wasn't close enough. Five months? Wow. 00:20:25.849 --> 00:20:28.269 The court held the ALJ wasn't required to apply 00:20:28.269 --> 00:20:30.930 the borderline rule. So the rule exists, but 00:20:30.930 --> 00:20:32.710 the agency's working definition of borderline 00:20:32.710 --> 00:20:34.769 is tight enough that a substantial number of 00:20:34.769 --> 00:20:36.549 claimants who are close to the next category 00:20:36.549 --> 00:20:39.910 still get their current chronological age outcome. 00:20:40.440 --> 00:20:44.180 So, Kathleen, at 54, how close to 55 does she 00:20:44.180 --> 00:20:47.079 need to be to get pushed over the edge? All right. 00:20:47.259 --> 00:20:49.819 54 and 11 months, she should be inside the borderline 00:20:49.819 --> 00:20:51.940 window. Any lawyer would argue for the older 00:20:51.940 --> 00:20:54.859 category and probably get it. 54 and 7 months? 00:20:55.079 --> 00:20:57.380 Well, we're really at the outer edge here. Some 00:20:57.380 --> 00:21:00.319 ALJs would apply it. Plenty wouldn't. The circuits 00:21:00.319 --> 00:21:02.319 are split on whether the failure to address this 00:21:02.319 --> 00:21:05.299 issue is itself reversible error. If she's 54 00:21:05.299 --> 00:21:07.339 and 2 months, the borderline rule isn't going 00:21:07.339 --> 00:21:09.730 to help her. She has to win on the merits at 00:21:09.730 --> 00:21:12.109 light work residual functional capacity in front 00:21:12.109 --> 00:21:14.309 of a vocational expert who's about to cite document 00:21:14.309 --> 00:21:18.309 preparer. So the grid's uniformity has a softening 00:21:18.309 --> 00:21:20.829 rule built into it. And the softening rule has 00:21:20.829 --> 00:21:23.289 its own categorical line. And Kathleen is on 00:21:23.289 --> 00:21:26.390 the wrong side of that one, too. Right. It's 00:21:26.390 --> 00:21:28.089 worth noting that somebody can also partially 00:21:28.089 --> 00:21:31.009 win. The ALJ might say they are disabled as of 00:21:31.009 --> 00:21:34.029 their 55th birthday. But this is how the system 00:21:34.029 --> 00:21:36.150 handles its own brittleness. It adds a rule. 00:21:36.859 --> 00:21:38.700 The rule has its own edges. The cliff doesn't 00:21:38.700 --> 00:21:41.619 go away. It just moves. Okay. We spent three 00:21:41.619 --> 00:21:43.539 episodes on how the structural picture works. 00:21:44.019 --> 00:21:46.539 How does it actually land here? I mean, what 00:21:46.539 --> 00:21:48.339 does this trilogy look like when you run it through 00:21:48.339 --> 00:21:50.980 two and a half million cases per year? So you 00:21:50.980 --> 00:21:53.019 get the numbers we were looking at in that first 00:21:53.019 --> 00:21:56.160 episode. The ALJ approval rates can range from 00:21:56.160 --> 00:22:00.200 below 20 % to above 80%. Same agency, same regulations, 00:22:00.460 --> 00:22:03.339 same five steps, same grid, which means that 00:22:03.339 --> 00:22:05.319 two claimants with essentially identical files 00:22:05.319 --> 00:22:07.299 can get different outcomes based on which judge 00:22:07.299 --> 00:22:09.539 they happen to draw. All right, we covered that 00:22:09.539 --> 00:22:11.839 as a structural claim, but what does it look 00:22:11.839 --> 00:22:14.480 like for Kathleen? Last week, we talked about 00:22:14.480 --> 00:22:17.359 Judge Williams. This was the approval rate around 00:22:17.359 --> 00:22:20.950 the office median. Disc disease claim that she 00:22:20.950 --> 00:22:23.009 had approved and the quality review memo three 00:22:23.009 --> 00:22:25.009 months later that taught her something about 00:22:25.009 --> 00:22:27.869 what the system considers adequate. So if we 00:22:27.869 --> 00:22:31.049 put Kathleen in front of her in 2026, after a 00:22:31.049 --> 00:22:33.609 decade of those memos accumulating, whereas if 00:22:33.609 --> 00:22:36.809 we put her in 2010 before any of them, same medical 00:22:36.809 --> 00:22:38.910 record, same testimony, this would even be the 00:22:38.910 --> 00:22:41.309 same ALJ, but we could have a different system 00:22:41.309 --> 00:22:44.170 around her. And the outcome then also moves. 