# Oral argument — North America's Building Trades Unions v. Occupational Safety & Health Administration

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/argument%3A31961

## Record

- **Collection:** Oral argument
- **Document type:** Oral argument
- **Published:** September 26, 2017

## Text

Case number 16-1105 et al., North America's Building Trade Unions Petitioner v. Occupational Safety & Health Administration et al. Okay, so we have a very complicated set of stage directions. I'm hoping you guys can all figure out when to stand up and when not. A significant risk, we begin with industry petitioners. Good morning. May it please the court, my name is Bill Wareham. I'm here to address the issue of significant risk. On this issue, I will be splitting time with my colleague, Mike Conley. Mike will be talking about so-called surveillance data. I'll be talking about everything else related to this issue. I've requested two minutes of rebuttal time. I'm here on behalf of Petitioner's National Stone, Sand, and Gravel Association and the Brick Industry Association. I also represent Petitioner Intervenors, Portland Cement Association, and the National Concrete Masonry Association. First of all, we do not dispute that respirable silica can cause disease. This has been known for centuries. The question here is the level of exposure at which respirable silica can cause disease. Specifically, the burden that OSHA bears is to show significant risk of material health impairment at the prior level of the standard, 100 micrograms, and significant improvement at the new level of the standard, which was set at 50. OSHA asserts it's taken an objective look at the data. It asserts that the weight of the evidence supports a finding that silica exposure creates significant risk, not just at the level of the prior standard, but even at the level of the new standard. In fact, the reach of the new rule is limited only by OSHA's findings that it's not feasible to achieve workplace exposure levels below the level of the new standard of 50. In contrast, we assert that OSHA had a thumb on the scale. We believe the record makes clear that OSHA came to this rulemaking with a determined goal of reducing the level of the standard. We believe the results oriented perspective clouded OSHA's judgment and caused it to lose objectivity, which we believe permeates the entire proceeding. Do you dispute that, according to Robbins, the Pathologic Basis of Disease 2015, that silicosis is the most prevalent chronic occupational disease in the world? Your Honor, we do not dispute the fact that respirable silica, in fact, can cause a variety of lung diseases, including silicosis. So, again, the question, Your Honor, is what is the level at which that should be expected to occur, the level at which significant risk of material impairment is expected to occur? You know, you say OSHA had a thumb on the scale, which is a curious statement to me given our standard of review. I mean, the question here is, is there sufficient evidence in the record to support OSHA's position? You can certainly point to contrary evidence, but OSHA has explained all of that. And, you know, from our perspective, it's kind of in, isn't it? Yes, Your Honor. You know, you have to make your argument in terms of our specific standard of review, which is a substantial evidence question. Our case law is very clear about that. Absolutely, Your Honor. I agree completely. So what's your best argument? The first best argument is renal disease, Your Honor. Which is kidney disease. In the record, OSHA committed dozens and dozens of pages and hundreds of pages of reporting documents to the proposition that exposure to respirable silica, in fact, causes kidney disease. Suppose the other two are okay. Do you have to show that all three of the diseases that OSHA looked at are unsupported? Or don't you have to show that OSHA's judgment on all three were unsupported, not just one? Yes, Your Honor. If one is supported, that's enough to sustain it. Yes, Your Honor, I agree with that. Okay. So to work through them, so you're absolutely right. So I'm setting the stage to try to create, to give a perspective on how we think OSHA viewed the evidence, because you're right. OSHA committed dozens and dozens, hundreds of pages to the proposition that what it's done here is supported. And what we're asking the court to do is take a hard look at the justification that OSHA put forward. And we believe that when you take a hard look, that, in fact, their proof does not rise to the level of substantial evidence. And, again, renal kidney disease is the very best first example, because the agency vociferously defended the position that silica exposure can result in kidney disease, reached that conclusion definitively in the final rule. And in the context of this litigation, in the full light of day, OSHA chose not to defend that position. So we believe that speaks volumes and, you know, dispenses of that particular disease endpoint, because they chose not to defend. But we think it's indicative of an issue that pervades the rest of the record, which is a failure to show substantial evidence and support the other disease endpoints. So that, Your Honor, that's renal, and you're absolutely right. OSHA set out several other disease points separately for the proposition that, you know, individually those, you know, their findings on those disease endpoints support the standard that they set. And in our brief, one by one by one, we go through each disease endpoint and explain why that's not the case. The second best example, Your Honor, is the so-called no effect threshold. And what we submitted evidence for the record that we believe conclusively demonstrated that there's a level of exposure below which respiratory disease related to silica exposure should not be expected to occur. And our evidence came in four pieces. The first piece is real world evidence that we pointed to, which Mr. Connolly is going to speak to a bit more here in a second. But when you look at what happened after the prior standard was set in 1971, the one that this rule replaced, the incidence of silica disease plummeted in the workforce in the United States by more than 90 percent over the period of time since the standard was put in place. So we believe that when you look at that public health experience, first of all, it raises the question as to whether anything needed to be done other than let that standard continue to work its magic, because it was doing a great job. But secondly, when you look at that experience, you have to ask yourself from a common sense standpoint, can there be effects at low levels as OSHA contends when there continues to be a reduction in disease incidence? The trend continues to this day. And, you know, common sense, just looking at the situation, stepping back from the guts of the analysis, says it seems to be working and low levels of silica exposure do not appear to be causing a real problem of disease in the population, in the worker population. So that's point one. Point two is we believe, and we set forth that we believe there's a disease mechanism that amply explains the existence of a threshold, which is people are designed to deal with dust. People are in dusty environments all the time, and it doesn't kill them. And part of the reason why is the lung has a mechanism available to it to grab onto the dust and to eventually remove it from the lung or to encapsulate it in a way that it's not harmful. So there are mechanisms, physiological mechanisms in the lung that are designed to deal with dusty environments, and we believe those mechanisms, as long as there's not an overabundance of silica exposure, are sufficient to prevent the silica disease that OSHA contends can occur at very low levels. So there's a plausible physiological mechanism, we believe. Is it enough that there's a plausible physiological mechanism? There's a dispute, a scientific dispute about the question of threshold, isn't there? Your Honor, it's necessary but not sufficient. There is a scientific dispute on this question, isn't there? OSHA would have you believe that it's a jump ball, and they probably wouldn't even agree with that, but I would characterize it as a jump ball. There are other scientists who believe the opposite of you, is that right? Correct. And how are we supposed to, you have what's described as almost a metaphorical analysis of how the lungs are going to filter out silica, which we are not really capable of evaluating. But we have scientists on both sides, and the law here is clear that when there are scientists on both sides, that OSHA is permitted to pick the ones that are most likely to protect worker safety. There's supposed to be a thumb on the scale in terms of worker safety, isn't that right? That's what the cases say. That's what our own case says, that it's perfectly appropriate for OSHA to wait in favor of worker safety. That's right, isn't it? Correct, Your Honor, to a point. And so that doesn't insulate OSHA from review. No, and that's what we're doing here. But it's not enough just to say there's a plausible mechanism. You have to be able to show that OSHA's studies are just not themselves substantial evidence. Correct, Your Honor. And there were two further points, and unfortunately this is very complicated stuff. I'm trying to hit the highlights in my limited time here. But I said mechanism is sufficient or necessarily not sufficient to prove the point I'm trying to make. A third lag on our stool, and we have a four-legged stool here, is the existence of so-called dose rate effect. And there are studies in the record that show a greater than proportional increase in health risk as exposure increases. So it's not just linear. You would expect if you get an increment more of exposure to have an increment more of risk. But the dose rate issue has to do with if you get an increment more of exposure to get more than an increment of additional risk. So is it not proportional? Is it more than proportional? And so dose rate effect is sufficient but not necessary. But it's a critical part of our argument because it shows that you should expect some difference in health effect as exposures change, and especially as exposures go up. And there's definitive evidence in the record that demonstrates the existence of a dose rate effect, which OSHA brushes off by saying, yeah, but that level is much higher than the standard we set, so it tells you nothing about what happens at the level of the