Oral argument — United Steel, Paper and Forestry, Rubber, Manufacturing, Energy, Allied Industrial and Service Workers International Union v. Mine Safety and Health Administration

Case No. 18-1116, United Steel, Paper and Forestry, Rubber, Manufacturing, Energy, Allied Industrial and Service Workers International Union et al. Petitioners v. Mine Safety and Health Administration et al. Ms. Collard, Petitioner of United Mine Workers, Ms. Edgar, Petitioner of United Steel Workers, and Ms. Scott, the respondent. Ms. Collard. Good morning. Good morning, Your Honors. I'm Laura Carr, representing Petitioner of United Mine Workers. This case concerns MSHA rulemaking regarding examinations of working places in metal and non-metal mines that is unlawful for two reasons. First, the new rule, the 2018 rule amendment, decreases the safety protections guaranteed to miners under the prior 2017 rule in violation of Section 101A9 of the Mine Act. Second, the amendments represent unlawfully arbitrary and capricious rulemaking by MSHA. I will address the Mine Act violation for the first six minutes of this argument. My co-counsel from the United Steel Workers will then discuss MSHA's arbitrary and capricious rulemaking for an additional six minutes, and we would like to reserve three minutes for rebuttal. Two points are key to understanding MSHA's Section 101A9 violation. First, the 2018 rule amendments decrease miner safety protections, and second, MSHA's claims to the contrary are not entitled to deference because they are not based on factual findings. It is also important to understand what is not at issue here, and that is the applicability of Section 101A9 to this case. The statute provides that no mandatory health or safety standard promulgated under the Mine Act shall reduce the protection afforded to miners by an existing mandatory health or safety standard. It places an explicit limit on MSHA's rulemaking power, reflecting Congress' recognition that mining is an exceptionally dangerous job and that miners, therefore, need to be protected to the maximum extent possible. MSHA has conceded the statute's applicability, both in the agency's brief and in the administrative record for the 2018 rule amendments. Yet, MSHA reduced miner safety protections via those amendments in violation of Section 101A9. The agency did this through two important and substantial changes to the protections in the 2017 rule. First, MSHA removed the requirement for mine operators to inspect all working areas of their mines before work begins on each shift, and gave operators permission to wait to start inspections until the same time that miners are beginning work. And that sounds problematic on its face, but if you look at the Federal Register explanation, the agency makes clear they expect that this will be implemented in a way that no miners are exposed to hazards before the inspection is complete. Yes, I'm familiar with that explanation. There are two points that I'd like to raise in response to that. The first is that, despite MSHA's discussion on the subject, it's manifestly impossible to prevent miners from being potentially exposed to hazards if operators can wait to start their inspections until the miners begin work. But the inspector goes first, and is the eyes and ears, and is looking for hazards, and the miners follow right behind. Well, the idea of the inspector going out ahead of the miners first is not what the text of the regulation says. When a mining crew goes in to work... Let's do the inspection before or as work begins. Right, and if the inspector is beginning the inspection at the same time that miners begin work, miners work in crews of several people. They're spread throughout the working area when they go into the mine. The inspector can only look at what's right in front of him at a time. He can only inspect what's in his field of vision as he moves through the working area. But if the crew has gone to work, they're spread throughout the working area. If it takes the inspector a half hour to get through the entire area and look at everything, then that means that at least some miners would have been working exposed to any uninspected hazards for 25 minutes, almost 30, which is... There's no question that the as work begins standard could be implemented in a way that puts miners at risk, as you just said. But if it could be implemented, I mean, suppose they're all working in one area on a given day, and they go in together and the inspector goes first and is looking around as the miners are turning the machines on, or whatever it is they do when they start work. Well, the problem with thinking about... I think that we're getting towards the best practices discussion that MSHA made in their administrative record. The problem with that is that advisory best practices are not a substitute for binding rulemaking under 101A9. Yeah, I'm not talking about best practices. I'm just trying to imagine the agency says this can be implemented in a way that doesn't put miners at risk if the inspectors and the miners go in at the same time. And that doesn't seem facially unreasonable. Well, even if it were possible under certain circumstances to implement the rule in a way that would be as protective as the 2017 situation, the problem is that following the letter of the regulation and doing the minimum, you don't get there. The reason that we regulate the timing of inspections and other health and safety standards under the Mine Act is to provide a floor beneath which unscrupulous, less careful operators cannot fall. Those are the people you have to be concerned about. And if it's possible to follow the letter of the regulation and to provide less protection to miners than they had in 2017, then that means the text of the 2018 amendments violates the Mine Act. Just one more question along these lines, which is might it not be safer to have the inspector in the mine looking for problems at the time the workers are going in if you have a mine that is not 24-7, right? The inspectors go in as the miners are starting work. Suppose the hazard only manifests itself once the miners turn on the machine and start working. They might miss that if they go in before when the mine is down for the evening. Before I answer, I do want to note that we would like to reserve three minutes for our time for rebuttal. I see that I've passed the three-minute mark already. Go ahead. Thank you, Your Honor. I think that the important thing to note in response to that hypothetical is that if you're having miners go in to begin work in an area that has not been completely inspected before they get there, they have access only to piecemeal bits of information about hazards that the inspector collects as he moves through