00:22:44.430 --> 00:22:46.930 Right. And it can move pretty dramatically. That's 00:22:46.930 --> 00:22:49.299 what we were talking about last week. Because 00:22:49.299 --> 00:22:52.279 Judge Williams in 2096 has absorbed a decade 00:22:52.279 --> 00:22:54.480 of guidance, training, and quality review feedback 00:22:54.480 --> 00:22:57.380 that the Judge Williams of 2010 hadn't seen yet. 00:22:57.519 --> 00:23:00.619 Same judge, same evidence, but the system around 00:23:00.619 --> 00:23:03.059 the judge tilted. Now let's hold the year steady 00:23:03.059 --> 00:23:06.039 and run the other axis. She's in 2026 with two 00:23:06.039 --> 00:23:08.000 different judges. She testifies she can only 00:23:08.000 --> 00:23:10.079 stand for about 20 minutes before the pain gets 00:23:10.079 --> 00:23:12.319 severe and that she has to lie down a couple 00:23:12.319 --> 00:23:14.039 of times during the day and her pain medication 00:23:14.039 --> 00:23:16.539 makes her foggy. The objective medical evidence, 00:23:16.660 --> 00:23:18.460 the MRI, is the same in front of both judges. 00:23:18.740 --> 00:23:21.400 But? But Judge A finds her testimony credible. 00:23:22.119 --> 00:23:24.299 The residual functional capacity ends up more 00:23:24.299 --> 00:23:27.220 restrictive. With her age and education and limited 00:23:27.220 --> 00:23:29.779 transferable skills from the warehouse job, she 00:23:29.779 --> 00:23:31.920 can land in a category where the - Grids direct 00:23:31.920 --> 00:23:34.680 to find disabled and she gets approved. Judge 00:23:34.680 --> 00:23:36.859 B might find the same testimony only partially 00:23:36.859 --> 00:23:40.059 credible. And so we'd come out as a residual 00:23:40.059 --> 00:23:42.779 functional capacity of light work with some qualifications. 00:23:43.059 --> 00:23:45.539 The grids say not disabled and she's denied. 00:23:45.819 --> 00:23:48.359 OK, so the same woman with the same medical record, 00:23:48.460 --> 00:23:50.500 but with different outcomes because of different 00:23:50.500 --> 00:23:53.220 years. So that was different outcomes because 00:23:53.220 --> 00:23:55.599 of different judges. OK, same woman with the 00:23:55.599 --> 00:23:58.000 same medical record, but different outcomes because 00:23:58.000 --> 00:24:00.210 of different judges. Right. And that judge who 00:24:00.210 --> 00:24:01.849 denied her wouldn't have inherently done anything 00:24:01.849 --> 00:24:04.210 wrong. They might have written a decision citing 00:24:04.210 --> 00:24:06.509 the medical evidence and explained why they found 00:24:06.509 --> 00:24:09.390 the testimony only partially credible and then 00:24:09.390 --> 00:24:11.450 apply the grids. On review, that credibility 00:24:11.450 --> 00:24:14.190 finding gets substantial deference. So all of 00:24:14.190 --> 00:24:15.710 the structural features we talked about across 00:24:15.710 --> 00:24:18.450 the trilogy, the horizontal variation, the weak 00:24:18.450 --> 00:24:21.190 internal review, the ex -ante mechanisms, they 00:24:21.190 --> 00:24:23.910 all run through the credibility finding and the 00:24:23.910 --> 00:24:26.690 RFC, which determine what the grids then output. 00:24:27.259 --> 00:24:29.880 Right. This is why social security outcomes look 00:24:29.880 --> 00:24:32.400 the way they do. The structural picture isn't 00:24:32.400 --> 00:24:34.319 abstract. We have this five -step framework, 00:24:34.480 --> 00:24:36.559 and the grids give it a specific funnel at the 00:24:36.559 --> 00:24:39.279 end. Credibility determines what evidence counts. 