standard. And we think that's an arbitrary rationale for setting aside these decisions. And the very last argument is evidence of a threshold itself using the epidemiological data. And Dr. Moorfield, on behalf of my client, submitted a study to the record where he did statistical analysis on an available cohort of data showing that the best fit statistically for the results from that data are that there's a threshold of exposure below which you shouldn't expect effect. So when you put those pieces together, you know, applying some common sense, thinking that there's a plausible mechanism, seeing that there's a dose rate effect, and seeing that there's good science that supports the existence of a threshold, we think that's very, that is substantial evidence for the proposition that a threshold exists. And, of course, OSHA disagrees with that, and they rebut each of the points that I made in the record. But I think the most telling thing is the very end in the final rule preamble where they state their conclusion on this threshold issue, this is a JA75, they say if OSHA were to accept our claim that exposure at 100 micrograms is safe, because we believe the threshold is above the level of the new standard, and if that claim is, in fact, erroneous, the consequences of that error to silica-exposed workers would be great. So we think that speaks volumes because, yes, OSHA is trying to make a health-protective decision here, but it's not a health-protective decision based on substantial evidence. It's a health-protective decision made out of concern that they might be wrong, and that's not substantial evidence. Your Honor, I see I've run over time. There's one further point I'd like to make. Your record is your rebuttal time. You're free to use it up if you want to, but that's your choice. One other example I'd like to raise is for the brick industry, Your Honor. And this, we submitted information to the record indicating that silica exposure in the brick industry is different than silica exposure in other industries, and the reason is the silica is caught up in clay, so it's occluded is the word that's used. So it's, the silica itself is harder to come into contact with the lung. So we, and OSHA and the record agreed that the silica in the brick industry is different and should present lower risk, but the study that we submitted to the record in support of the proposition that there should be, you know, OSHA should, there should be no risk at the level of prior standard, OSHA rejected that study on the grounds that it didn't meet their criteria for a valid study. The problem here is then OSHA turned around and said, but nevertheless, we're going to look at that study and we're going to rely on that study to conclude that there continues to be significant risk from exposure to this industry, and this industry should continue to be covered by the general industry standard. So we believe that's not substantial evidence, because on the one hand, they reject our study, you know, on the grounds that it doesn't meet their validity criteria, and on the other hand, they turn around and use that study to support their finding that there's significant risk here. That, we believe, is patently arbitrary and supports a remand of that issue. Thank you, Your Honor. I think we have Mr. Connelly next. Yes, sir. May I please court, Michael Connelly for the Chamber of Commerce. We were given three minutes to talk about one issue, and that is the importance of CDC government surveillance data, which is a comprehensive record of the number of silicosis deaths that occur in this country each year. We focus on this data because we think it fundamentally undermines OSHA's significant risk. And we think it does so in two ways. First, since 1971, when the prior regime went into effect, there's been an enormous decline in the number of silicosis deaths in this country, according to this data. It's fallen by more than 90 percent, which shows that the prior regime was working and was effective. The second thing that this data shows and that OSHA agrees is that of the deaths that were occurring, many can be attributed to exposures above the prior limit, either because they were exposed before 1971, before the prior limits went into effect, or they were exposed in violation of the prior standard. And so the new rule that OSHA has put into effect wouldn't have done anything to prevent those silicosis deaths. OSHA's main response to this, to the CDC data, is that it underreports. But this really is pure speculation. OSHA concedes in the preamble that it really has almost no empirical evidence of how much underreporting is occurring. Well, is it true, as OSHA says, that the death certificate information does not include information about exposure levels? That is correct. That's actually true, right? That's correct. So then why isn't it logically followed that, therefore, it doesn't really relate to the question of risk assessment? Because the CDC data has been recording the number of silicosis deaths under the prior regime, and the prior regime was 100 micrograms per cubic meter. And even if you assume that all of those deaths were attributable to an insufficient standard, which clearly isn't the case, it still doesn't rise to the level of significant under the benzene standard, given that there are 2.3 million people that are exposed to silica in the workplace, according to OSHA's own data. And that is the main takeaway, I think, from the CDC data, is we have real-world evidence of the effect. I'm not sure I'm following. You keep calling it CDC data. This is just the collection of death certificates, right? CDC hasn't actually examined the bodies or done any of its own analysis. It's just collected death certificates. Correct. This is a partnership between the CDC, the federal government, and the state. That is a record of death certificates. And is the right division, the number that you're dividing, the total number of deaths that are reported on the death certificates by the total number of workers in industry, or is the right number the total number of deaths at a certain level of exposure? That is in terms of the 1 in 1,000 tests. Sure. That's not supposed to be just 1 over the entire population of the United States or 1 over everybody who works. It's supposed to be 1 over 1,000 people who work at a certain exposure level. Isn't that right? For what the meaning of significant is. Sure, that's correct. So then the number you're talking about, which is not, I think this is following up on what Judge Taylor was saying, the number you're talking about is not the percentage at an exposure level. It's the percentage of all workers in industry, some of whom may have no exposure whatsoever to silicon, right? Correct. So I think it is relative to the number of people exposed to silicon, and that's 2.3 million. Corey Nosh's calculus isn't exposed to silicon at a certain exposure level. That matters not all people who may have been exposed to silicon. You could break it down that way too, and those numbers I believe were in the NIOSH report and were in the preamble. There are, and I believe there are hundreds of thousands, that were exposed above the prior limit. So I think those numbers do exist, that even if you reject the 2.3 million, that it's not an insignificant number. So OSHA gave two other reasons for being unpersuaded by this, and I want to ask you about both of them. One was that they said, according to record evidence, that decline resulted from decline in employment in heavy industry, not from improved working conditions. That was one of their reasons. What do you think about that? So there is no sort of nationwide study on comparing the decline in workforce versus the number of deaths that have declined. Well, they said record evidence for this. Yes, and that comes from one comment from Dr. Rosenbaum. It was, to be frank, back of the envelope calculation. He compared certain decline in workforce in Michigan to the number of deaths declining in the United States, and he ballparked that maybe they might be related. And the second point is that... No, no, no. Is it true that employment in heavy industry has, in fact, declined? Industries where silica dangers are greatest. That's actually true. It is true that the number of... Okay. So what's the problem with the agency concluding from that that, you know, it's logical to think that maybe these deaths, declining death rates, come from that and not necessarily better working conditions? Is that a logical... What's the matter with that? The fundamental problem with that is that they're making that argument to suggest that the prior rule was not effective, but the CDC was found that it was effective. In the preamble, they found that it was effective. They argue it can be more effective. But to simply argue that these deaths, the decline in deaths, was solely because of declining workforce is just not supported. Yeah, but what about... Is OSHA wrong when they said that the decline in mortality has leveled off since 2000 and they concluded from that that, you know, maybe the effect of the old, of the prior rule was its effectiveness has ended? Yes. That's factually true, correct? No, I don't believe so. It's not true that the decline in mortality has leveled off since 2000? No, I don't believe so. The data is in the record. If you look at it, decade over decade there's been a decline. Now, OSHA cherry-picked one year, last year, 2015, but 2014 was the lowest year on record. They focused on a slight uptick in 2015, but even 2015 was the fifth lowest year on record. So, and the data is in the record, so you can see there has been no leveling off. Okay, further questions? Thank you. Respondents, Ms. Lindberg for OSHA. May it please the Court. Good morning, Your Honors. My name is Kristin Lindberg and I'm here on behalf of the Secretary of Labor on the issue of risk. Your Honors, OSHA's risk assessment for silica is thorough, reasonable, and based on the best available evidence in the record. Industry petitioners scatter shot the facts on almost every aspect of the analysis do not disturb OSHA's findings. In the ethylene oxide case, this Court heard strikingly similar claims from the Association of Ethylene Oxide Users. Here's what the Court said. The Association's attack on the evidence in this rulemaking fundamentally misconstrues the role that OSHA and reviewing courts each must play in regulating cost and exposure. The Association attacks each piece of evidence suggesting that