the working area. If the inspection is done before miners begin work, they have a complete picture of everything that the inspector found throughout the work area. So it's the difference between being given a fairly comprehensive report of what kind of potentially fatal hazards you might encounter in your work day versus going in with nothing. Let me ask you, do we know from the record how long these inspections take? I know you said 30 minutes. That's a hypothetical for the sake of this argument. I am not aware of any notation in the record of the average length of inspections, but I do know from experience in representing miners that working areas can be quite large, and it can take a considerable amount of time for any one individual to make their way through the entirety of it and look at everything. The concern that we have is for what happens to people who are working at the end, right? The last place that the inspector gets to, they've been there for 15, 20, 25, however many minutes it is, and you find evidence in the administrative record of people suffering fatal injuries and accidents from uninspected hazards in 10, 12, 13 minutes. It's enough to kill someone. Let me ask you this. Do we know from the record if the shift starts at 8 a.m. that the inspector is there at 7 a.m. under the old rule? I don't believe we know from the record. I mean, the 2017 rule was only effective for three days, so we don't have a whole lot of data to go on. Under the prior 1979 standard, operators had considerable leeway when they conducted inspections, and I'm not certain what the average lead time was. They could do it any time, couldn't they, under the 79? Yes, that's correct. Okay. I have a question. It's sort of a turn of art here. When we use the word inspection, what are we talking about? In other words, if you can use an analogy to I have my house cleaned, it is thoroughly cleaned, but it is also dusted. It is also mopped. So when we're talking about this type of inspection, are we talking about what I would call the comprehensive inspection as opposed to an area pinpointed inspection or a problem pinpointed inspection? In other words, just sitting on this court, I've seen all types of inspections described, so I want to know what we're talking about in this particular area that's at issue here. Yeah, of course. The type of inspection that we're talking about is much more similar to the comprehensive inspection that you described. The responsible, competent person under the regulations is supposed to go through the entire working area of the mine and look for any hazards that have the potential to endanger miners. Under the 2017 rule, that person was supposed to make a record of every hazard that they found under the 2018 rule amendments. They are able to avoid recording certain hazards as long as they're corrected within a quick period of time. But the goal of the inspection in any case is to get a comprehensive picture of what are the conditions underground or on the surface, depending on what type of mine it is. The nature of mining is that the physical environment is constantly changing. It's not like going into work in an office or even in a warehouse or a factory where things stay put. Equipment tends to do the same things day in and day out. You have a constantly shifting physical environment. You have constantly moving people and equipment. And so what the inspector is doing is taking a look at everything to say, okay, based on the conditions that we have today versus what they were on the last shift yesterday, what are the dangers that miners might encounter so that they can be aware of those and avoid being injured or possibly killed at work. All right. Thank you. Thank you, your honors. Ms. Eckert. Good morning. My name is Susan Eckert. I'm here on behalf of the United Steelworkers. I'm going to address the second part of the argument, which is the arbitrary and capricious challenge, which is, of course, tied into the Section 101A9 argument as well, since the statute and arbitrary and capricious decision-making work together to create a backstop for how the agency must review what is more protective or as protective under the language. We have asserted that the 2018 rule is not the product of reasoned decision-making because HMSA has not provided in the record adequate reasons for the changes they've instituted. And there are unexplained inconsistencies between the 2017 rule and the 2018 rule. The record for the 2018 rule does not support those changes and, in fact, at times contradicts the much more comprehensive administrative record that the agency put together for the 2017 rule. Secondly, we have asserted that HMSA is not entitled to an unfettered deference just because of its expertise and experience, where HMSA fails to explain what its experience is and how that has informed its policy. You have an arbitrary and capricious decision-making. This is particularly problematic here because HMSA previously relied on its expertise and experience in developing the 2017 rule. And so you see a contradiction because in 2017, of course, they're asking for deference to their expertise and experience, and now yet again they're asking for deference. And we don't believe it can be so wide open that we just give HMSA sort of a blank check to make changes as they believe, with a change in the new administration, with changes in priorities. Another agency might be able to make those kind of changes more freely, but here we've got that backstop of Section 101A9 that limits what an agency, in this case HMSA, can do, which is somewhat of an unusual situation compared to other more plain, vanilla, arbitrary, and capricious cases. They don't question your premise that the new rule needs to be at least as protective as the old rule. Correct. They say they meet that standard. Correct. Correct. HMSA has agreed that Section 101A9 is a requirement. So suppose, just for the sake of argument, assume that the new rule could be implemented in a way that is as protective of minors or not. Why wouldn't we give some weight to the agency's clear statement in the Federal Register that they want it implemented in a way that protects minors and I can imagine ways in which that might work, right? The inspector goes in the mine with the workers right behind him. Why wouldn't we give the agency the benefit of the doubt in that sense, whether it's our deference or crediting their explanation for State Farm purposes? That doesn't seem unusual. Right. I mean, I think in general an agency does get a certain level of deference, of course, but here, in effect, we have HMSA taking on HMSA. In 2017, we had findings that were made that said the safest way to conduct the exam is before. So I see a