00:24:39.700 --> 00:24:41.859 That leads to what you have as a residual functional 00:24:41.859 --> 00:24:44.700 capacity, and that determines where we land on 00:24:44.700 --> 00:24:47.400 the grids. So whatever is shaping the ALJ's approach 00:24:47.400 --> 00:24:49.480 to credibility and the residual functional... 00:24:49.740 --> 00:24:52.359 capacity, like their experience in the local 00:24:52.359 --> 00:24:54.519 office culture and quality review environment, 00:24:54.779 --> 00:24:57.480 those shape the outcomes. And the data on claimants 00:24:57.480 --> 00:25:00.599 who don't make it to the end? So the GAO and 00:25:00.599 --> 00:25:02.559 Social Security's own inspector general have 00:25:02.559 --> 00:25:04.700 both reported on claimants dying while waiting 00:25:04.700 --> 00:25:07.400 for a hearing, which makes sense. This is a process 00:25:07.400 --> 00:25:11.299 people only start when they're not 100%. So at 00:25:11.299 --> 00:25:13.579 current processing times with a nine -month average 00:25:13.579 --> 00:25:17.069 and some offices running longer than that, the 00:25:17.069 --> 00:25:18.630 number of people who die during the pendency 00:25:18.630 --> 00:25:20.109 of their claim is going to be meaningful every 00:25:20.109 --> 00:25:22.390 year. And that's not the system failing in some 00:25:22.390 --> 00:25:24.869 exotic way. That's the system working as described 00:25:24.869 --> 00:25:27.970 at volume with a better wait time than it used 00:25:27.970 --> 00:25:30.890 to be. Okay. Everything we built this season 00:25:30.890 --> 00:25:33.990 shows up in this one system. Let's walk through 00:25:33.990 --> 00:25:38.789 how it does. What a great idea. Here's the doctrinal 00:25:38.789 --> 00:25:42.079 point anchoring all of it. Once disability benefits 00:25:42.079 --> 00:25:44.700 are awarded, they are a statutory entitlement 00:25:44.700 --> 00:25:46.940 that the Constitution is going to protect. That 00:25:46.940 --> 00:25:49.299 was Goldberg, Roth, and then Matthews itself. 00:25:49.460 --> 00:25:51.559 If you have benefits and the government wants 00:25:51.559 --> 00:25:54.359 to take them away, due process attaches. You're 00:25:54.359 --> 00:25:56.200 entitled to notice. You're entitled to be heard. 00:25:56.420 --> 00:25:58.640 The Constitution has things to say about how 00:25:58.640 --> 00:26:02.079 that has to happen. And once it's awarded, that's 00:26:02.079 --> 00:26:04.839 the termination cases? Right. So Matthews v. 00:26:04.859 --> 00:26:06.799 Eldridge was a termination case. He was already 00:26:06.799 --> 00:26:08.740 receiving benefits when Social Security moved 00:26:08.740 --> 00:26:11.150 to cut them off. That's when the three -factor 00:26:11.150 --> 00:26:13.369 balancing test got applied. What the court was 00:26:13.369 --> 00:26:15.750 deciding was how much process was constitutionally 00:26:15.750 --> 00:26:17.650 required before the government could take away 00:26:17.650 --> 00:26:20.410 something elders already had. Okay, and the initial 00:26:20.410 --> 00:26:22.950 application is different. Constitutionally different. 00:26:23.150 --> 00:26:25.470 So the first -time applicant doesn't have a property 00:26:25.470 --> 00:26:27.670 interest. There's no entitlement to a benefit 00:26:27.670 --> 00:26:29.990 you haven't been awarded. You have a statutory 00:26:29.990 --> 00:26:32.130 right to whatever process Congress put in the 00:26:32.130 --> 00:26:35.369 statute. Paper review, reconsideration, the ALJ 00:26:35.369 --> 00:26:37.710 hearing, appeals counsel, eventually federal 00:26:37.710 --> 00:26:39.950 court. But those are procedural rights Congress 00:26:39.950 --> 00:26:42.490 chose to give. The Constitution isn't requiring 00:26:42.490 --> 00:26:44.589 them. It's not requiring much of anything at 00:26:44.589 --> 00:26:46.730 