no individual piece proves the relationship between ethylene oxide exposure and various adverse health effects. This approach disregards the marginal contribution that each piece of evidence makes to the total picture. While some of OSHA's evidence suffers from shortcomings, such incomplete proof is inevitable when the agency regulates on the frontiers of scientific knowledge. Here, as Counsel for Industry Petitioners acknowledged, scientists have known for centuries that silica causes illness and death. And this standard itself is decades in the making. The National Institute for Occupational Safety and Health recommended this silica pell of 50 micrograms per meter cubed in 1974. OSHA officially began this rulemaking effort in 1997 and adopted that pell of 50 just last year. The science in this area has developed to the point where it tells a consistent story. Workers exposed over their working lives to respirable crystalline silica at the prior permissible exposure limit and even at the new exposure limit of 50 get sick and die at rates that are far above what the Supreme Court deems significant. Silica-related health effects are not diseases of the past, as industry petitioners have you believe. Study after study in this record supports the conclusion that workers get silicosis, lung cancer, other nonmalignant respiratory diseases, and renal diseases as a result of their occupational exposure to silica. But your, I would say the vigorousness of your support on the renal disease appears somewhat less than the vigorousness of your support on the other diseases and morbidity. Wouldn't you say that's fair? I would say that's fair, Your Honor. Given that you only put it in one footnote. We didn't spend a lot of time on the renal disease in our brief, as you noticed. OSHA has said from the beginning that the evidence on renal disease is less robust than for the other health endpoints here. It said it in the proposal. It asked questions about the study. It ran it by its peer reviewers. And after it got support for this study, it decided to include it in the final risk assessment. But yes, I think it's correct to characterize the evidence there as less robust or maybe with more uncertainty. There's just less evidence out there. Is your response also that from our perspective it doesn't really make any difference because the other two are sufficiently robust? Is that your point? Yes, Your Honor. That's a very good observation. OSHA's risk assessment is based on a number of health endpoints. Lung cancer, sclerosis, nonmalignant respiratory disease. Is there any possibility that conclusions on renal disease influence the other two? In other words... Not that I know of, Your Honor. Sometimes we'll send something back because of the possibility that we don't know whether the agency's conclusion on... And three is a problem. We'll sometimes send that back and say, well, even though the other two look okay, we need to know whether the conclusions on one or two were influenced by three. I believe they were independent analyses. They were based on different studies in the record. So our conclusions are not interdependent for those health endpoints. And the risk for the other health endpoints are so clearly significant, Your Honor, that a remand wouldn't change that conclusion. What do you say about the death certificate data? The surveillance data that they're focusing on do not and cannot undermine OSHA's risk assessment here. I believe Your Honor has picked up on some of those reasons, but another is related to what we were just speaking about, that OSHA's risk assessment looks at a number of health endpoints that result from silica exposure. The silicosis mortality data only look at silicosis mortality. So just as an example, at 100 micrograms per meter cubed exposure over 45 years, OSHA's risk estimates show that 11 out of 1,000 workers will die from silicosis. Up to 54 will die from lung cancer, 85 will die from nonmalignant respiratory diseases, and hundreds will get silicosis. So those data are sort of the tip of the iceberg in terms of health effects from silica exposure. In terms of underestimation of silicosis mortality, you talked about that already, but there is evidence in the record that death certificates do significantly undercount silicosis mortality. There was one study done by Rosenman which found that only 10% of people with confirmed Category 3 silicosis, which is the most advanced debilitating stage of the disease, had silicosis on their death certificate. So those data represent a huge undercounting of just that one health endpoint. And because of their lack of exposure data, they can't be compared in any way to OSHA's risk assessment or used in any way for risk assessment. And the National Institute for Occupational Safety and Health, who publishes that data, agrees with that assessment. Council said that when I asked him about one of the OSHA's reasons for not relying on the census data was that the decline leveled off after the year 2000. And he said that's false. Experts are concerned that there seems to be a plateau in that data. I, to be honest, don't have it right in front of me, but they have identified that plateau as starting between 2000 and 2010. And they're concerned that new types of exposures that have come into use over the past several decades may be having an effect on health at this point. What do you say to the question raised about the brick industry study, that you used it when you wanted to and didn't use it when you didn't want to? That's not accurate, Your Honor. I figured you'd say that, but I need a little more information. So they're referring to the study by Love. And maybe getting some context here helps. OSHA did its risk assessment, and the studies that it used reflect workers in a number of different industries and a number of different countries under a number of different working conditions. It's a very diverse group of studies, and it shows highly significant risk for all of those workers. The brick industry urged us to look at a number of studies that they said demonstrated risk or lack of risk in the brick industry. OSHA looked at all those studies, discussed them in the record. The only one that had any exposure response information, which is what OSHA needs to meet its burden of demonstrating risk for given exposure levels, was the Love study. So OSHA examined that study more deeply, and it found that although the Love study underestimates risk for a number of reasons, including that it didn't examine health effects among retirees and it only looked at workers at one point in time, the study still reported sufficient cases of silicosis to meet OSHA's benchmark of 1 in 1,000 workers. In fact, in workers ages 55 and older, the age category most likely to have sufficient time of exposure to develop lung abnormalities. The prevalence of abnormal X-rays ranged from 29 for 1,000 workers at very low exposures, around 11, to 164 out of 1,000 workers at exposures just around the prior fell of 100. So once OSHA evaluated the study to the extent it was possible based on the study, it held that this is not a close call on the brick industry. This study, despite its flaws, demonstrates significant risk, as do the rest of the studies OSHA used for risk assessment. I think it would be useful here, Your Honor, in view of the overall attack petitioners make on OSHA's risk assessment here, to just step back a little bit and review the support OSHA had in the record for its findings. Their risk assessment findings were supported by nearly all of the occupational health and medical organizations that commented on the rule, including the National Institute for Occupational Safety and Health, the American Cancer Society, the American College of Occupational and Environmental Medicine, the American Thoracic Society, the Association of Occupational and Environmental Health Clinics, and the American Public Health Association. OSHA's determination that silica exposure causes lung cancer is consistent with the findings of the National Toxicology Program here in the U.S. and the World Health Organization's International Agency for Research on Cancer. Industry petitioners want you to reject conclusions that have overwhelming support among scientists and that were supported by the independent peer reviewers who scrutinized OSHA's risk assessment. They want you to reject this extensive body of scientific evidence on the flimsy basis that there are flaws in some of the studies OSHA relied upon and that there is uncertainty in epidemiology. They want you to impose a legal burden on OSHA that the agency could never meet. I thought it might be helpful to look at a few of the cases in this area just to assess the burden. In the Benzene decision, the Supreme Court stated that OSHA is not required to support its findings if significant risk exists with anything approaching scientific certainty and is permitted to rely on a body of reputable scientific thought. Here, that is exactly what OSHA has done and the broad support for OSHA's conclusions within the scientific community should increase the court's confidence that OSHA's analysis is balanced. In addition, the court's role in substantial evidence review is not to reweigh the evidence and make an independent decision as the industry petitioners imply. Rather, as this court stated in the first lead case, the court expects OSHA to identify relevant factual evidence, explain the logic underlying its choices, state its assumptions, and present its reasons for rejecting significant contrary arguments and evidence. OSHA has done all of those things here. OSHA considered each of the industry petitioners' arguments regarding such issues as the threshold for silica-related health effects, the mortality surveillance data, uncertainty in exposure estimates. The agency explained why petitioners' arguments do not disturb OSHA's finding that silica exposure affects worker health at rates that are highly significant. It was reasonable in its decisions and explanations. Your Honors, the courts understand that OSHA, in marshaling scientific evidence to support a risk assessment, cannot ever achieve perfection because the science those risk assessments are based on is not perfect. There will be flaws in studies. There will be stronger and weaker studies. There may be some uncertainty. But what OSHA has done here is extensive analysis based on a huge body of evidence, conforms fully with the OSHA Act and with the requirements of courts that have interpreted the OSHA Act. Your Honors, I ask that this court deny the petition for review and allow OSHA to protect the millions of workers in this country whose health and lives are at risk from silica exposure. If you don't have any more questions. Do you have any questions? I guess my question is, I always thought it was the OSHA Act. It's actually the OSHA Act? We say the OSHA Act. I see. Okay. I'll have to ask the industry what they call it. Thank you, Your Honors. Okay. Does Mr. Wareham have any time left? Do you want another minute or are you satisfied? Just one minute, please. Thank you, Your Honor. I'd like to make three quick points. One, we are asking this court to look with skepticism at OSHA's analysis and OSHA's finding. There's no doubt about it. We believe the skepticism is well warranted and you have to look no further than page 28 of their brief where they reproduce the summary of the risk assessment that supports this final rule. What you see there, and I'll just look at renal, but for each of the other disease points, it's the same. They predict 32 out of 1,000 risk of fatal renal disease at the level of the new standard. So 32 out of 1,000. Three percent of workers should be expected to die of renal disease. Even at half of the new standard, 25, they predict 25 out of 1,000. Two and a half percent of employees should be expected to die of renal disease at a level of half the new standard. Presumably, if they had gone down to even lower levels of prediction to two and a half micrograms instead of 25, the numbers would still be significant. That can't be. This predicts bodies in the streets and objective evidence just does not support the fact that silica, it presents that level of risk to workers and employees. That's point number one. Point number two, to Judge Tatel's point, absolutely right. And as OSHA's counsel aptly said, the agency analyzed five separate disease endpoints in the final rule, and we provided rebuttal to every single one of those. So, again, given the extreme timelines we have here today, all I could give you is representative examples of where we think some of the big problems are. We read the briefs, so you don't have to cover everything. I do have one question. And the reason I brought up the statement that silicosis is the most prevalent occupational disease in the world is the way you begin your argument, which is this regulation is a solution in search of a problem. And I'm just asking, the deaths have come from silicosis, and this may be outside the record, but are we that much, is our country that much more protective of workers who are exposed to silica than the rest of the world? If it is, maybe you can't answer that. Well, that's a difficult question to answer, but that's why we made a point to the so-called surveillance data, because the surveillance data showed vast improvement in rates of silica disease, and that could be affected by a variety of factors, as we talked about, but undeniably one of those factors is the imposition of the prior standard of 100 micrograms. So, again, we're just asking the court not to turn a blind eye to what appears to be great success. My last point... I'm afraid you're out of time. All right, thank you very much, Your Honor. Your briefs are very good. We'll promise to read them again. Thank you. All right, technological feasibility. Mr. Hammack for the industry petitioners. Thank you, Your Honor. May it please the court. My name is Brad Hammack, representing industry petitioners, American Foundry Society, National Association of Manufacturers, and our numerous construction industry trade associations. As you point out, I'm Chief Judge Garland. This is related to technological feasibility. The panel is unlucky enough to get me and Ms. Goodman twice today, first on technological feasibility and then second on economic feasibility, which, as this court knows, is one of the constraining factors of the OSHA rulemaking process. Quite simply, Your Honor, it's OSHA's determination that the rule is technologically feasible, capable of being done for the construction, foundry, and hydraulic fracturing industries is not supported by substantial evidence in our view. We give several reasons for this in our briefs, which we won't repeat here, but I would like to focus on just a couple of points, starting with the construction industry. For construction, I want to focus on Table 1. Table 1 is something that you've read a lot about in the briefs and really is the linchpin of OSHA's construction standard. In our view, above all else, Table 1 demonstrates convincingly that the rule cannot be met in most operations most of the time in construction. Table 1 is the embodiment of OSHA's technological feasibility analysis. It is designed, in OSHA's own words, to meet a PEL of 50. If there is a control measure that is not on the table, OSHA concluded, based upon its review of the record, that it could not reliably reach a level of 50. Furthermore, if respirator use is required by Table 1, that means that OSHA has determined that a PEL of 50 cannot be met without regard to the use of respirators. And this is where we start to get into OSHA's fundamental burden to prove technological feasibility without respect to the use of respirators. Table 1 demonstrates infeasibility by requiring extensive respirator use on its face, which I will talk about, as well as omitting numerous control measures that will have to be used by employers in construction, but that by the agency's own analysis cannot reach a level of 50. Let's start with the obvious respirator use. Of all of the tasks and locations for those tasks analyzed on Table 1, by our count, and I admit there are different ways to count this, but by our count, one-third of them require some form of respiratory protection. When the task is performed for just over four hours. And this does not include abrasive plastic, which the agency has already found is technologically infeasible. When costing the number of construction employees that may need to be provided medical surveillance under the rule because they have to wear respirators for at least 30 days per year, OSHA assumes approximately 300,000 construction employees will have to wear respirators if they follow Table 1. In spring, OSHA says, the Secretary says, that this amounts to only 13% of construction employees. Assuming for the sake of argument that that is true, we contend that that is an opposite position, a contrary position that the Secretary has taken in other rulemakings and in other litigation, mainly recently the hexavalent chromium rule, where our read is that OSHA concluded that 9% of respirator use for the affected employee population was too much, and that rendered the pill that they had initially proposed in that rule to be technologically infeasible. I hadn't thought about this before, but you related it to the cost of medical surveillance. Is it technologically infeasible, are you saying, or is it economically infeasible? I'm saying it's technologically infeasible. Could you explain why use of respirators is technologically infeasible? I'm sorry, I misunderstood the question. No, the use of respirators is not technologically infeasible. What we are saying is based upon this court's precedent, though, OSHA has to make an initial technological feasibility analysis that focuses first and primarily on engineering and work practice controls to reach a level that they are seeking to reach without regard to the use of respirators. Now, there can be isolated respirator use, Your Honor. That is absolutely the case. But what we are suggesting here is that in the construction industry, we have more than isolated respirators, and we think the agency has taken different positions in different rulemakings to reflect this. The reference to the medical surveillance and the cost study was just a way to get our handle on how many construction employees are out there that are potentially affected by this, and therefore try to get an idea about the number of respirator uses. Furthermore, I'd like to make another point. For many of the operations in Table 1 where OSHA is claiming feasibility, the engineering controls cannot be used. And, for example, take the first six tasks on Table 1. OSHA has indicated that they have to be performed wet. So if someone is using a stationary masonry saw cutting a piece of granite in your kitchen, for example, OSHA has said in order to follow Table 1, you have to cut wet. Well, what happens in that situation where you may decide for whatever reason you're not going to let the employer cut wet? In that situation, the employer has to seek another engineering control, basically cutting dry in that situation. Those types of other engineering controls... Sorry, Ron, yes? I'm sorry. I don't have granite in my kitchen. Why would someone not let them do that? Well, there are a number of reasons in the rulemaking record itself as to why an employer would not be able to cut wet. And so I would obviously refer the court to that. Just to name a few, for example, there are situations where adding water to certain types of aggregate will change the face and the color of the aggregate. So you can't apply water in that situation. There are certain types of facilities, health care facilities inside of people's homes where they don't want to introduce water because of mold. So the point is there are situations that have been identified in the record where it is infeasible to use wet methods, and then you have to use a dry method, and that's the point. OSHA has not found that those are technologically feasible as well. So you have a number of instances in play in construction where respirator use will be required, and we contend that threshold is not isolated respirator use, but in fact something that's more broadly that really goes to technological feasibility. What's the data on respirators having to be used more than four hours? The data varies, Your Honor. In OSHA's assessment of that, and it's found in Table Roman 5-45, OSHA looks at the number of tasks and projects the extent to which those tasks are estimated to occur more than four hours and whatnot. And what you will see is, in fact, and this is true, their belief is that the length of tasks differ per operation. So if you're jackhammering, for example, OSHA estimates that 50% will be above four. And we contend it shows several examples of when a full shift work would have to take place in exposures of four hours. So several examples is not the same thing