the initial stage. So the two and a half million 00:26:46.730 --> 00:26:49.410 initial applications aren't getting a constitutional 00:26:49.410 --> 00:26:51.970 hearing. They're getting a statutory one. Yes. 00:26:51.970 --> 00:26:53.970 And it's going to be thinner. And that's a design 00:26:53.970 --> 00:26:57.559 choice that is driven in part by volume. So if 00:26:57.559 --> 00:27:00.000 we run Matthews over it, not for those initial 00:27:00.000 --> 00:27:02.220 applications, but for what happens when someone's 00:27:02.220 --> 00:27:04.759 in the system and the agency wants to deny continuing 00:27:04.759 --> 00:27:07.539 benefits or try and terminate them or recoup 00:27:07.539 --> 00:27:09.819 overpayments, those are the cases where the three 00:27:09.819 --> 00:27:12.460 -factor test actually does work. So Matthews 00:27:12.460 --> 00:27:14.619 balances the private interest, the risk of erroneous 00:27:14.619 --> 00:27:16.980 deprivation with additional procedure, and the 00:27:16.980 --> 00:27:19.140 government's administrative burden. Right. We 00:27:19.140 --> 00:27:21.660 plug the numbers in. The private interest, pretty 00:27:21.660 --> 00:27:23.299 big. These are people losing benefits they've 00:27:23.299 --> 00:27:25.059 been living on. I would say enormous, but the 00:27:25.059 --> 00:27:26.880 court has said. It is inherently not enormous. 00:27:27.200 --> 00:27:30.640 The risk of erroneous deprivation. It's demonstrably 00:27:30.640 --> 00:27:33.019 real. The GAO has documented that it's real. 00:27:33.279 --> 00:27:35.799 And the administrative burden of more individualized 00:27:35.799 --> 00:27:38.740 process. This is where the burden can get staggering. 00:27:39.519 --> 00:27:41.839 Continuing disability reviews alone run somewhere 00:27:41.839 --> 00:27:44.380 between one and a half and two million a year. 00:27:44.539 --> 00:27:47.279 So overpayment cases, there's hundreds of thousands 00:27:47.279 --> 00:27:49.980 more of those. The post -award population that 00:27:49.980 --> 00:27:52.660 Matthews actually governs is its own multi -million 00:27:52.660 --> 00:27:56.019 case per year system. And when you balance those 00:27:56.019 --> 00:27:59.359 three things honestly at that volume, you tend 00:27:59.359 --> 00:28:02.059 to get this. Yes, you get these truncated termination 00:28:02.059 --> 00:28:05.019 procedures. You get CDR processes that mostly 00:28:05.019 --> 00:28:07.680 run on paper, overpayment recruitment that starts 00:28:07.680 --> 00:28:09.799 before the claimant has meaningfully been heard, 00:28:09.920 --> 00:28:12.640 and an appellate path that can take years. The 00:28:12.640 --> 00:28:14.779 Constitution is in the room applying Matthews, 00:28:14.779 --> 00:28:16.839 and Matthews is doing what it does, which is 00:28:16.839 --> 00:28:19.200 ballots. And what it produces at this volume 00:28:19.200 --> 00:28:21.259 is a process that's much thinner than you might 00:28:21.259 --> 00:28:24.089 imagine. Matthews itself was a Social Security 00:28:24.089 --> 00:28:26.970 disability case. The court knew it was setting 00:28:26.970 --> 00:28:29.329 the rules for federal benefits program processing 00:28:29.329 --> 00:28:31.809 hundreds of thousands of terminations a year. 00:28:32.250 --> 00:28:35.509 It wasn't sitting there imagining one isolated 00:28:35.509 --> 00:28:38.190 termination case. Right. The court built the 00:28:38.190 --> 00:28:40.230 administrative burden into the test as the third 00:28:40.230 --> 00:28:42.450 factor precisely because it was thinking about 00:28:42.450 --> 00:28:45.109 these high volume benefit systems. Scale isn't 00:28:45.109 --> 00:28:47.309 an accident in Matthews. It is part of the intentional 00:28:47.309 --> 00:28:50.250 design. And so the third factor was designed 00:28:50.250 --> 00:28:53.599 to then count the volume. Yes. And it was designed 00:28:53.599 --> 00:28:55.640 to do that from the start. But what has changed 00:28:55.640 --> 00:28:59.319 since 1976 is the ratio. The disability adjudication 00:28:59.319 --> 00:29:01.779 system today is an order of magnitude bigger 00:29:01.779 --> 00:29:04.019 than the one the court was looking at. The initial 00:29:04.019 --> 00:29:06.420 applications, the CDRs, the terminations, the 00:29:06.420 --> 00:29:09.259 overpayments, same test, but a much bigger denominator. 