as in many cases. So what's the evidence? I understood from OSHA's brief that they don't think that there are many circumstances where more than four hours are required. Now, obviously respirators can be used in limited circumstances, and the question is, what is your evidence that, in these circumstances where they've identified use of respirators, that it's going to have to be used for more than four hours? Your Honor, I would ask that you look at the same evidence that actually OSHA is presenting, because I think there are different ways to look at it. But I think you will notice, and I don't know all the operations off the top of my head, admittedly, that a number of them have significant, at least 50% of the operations are estimated to take place more than four hours. I also took their point to be it's not the gross number of operations, it's how many employees are in each of these different fields doing these tasks, and that's where they get a lower number than you get. Yes, Your Honor, that's a good point. They do focus on the number of employees exposed. I would pose this point to the court. Again, the legal test here is in most operations most of the time, and I suppose there are a number of ways to look at that. However, you don't hear what OSHA wants to talk about in terms of employees. A reasonable possibility in most operations. I'm sorry, Your Honor, excuse me. A reasonable possibility. I was just adding to your standard. Yes, a reasonable possibility, that's right. But I ask that because, I mean, given the standard, the reasonable possibility that the typical firm will be able to install controls that can meet the PEL in most of its operations, where does the respirator, how does that fit into that standard? I'm sorry, could you repeat the question again? The standard is, it's technologically feasible if there is a, quote, reasonable possibility that the typical firm will be able to install controls that can meet the PEL in most of its operations. Right? That's the standard. Yes. Okay, so how does it, how does, how do the aspects of Table 1 fail that test? Exactly, Your Honor, and I would raise two points. No, that's my question. Right? How does it fail? How do they fail? You can say it. I should have said, good question. Well, what's the answer? I think your question raises one point which I'd like to make and then I'll answer it. I promise I'm not avoiding your question, Your Honor. Okay, go ahead. I think one of the things that this case presents, in Table 1 in particular, is what does really in most operations most of the time mean? Because I think you have a standard here where you've got, in our view, something more than isolated respirator use. And OSHA's saying here that it's not in most operations most of the, it is in most operations most of the time, but in other rulemakings they said something different. But I contend here, just based on the data and the volume and the record, that one-third of the tasks on Table 1 requiring respirator use when done for more than four hours, plus abrasive blasting requiring respirator use, plus a whole host of tasks that are not performed wet but are performed dry and will presumably require respirator use, reaches that level of not meeting in most operations most of the time with only isolated respirator use. Your Honor, if I might just turn quickly to the foundry industry. You're out of time. If you want to take up your bubble, you can. I'd be happy to. Thank you, Your Honor. I would just like to note with respect to the foundry industry, in our view, substantial evidence does not support OSHA's finding that the rule is technologically feasible as well. And this is for a different reason. This relates to exposure variability. To reach a level of 50 in most operations most of the time, evidence demonstrates that foundry employers must attain a level far below 50. This the agency has not done. We believe the best available evidence of this exposure ability was a study performed by the American Foundry Society and submitted to the rulemaking record, which looked at the variability and concluded that for an employer to be even 84% confident that they would reach a PEL when implementing engineering controls, they had to target a level of 20. OSHA, in our view, did not address this study in the technological feasibility analysis of the rule. And as a result, their finding that it's technologically feasible for the foundry industry to meet the PEL is not supported by substantial evidence. They have a study showing up to 80% of the variability could be eliminated, right? Yes, Your Honor, except those studies that they cite, none of them are in the foundry industry. And the foundry study itself looked at, and it was unique in this respect, because it looked at the same job in the same facility over a certain amount of time. Presumably, some of those facilities were controlled. Presumably, some of them may not have had great controls. But either way, that study, Your Honor, directly addressed the foundry industry, and the studies cited by the agency do not. And so we would suggest that they needed to address that study specifically as it relates to technology. And what about their answer that if it's really not feasible in particular foundries, that there's enforcement flexibility? I know, Your Honor, and this court has held three times in this case. But I would make the following points. Number one, in those cases, I do not believe that a study of this specificity was introduced and how it relates to the foundry environment. And I would also ask this court to review the enforcement policy that OSHA is stating here. Because I would assert it's really not a policy at all, Your Honor. It's really simply just stating in the preamble to the final rule that in certain circumstances. And that's what was done in the other three examples, and we upheld them in those cases. So you're putting us in a tough position to criticize or to overturn. I understand that, Your Honor. But I also think in some of those prior decisions, what we've seen in terms of the agency progressing is announcing things like enforcement policies in different avenues as opposed to the preamble to the final rule. And I would just make one. I know my time is up, Your Honor. Yeah, it's four minutes up. Okay. Let's hold your final point for the briefs. All right. Thank you, Your Honor. I appreciate it. Thank you. Now, Ms. Goodman, again. Not again, but to come again. Good morning. May it please the Court. My name is Lauren Goodman. I'm Counsel for Respondents, OSHA, and the Department of Labor. And I will be addressing the issues of technological feasibility. Now, industry challenges OSHA's technological feasibility findings only with respect to three sectors, boundaries, tracking, and construction. And I want to start by pointing out that many of industry's arguments, including the argument Mr. Hammack just discussed about the use of wet control methods under Table 1, many of their arguments boil down to one common refrain, which is that some employers may be unable to comply with a new exposure limit using engineering and work practice controls some of the time. But even if that's true, for whatever reason, that assertion does not give credibility to the industries in question. The legal test for technological feasibility that Judge Tatel stated just moments ago presumes that an OSHA health standard can be technologically feasible, even when there are times when some employers cannot use engineering and work practice controls to comply with this health. And this Court's two decisions on OSHA's lead standard explain why OSHA's establishing technological feasibility in this pre-enforcement challenge is so limited, or in the words of the Court, greatly eased. One, a finding of technological feasibility at this stage simply establishes a presumption of feasibility that individual employers can rebut an individual OSHA enforcement action. Furthermore, by its express term, the rule itself requires only that which is feasible and allows for respirator use if it turns out that an employer cannot comply with Tatel using engineering or work practice controls. What about their, if you can address their respirator argument in a little more detail. So their argument is, yeah, that's fine in isolated cases. It's not fine in the circumstances that they're describing. They say most third of the tasks in Table 1 require it. And then you add up those additional things that he mentioned. I don't know how many times you multiply that and divide it, but somehow that gets to a majority. Yeah, I disagree. What's the answer to those points? I disagree with their assessment first with respect to Table 1. They get to one third of the tasks on Table 1 using perhaps what is some fuzzy math. They're essentially double counting each entry on the table, even when those entries all relate to the same task. OSHA views the table differently. And if you break down, OSHA has 19 tasks on the table. For 13 of those 19 tasks, respirators are never required. It does not matter how long the work is performed. It does not matter where the work is performed. And even for the remainder of the tasks on Table 1, for which some respirator use is required, OSHA's view is that they are only required, in fact, they are only required in limited situations. So for almost, for over half of the tasks, respirators are never required. For the remainder of the tasks, they are required only in some situations. And I do want to clarify something with respect to the chromium standard, which Mr. Hammack referenced. They point to OSHA's allowance of 13% respirator use here in construction and suggest that that is somehow inconsistent with OSHA's position in the chromium litigation and in the chromium rulemaking. And it is true that in chromium, OSHA elected not to adopt an exposure limit at which 9.5% of workers would have needed respirators. But that decision was not based on that 9.5% figure, and certainly not based on a determination that that 9.5% respirator use was accepted. I encourage the Court to look at the page of OSHA's brief in the chromium litigation that petitioners cite and have included in the appendix, because very clearly on that very page of OSHA's brief, it states that 9.5% respirator use figure did not factor at all into OSHA's technological feasibility findings for that rule. In that case, OSHA rejected an exposure limit, not because 9.5% of employees would have been in respirators, but rather because respirator use would have been significantly higher than that, upwards of 30% or more, for certain application groups that together accounted for the majority of employees covered by that rule. Let me just go back to, in this rule, the four-hour question. What's the evidence on whether where it is required, it's going to be required for more than four hours? OSHA believes that most of the time, and in fact, found that most of the time, construction work is going to be performed intermittently or for short periods of time. That evidence was in the form of testimony from construction workers, from construction trade associations, from representatives of the relevant labor unions. There were also studies in the record, in particular, a Flanagan study from 2003 that looked at a number of construction tasks and concluded that, on average, construction tasks that generate filter are performed from 150 to Can you say something about the variability in the foundry industry? Absolutely. At the outset, with respect to the AFS study that they were relying on, I want to make clear, that study was designed to address a mathematical principle. It simply did not address how much variability can actually be controlled. The evidence in the record, including the studies that you mentioned, Judge Garland, show quite clearly that a substantial amount of variability can indeed be controlled because it's based on factors that employers are aware of and employers are able to predict what their exposures will be. I also want to make clear that variability is not a concern with respect to the construction standard. OSHA's finding with respect to the construction standard was that a typical employer is going to follow table one. Table one does not require separate compliance with the new exposure limit. Therefore, variability cannot preclude construction employers, at least a typical construction employer, from complying with the standard. I also want to emphasize, with respect to variability, that this court's decision in the effective case made quite clear that nothing about the potential for variability, no matter what that potential is, is going to change OSHA's legal burden. Ultimately, the most operations, most of the time, does not ask, as petitioners would have the court do, whether every employer can comply all of the time, or even whether a given employer can comply with 84% or 95% confidence. The final point I'll make on variability is simply that the court should look at the data here. OSHA has extensive data on foundry information specifically, over 1,200 samples in OSHA's exposure profile, as well as data submitted by industry. That data inevitably accounts for any variability these foundry facilities are facing. And those exposure profiles show quite clearly that employers are not precluded from reducing exposures below the new exposure limit. I see my time is up. Can I ask you a question about fracking? And that is, at the time of the rule, my understanding is they hadn't even made the 100 Pell. It's been a year. Are you following the progress? Because it seems to me if they haven't met it at the time of the rule, and you give them five years, I mean they haven't met the original at the time of the rule, and you've given them five years, how are they doing so far? I don't have any evidence on that, Your Honor. What I have is the evidence and the record that OSHA relied on discussed extensively in the technological feasibility analysis, explaining that new controls have been developed, are starting to be rolled out in the industry. In fact, some of them are already in use in a number of facilities. And while data has been limited, what data we have shows that they have been effective in at least reducing the exposures below the prior exposure limit of 100 micrograms per cubic meter of air. And OSHA identified additional controls, simpler controls that are readily available today, things like operator booths and water misting, that they believe will allow the industry to reduce exposures further to the new exposure limit. And a final note on that issue is I just want to point to the fact that this court's decision in the first lead case, indicating that OSHA is not bound to the technological status quo, and that because OSHA standards can be technology forcing, OSHA can require industry to meet a toll that has never previously been attained anywhere, as long as it provides reasonable time for employees to commit to compliance. But we gave them 10 years in lead. That's correct. In this case, OSHA found that given where we are in terms of the technology development and the relative technology, that five years will be sufficient. And industry has not seriously challenged that contention. There's certainly no evidence to the contrary. Thank you. Thank you. Does Mr. Hammack have any time left? Oh, I'm sorry. I'm sorry. I apologize. I apologize. A little hard to keep all the players. This will be Ms. Bohr. Yes. Yes, thank you. Good morning. May I please record my name is Victoria Bohr. I'm a counsel to North America's Building Trade Union. And I'd like to briefly address the significance of Table 1 to OSHA's technological feasibility analysis and the construction standards. By way of context, Table 1, which is the centerpiece of the construction standards, is a gift to the construction industry. Most OSHA standards set a permissible exposure limit and require employers to monitor their workplaces and devise their own strategies following the hierarchy of controls to bring exposures below the permissible exposure limit. The SILCA standard gives employers options. They can follow the traditional approach, or they can follow Table 1, which is in effect a manual that lists 19 of the 23 construction paths that most commonly generate significant SILCA exposure and specifies control strategies for each. Employers who fully and properly implement the controls listed on Table 1 are free from monitoring their workplaces and have a safe harbor for complying with the PEL. Table 1 was devised as a response to the small business community and developed with input from the government, industry, and labor through the protracted rulemaking process. As Ms. Goodman said, OSHA assumes that most employers will follow Table 1, which is a completely reasonable assumption because it tells employers exactly what they have to do, frees them from monitoring, and gives them a safe harbor for complying with the PEL. Now, rather than accepting this gift, as Mr. Hammack already explained to you, the industry petitioners point to it and argue that to the extent it requires the use of respirators, OSHA is in effect conceding that the standard is infeasible. In the first led standard, this court made clear that the starting point in the feasibility analysis is what the standard requires of the regulated community, a step that this argument, petitioner's argument, completely ignores because Table 1 does not require employers to comply with the PEL. What it requires is for employers to implement the listed controls. So whether the PEL can be reached without the use of respirators, is the question that the industry petitioners focus on, is actually completely irrelevant. What is relevant is whether, as Ms. Goodman said, the typical employer can comply with Table 1 most of the time. And on this question, petitioner's infeasibility argument rests on vague assertions that in certain circumstances, certain employers may not be able to use certain of the wet methods on Table 1 at some time. As we explained in our brief, and there's substantial record evidence, including industry testimony, that refutes these claims. And petitioners have pointed to no evidence that undermines OSHA's conclusion that most employers will be able to comply with Table 1 by utilizing those controls most of the time. Thank you. Now I'm back to Mr. Hammond. Do you want one minute? We're going to hold you to one minute. Thank you, Your Honor. And just very quickly, I appreciate having one more minute. I just would refer the court, and like Ms. Goodman, I ask that the court looks at the briefs that were filed in the HEXCROM case as part of the joint appendix. I do think the HEXCROM situation matches up quite perfectly with this situation in the sense that in HEXCROM, OSHA conducted its analysis based on application groups. They looked at overall respirator use within those application groups, and they determined an overall percentage of respirator users. In addition, they were influenced by the fact that several sub-application groups would require much more respirator use. That, we contend, is exactly what they did here, except they used a different metric. And I would also just point out, Judge Henderson, with respect to fracking quickly, I would also encourage the court to look at those industries in the initial lead decision, which found that there wasn't enough evidence to justify the technology-forcing aspect of the stand. Thank you. Only six seconds over. Thank you, Your Honor. It's me again. Starting over on economic feasibility. Over on economic feasibility. For several reasons, we also assert that OSHA's finding that the rule is economically infeasible for construction, foundry, and high-product factory industries is, again, not supported by substantial evidence. And at the outset, I would encourage the court to look at the record and look at some of the overall findings of costs that OSHA has put forward. And I think that underscores that many of the assumptions that they've used are not supported by substantial evidence. And as we note in our briefs, you know, this is a comprehensive standard. There's medical screening and surveillance. There's respiratory protection. There's engineering controls. There's rain exposure control plans. Very extensive requirements. For certain industries in the construction industry, for example, OSHA estimates annual cost per establishment at really low levels. I mean, if you're a residential home builder, cutting granite in your kitchen again, the costs of the rule are going to be estimated to be about $360 a year. We believe that these costs are low because the fundamental inputs are not based upon substantial evidence. And I'll start quickly with construction and then like to turn to boundaries and only spend a minute on hydraulic fracturing if there's time. Two critical assumptions in construction that we believe OSHA used are not supported by substantial evidence in the rulemaking record. And we've already talked about one in the context