00:29:10.029 --> 00:29:12.569 So the doctrinal question isn't whether Matthews 00:29:12.569 --> 00:29:15.170 was misapplied. It's whether a test that counts 00:29:15.170 --> 00:29:17.930 administrative burden as a factor at all can 00:29:17.930 --> 00:29:20.130 ever produce a result that requires more process 00:29:20.130 --> 00:29:22.730 once the system is big enough that the burden 00:29:22.730 --> 00:29:25.710 dominates the math. Yeah. If one of those factors 00:29:25.710 --> 00:29:28.049 is going to be the administrative burden, then 00:29:28.049 --> 00:29:30.450 eventually at a sufficient scale, that factor 00:29:30.450 --> 00:29:33.430 will tend to outweigh or almost outweigh everything 00:29:33.430 --> 00:29:35.750 on the other side. Not because the test is being 00:29:35.750 --> 00:29:37.650 misapplied, but because this is the way it was 00:29:37.650 --> 00:29:40.460 built. The court accepted scale as a legitimate 00:29:40.460 --> 00:29:42.619 counterweight from the beginning. And the test 00:29:42.619 --> 00:29:45.799 doesn't have a built -in floor. Right. It tells 00:29:45.799 --> 00:29:47.660 you you need to balance. It doesn't tell you 00:29:47.660 --> 00:29:50.000 that some minimum amount of process is required 00:29:50.000 --> 00:29:52.519 regardless of what the balance produces. There's 00:29:52.519 --> 00:29:55.200 no real doctrinal limit on how thin process can 00:29:55.200 --> 00:29:57.140 become if the system is big enough to make the 00:29:57.140 --> 00:30:00.319 burden factor dominant. And let's be clear, what 00:30:00.319 --> 00:30:02.500 we've been talking about so far this episode 00:30:02.500 --> 00:30:05.660 is the largest administrative adjudication system 00:30:05.660 --> 00:30:08.410 in this country. It is. It is big, but it is 00:30:08.410 --> 00:30:11.990 not the only big one. So next episode, the last 00:30:11.990 --> 00:30:14.769 official episode of the season until slaughter 00:30:14.769 --> 00:30:17.490 comes out, is the other major deep dive we're 00:30:17.490 --> 00:30:20.029 doing on immigration adjudication. And specifically, 00:30:20.089 --> 00:30:22.650 we're talking about asylum. It's a different 00:30:22.650 --> 00:30:24.829 system, a different structure, and the constitutional 00:30:24.829 --> 00:30:27.650 stakes are higher. Social security is about income. 00:30:27.930 --> 00:30:29.589 Immigration is about whether you get to stay 00:30:29.589 --> 00:30:31.769 in the country. You might accept the procedural 00:30:31.769 --> 00:30:34.619 protections would be stronger there, but. Surprise, 00:30:34.700 --> 00:30:36.839 they are not. They are weaker, and the volume 00:30:36.839 --> 00:30:40.319 problems get worse, not better. So that does 00:30:40.319 --> 00:30:42.039 it for today's episode on administrative remedies. 00:30:42.220 --> 00:30:45.140 Thank you for joining us today. Please, if you 00:30:45.140 --> 00:30:47.519 enjoy this podcast and enjoy this episode, give 00:30:47.519 --> 00:30:51.319 us a like on Spotify, iTunes, or whatever platform 00:30:51.319 --> 00:30:54.180 you're listening on. And be sure to tune in next 00:30:54.180 --> 00:30:56.339 time where we'll continue to dive into the contours 00:30:56.339 --> 00:30:58.559 of administrative law, because remember, you 00:30:58.559 --> 00:31:00.200 can't fix what you don't understand.