of technological feasibility. And that is that, like technological feasibility, OSHA assumes for purposes of its cost analysis that essentially everyone is going to follow Table 1. This assumption is despite the fact that there are numerous excerpts in the record of people saying Table 1 will not be used. So OSHA's analysis is fundamentally flawed. Will not be used because they can't use it or because they prefer something else? Right. Because they can't use it, Your Honor. Exactly. I will say, Your Honor, this is actually evidence of both. There are some people that the choice question doesn't matter, right? Because if Table 1 is cheaper and they decide to pick a more expensive choice, that's basically their problem, correct? Correct. That's absolutely correct. It's those situations where they can't use those tasks. So we're back to the same question we were at before. Correct. And they say 13 out of 19 and you say one-third. And then so it's really the same question? In part, it's the same question. This filters through both analysis, Your Honor. That's correct. So in addition to that, I would like to point out one other assumption that we believe is not supported by substantial evidence, and that is OSHA's assumption that construction work will be done and performed throughout the country only 150 days a year. This is something that OSHA relies on throughout its psychological feasibility analysis. We have tried to find in the record evidence that supports that 150 days, and we have been absolutely unable to find it. We believe that that assumption is not supported by substantial evidence, and we would suggest the agency needs to reexamine its cost analysis pursuant to a more reasonable estimate when it's based upon substantial evidence of the number of construction works. So the test that you're faced with, costs it imposes do not, quote, threaten massive dislocation to or imperil the existence of the industry or threaten the competitive structure of the industry. That's a test, right? Correct. Is there a study in the record that says that the construction industry will be imperiled by this? There are two major studies in the record that reevaluate the costs. I got that part. I know that there's a debate about costs. Right. And then both of the courts said this isn't cost-benefit. This is really end-of-the-world cost. Right. Is there a study on that? Both of those studies filter through their cost estimates to the impacts on revenues and profits of various industry sectors and concluded that they exceeded the thresholds for industry profits and revenues, impacts on profits and revenues for most industries. I don't know the exact percentage right here at the top of my head. Most industries under their adjusted cost figures. But I would also like to... Before you move on, I just want to follow up a question to what the Chief Justice said. Yeah. I understood your argument to be a substantial item. I didn't understand that you were arguing that this would result in massive dislocations in the industry. In fact, I didn't even see any reference to that until your reply. Is that right? I think that, well, Your Honor... I think that it was only with respect to... And you didn't even make that argument with respect to... I'm looking at page 38 of your reply. That's the first time I saw it. And it's only there with respect... You didn't even make that argument with respect to fracking infrastructure. So have I misread your position? Our position, principally, Your Honor, is just that. You did not read it. Our position, principally, is that OSHA's assumptions are not based upon substantial evidence. So we think that's an independent... I understand that, but I had the impression from your answer to Judge Garland that you were also arguing that this standard would result in massive dislocations in the industry. And I didn't see that in your rubric. I understood. Our principal argument, Your Honor... No, only argument, right? Yes. I think Judge Garland was... Only argument. Am I right about that? You're correct about that, Your Honor. Okay, fine. You're correct about that. Okay. On the... Well, go ahead. Go ahead. Okay, go ahead. I would like to turn now quickly to the foundry industry. And I'm conscious of my time, Your Honor, so I won't go over. For two separate reasons, we also believe that OSHA's economic feasibility assessment for the foundry industry is not based upon substantial evidence. One relates to the model that OSHA adopted to cost engineering controls implemented in the foundry industry as well as other general industry sectors. And one relates to the actual types of controls that OSHA costed. First, as we set forth in our brief, OSHA adopted a per worker model of assessing cost of controls in general industry. This essentially groups numbers of overexposed workers and assigns control costs based upon those numbers of workers. That cost evidence was submitted in the... That model evidence was submitted in the record understate the true cost of controls because it suggests in a number of situations that you're going to have multiple overexposed workers protected by every single control fund. You know, under the EPA, we are required to apply a harmless error principle. And I wonder about that with respect to foundries. Because even if they're off a little bit, or even a lot, for the foundries, they had this 1% threshold. They were well below that. In other words, I mean, even if they were doubled, I don't think they'd reach the threshold. So, how do you... How do we think about that? How do you think... Yes, Your Honor. Another good question? How many minutes have I been up here, Your Honor? Good question, Your Honor. So, a couple of responses to that. First of all, I think when OSHA looks at the percent impact on revenues versus profits, I see for those same foundries, even OSHA's estimates of impacts on profits are much higher. But I would also say that adoption of the... The reason I ask the question is, given this standard, and given how far below the threshold two of these industries are, don't you have an obligation to show us that if OSHA's cost estimates... Not only that you don't think they're supported by substantial revenues, but that they're so far off that if they were proper, you'd be way over... You'd be over the threshold. In other words... Do you see my point? Yeah, I do see your point. I didn't see that anywhere in your brief. I didn't see that. We did say in the brief, Your Honor, that the contrary model put into the record by URS and Environomics, which went through the entire process of adjusting the cost, did threaten those thresholds that you speak of. I would also note that the contrary model that was implemented, as opposed to the per worker model, which is called the facility model, which is what we've said, estimates that for engineering control costs, those would be at least three times higher than the per worker model. So I think everyone can see, I think, in this room that most of the costs of this rule are driven by control measures. And as a result, that would have a significant impact, we believe, on the overall cost. But your model assumes a random distribution of exposed workers, right? That's the point of this binomial model, right? Yes. Yes. Well, that's what a binomial distribution is supposed to be. Right. Like a normal perp, right? As was made clear in Respondents' Opposition. Yeah. So... And also in my Statistics 101. That's right. So... So we have two different explanations. One is that they're already subject to control, so they're going to be clumped already. And we have another thought that they're going to be random. We don't really have any basis for deciding between these two models. Both are supported by experts, but in the context of this industry, we don't have a way of deciding. That seems like substantial evidence in favor of OSHA's expert judgment about where exposed workers are most likely clumped. The one distinction I would make, Your Honor, about the two models is the second model, what's called the URS model... That's our model. That's our model, correct. We believe this more accurately reflects how an employer would have to implement controls because it would necessarily take account of two things. One, whenever an employee is overexposed, regardless of whether it's clumps of workers, two workers, three workers, four workers, if it's one worker, then in fact the controls are going to have to be implemented. We believe the URS model does a better job of that and so should get preference in that respect. In addition, it also better considers the situation where you'll have one employee doing different tasks throughout a foundry, for example. But you see, once again, let me interrupt you, but you see, to win, you need to demonstrate not that your model is better, but that OSHA's is full. That's what you have to show. That's right, but I think... So you need to talk about OSHA's study and why you think it's flawed and also why it's reasons for not accepting your model are flawed. That's the way you win your case. You can't win your case by telling us yours is better than theirs. It doesn't work that way. You agree with that, right? I agree with that, and I think in our briefs we have set that out. The idea being that their model... I was just responding to what he was saying. Their model doesn't really reflect the way that economic decisions are made by employers. This contrary model does, and as a result, OSHA's failure to implement that model and carry through its economic analysis is not based upon substantial evidence. Are there questions? Okay, thank you. Thank you. All right, we'll hear from Ms. Goodman this time again. Thank you, Your Honor. Lauren Goodman again for Respondents OSHA and the Department of Labor, this time on economic feasibility. As the Supreme Court explained in its 1981 decision on OSHA's Cotton Dust Standards, Congress, in passing the OSHA Act, intended for employers to bear the cost of ensuring safe and healthy workplaces for American workers. And in fact, as the Court explained, Congress viewed the cost of worker safety and health, even when substantial, as standard cost of doing business. And because of the preeminent emphasis the statute places on worker protection, this Court and others have consistently held that an OSHA rule is not infeasible just because it is costly or even very costly. And as the Court has already noted today, the bar for infeasibility is extremely high. An OSHA standard is economically infeasible only if it threatens the downfall of an entire industry. Thus, OSHA's burden of establishing economic feasibility for a new standard like this one is not particularly demanding. OSHA need only show that it used the best available evidence to develop a reasonable estimate of compliance costs and that the projected costs are not likely to threaten the existence or competitive structure of affected industries. And that's precisely what OSHA did here. OSHA used the screening threshold for economic impacts of 1% of revenues and 10% of profits, thresholds petitioners have not challenged, to distinguish between industries for which it can assume feasibility and those that require more scrutiny. And as the Court has already noted, the three sectors for which petitioners are challenging OSHA's finding of economic feasibility, foundries, construction, and fracking, the projected costs of this silica rule are below, and specifically with respect to foundries and construction, are below even OSHA's screening threshold. Well, I have a question about the record with respect to both of those industries. For foundries, what evidence supports the 50-50 split between the old town and the new one? What evidence in the record supports that 50-50 allocation? And for construction, what supports the 150-day assumption? Absolutely. I will tackle your 50% question first, Your Honor. OSHA assigned, with respect to... Because if those are really all, that would dramatically affect the cost, right? So... I would not... Let me actually respond to that. Why don't you answer my question before asking the other one? Okay. Okay. Please come back to that, because that is the point I would like to make. You may not have to get to the second. If you've got record evidence to support both of those. With respect to the 50% assumption, Your Honor, OSHA assigned half of control costs in general industry to the prior costs, costs that are not attributable to this rule, and half to the new cost. Right. Industry's sole criticism of that assumption was that it didn't account for the rising incremental cost of controls. And OSHA clearly stated in the preamble that, in fact, it does, by its very nature, take that into account. By assigning equal costs to reducing exposures from uncontrolled situations to 100, when uncontrolled we usually refer to as about 250 to 300 micrograms per cubic meter there. Reducing those exposures to 100 and assigning equal costs to going to a more limited extent from 100 to 50, OSHA inherently accounted for the fact that it will cost more for each incremental reduction in exposures at the lower level. But beyond that, exposure data show that the average general industry worker exposed above the prior pill is exposed above roughly 300 micrograms. And typically exposures that high can only be addressed through using the most extensive types of controls. Things like local exhaust ventilation or wet methods, possibly with improvements in housekeeping. And if you look at the breakdown of costs, things like LED account for over 60% of control costs. Wet methods and ventilation together, over 75% of control costs. So in OSHA's view, this 50% breakdown is actually a conservative estimate. If you talk to OSHA's economists about this, they will tell you that more likely than not, it's going to cost significantly more. I talk about this all the time, Robert. Are any of those conversations reflected in the record? The evidence in the record, OSHA confirmed its belief about the controls that would be necessary in two ways. First, it looked at a comparison of four well-controlled foundries and four poorly controlled foundries and found that facilities with few exposures above the new pill were more likely to have things like working LED and good housekeeping. And the facilities with many exposures above even the old pill tended to have little or no LED for housekeeping, no worker enclosures, poor maintenance practices. They looked at that and found that that substantiated their belief that it would take the most expensive controls to get down to the prior pill. OSHA also looked at the exposure data in the record that contained control descriptions and found that exposures above 250 micrograms per cubic meter of air typically occurred in uncontrolled situations. And when exposures hovered between the old pill and the new pill, they were typically associated with the use of some controls, usually things like LED. So OSHA used the record evidence to substantiate its belief that, in fact, a 50% assumption was a conservative one. And you also asked about the 150-day issue for construction. And what I'd like to say with respect to 150 days is that was a piece of OSHA's analysis that OSHA adopted in response to comments it received during this briefer process, that, in fact, OSHA was not taking enough costs for engineering controls because it was not accounting for inefficiencies in control use. In other words, that there was going to be downtime at which, especially in construction controls, they're not going to be used. So OSHA made the adjustment. The effect of that adjustment was only to increase the estimated cost of controls overall. It actually surprises me that petitioners have chosen to focus on this issue. Could you just say again why that is? I understand it's because you first figured out the PERG employee cost. Well, that's with respect to the foundry industry. We're now talking about the construction sector. So why does it increase the estimated cost? Because, especially with respect to costs for engineering controls, and it's not even all the engineering controls, OSHA had multiple ways for estimating the cost of controls depending on the data they had available. When OSHA did not have monthly rental rates, it used its assumption about the number of working days only as a divisor. So, in other words, it took the annual cost of the control, divided it by 150, and then multiplied by a constant that would not change irrespective of what assumption it used. So by lowering the estimate from 250 days to 150 days, the result was only to increase the daily cost of control. And as we explained in our brief in a footnote that contains more math than I typically like to do, the end result of changing the assumption would actually be to decrease OSHA's estimated cost. So I am not entirely sure why this was an issue industry decided to pick up on, because in the end, if we change the assumption in the direction they're requesting, it would simply make OSHA's case stronger. I would like to take a moment, a bit off track, to address one issue that Mr. Hanek did not raise, but I want to make abundantly clear to the court. Industry made an argument in its opening brief with respect to the brick manufacturing sector, arguing not that the rule was infeasible under accepted tests for economic feasibility, but rather only that OSHA could not apply the standard to that industry because it imposed substantial costs without producing any health benefits. In the reply brief, they note that OSHA conceded that point, and I want to make sure the court understands OSHA emphatically did not concede that point and instead addressed it in a lengthy footnote, I believe it's footnote 42 in our brief. Ultimately, that argument was based on a faulty premise about risk, and in any event, the suggestion that OSHA rules are set by a balancing of costs and benefits is patently inconsistent with the Supreme Court's decision in conduct. And finally, I do want to quickly address Mr. Hanek's point about construction. There's just no merit to the argument that OSHA erred by failing to account for monitoring costs for employers that do not or cannot use Table 1 in construction. OSHA's economic analysis was, as the court's lead decision permits, based on the typical behavior of the typical employer in each industry. And OSHA found that most construction employers, and indeed the typical construction employer, is going to elect to follow the Table 1 compliance option and therefore will not incur monitoring costs. I see my time is up. Further questions? Thank you very much, Your Honor. I'm sure Mr. Hanek did not have time again. I have a question. Well, we have a question for him, so you're up. Yes. No, we really want to see your name. I'd like you to answer fellow counsels' curiosity about why you're challenging the 150 out of the assumption. You want me to do the assumption? Yes, Your Honor. Well, first, we believe that all OSHA assumptions have to be based upon substantial evidence. Okay, but her point is that this benefits your clients. And we disagree because we think the reason that they have said the costs go down is just a function of the model that they happen to choose, which was essentially assuming a particular cost for a particular control and then converting that in some way. We believe their overall assessment would be much higher if they actually increased the number of days. For example, OSHA talks about labor productivity penalties. They calculated a certain number of labor productivity penalties, right? Would those be affected by work being done 150 or 250 days? What about increased maintenance costs that would occur because these control measures would be used more frequently? There are so many different variables that should be, in our view, considered by OSHA in this particular case that isn't. And so I don't think that simply suggesting that under this same model it would reduce in cost fully answers the question, Your Honor. Okay, I just would like to hear OSHA's response to your response to OSHA's response. Thank you, Your Honor. My response to that is simply to say, A, that was not an argument made in their brief. In their brief, they were simply talking about the effect of 150 days on the cost for controlled respirators and exposure control plants. To my knowledge, and I would probably have to confer with my comment to confirm this, to my knowledge, that is the only place where this assumption came into play in OSHA's economic analysis. So without them pointing to additional places in the analysis where this was used and it was problematic, I don't have any additional response. Do you want me to reply to this? No. Thank you. Thank you. I think that finishes feasibility. Is that right? Right. We'll take a brief, very brief break before beginning the union challenges. You can all trade chairs or whatever it is you need to do. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A31961. Public record. Not legal advice.