problem in that we do have this record from 2017 with studies and model rules and great detailed description of fatalities, and now we have HMSA contradicting itself. So we don't have to even look to what the STE workers might think is a good rule. We look to what HMSA has clearly said in 2017 is a good rule and rely on their own evidence and their own findings that are far more detailed. And once you look at that and buy into that, then it's very difficult to get to where we're at in 2018 without some additional analysis, information, something that says we understand that level of protectivity that we need and here's why we think it meets this new standard. Instead, we've got a sort of wishful thinking that we believe this will work, but we don't have to go there because they already created a rule that addresses it. So why go to this new place when we already know the answer and both the record keeping and the timing of the workplace examination, given the record, given a lot of testimony that we received in the 2017. And if you're going to go in a new direction, and I'm not saying that agency can't go in a new direction, and that especially when you have a change in administration, you're going to have different priorities and policies. But first of all, you've got to develop that record very, very thoroughly to do that when you've already developed a record that goes in a different direction. So I think we're at a different place than when we're starting with a blank slate, creating a new rule, what's good, what's bad. We've already set the floor, and so why are we going to go back to a place where there's more discretion and more opportunities for problems, given what we've already seen from MNCHA in deferring to its expertise from 2017. So in that regard, we would ask that you vacate the rule and return the 2017 rule in place, and we'd like to then reserve the time, as we said, for our rebuttal. Are there any other questions? Thank you. Thank you. Ms. Scott? Thank you. May it please the Court, Emily Toler-Scott for the Secretary. There are more than 11,000 metal and nonmetal mines in the United States. Some of them resemble the popular conception of a mine, vast networks of underground tunnels or surface operations that cover many square miles that employ hundreds of people to remove gold or silver or copper from the ground. But the vast, vast majority of metal and nonmetal mines are not like that. They may be small granite quarries where operators remove slabs of granite from shallow pits. They may be dredging operations that remove sand and gravel from riverbeds. They may be mobile operations where pieces of equipment are moved from worksite to worksite to crush stone on an ad hoc basis. More than 90 percent of all metal and nonmetal mines employ fewer than 20 people, and more than half employ fewer than five. Needless to say, developing standards that not only are workable at such diverse operations but also that will protect miners at such diverse operations requires expertise. MSHA has that expertise and, subject to the Mine Act and to the APA, is authorized to use it to determine how best to protect miners' safety and health. MSHA has done that in this case and has determined that the workplace examination rule will protect miners' safety and health. That determination complied both with the APA's requirement of reasoned decision-making and with the Mine Act's requirement that no standard can be less protective than one it replaces. The Court should not second-guess those determinations but should instead deny the petition for review. The problem is I don't see any explanation of why the new rule will necessarily be as protective as the old rule. I mean, this is very common-sense notion, right? You need to finish the inspection to find out what the hazards are to tell the miners before they go in and expose themselves, right? And there's a finding in 2017 that that's the way to make inspections as safe as you can. And you come along and water that down and the only explanation is some language in the Federal Register that maybe you can do this in a way that's as safe, but maybe you can't. Well, I have a few points in response to Judge Cassis. I think, first, I don't think it's necessarily accurate to characterize the as-work-begins option as watering down the standard. I mean, the standard does retain the before-work-begins options as well, and MSHA explained that in mines that do have dynamic mining conditions, this will enable operators to choose which works best at their mines. As you pointed out to Ms. Carter, there may be mines where it makes sense to do, or not makes sense, but is safer, in fact, to do the examination as miners begin work. And the language of the standard itself, in fact, if you look at it as a whole, makes it clear that the standard does not permit a situation where an entire crew of miners is coming into a workplace that has not been examined. Why not? Because the standard requires operators to either promptly correct hazards, that is, correct them before miners are exposed to them, or notify miners that there are uncorrected hazards so that miners can take them. Notify miners of hazards that were found by the inspection. Right. That's not going to help miners if they're going into five different places at once and the inspector is simultaneously going into one of those five places to start the inspection. But the critical language is promptly notify miners, and in the preamble MSHA explains that promptly means before miners are exposed. And so, yes, the language of the standard is as work begins, but that does not mean or that requires some distance as some distance in time between when the competent person is inspecting the working area and when other miners are coming in. It's not because it is not possible for a competent person to find hazards to notify miners about them unless there is some distance in time there. And so to comply with the other pieces of the standard. But either the notice requirement, you're saying the prompt notice requirement solves this problem. Either it doesn't because there will be hazards that are not found in time, or there's just this very strange operation of the regulation that you expressly change the before work begins standard to an as work begins standard, but that can never become effective because there's no way to give sufficiently prompt notice. I think that might be, I'm not totally sure that I understand your point, Judge Katz. I'll try to respond. I mean, let me try it again. If the notice, if you can't send miners into an area of the mine unless and until they are informed of hazards found by an inspector, then what your rule of before or as miners begin would seem to just collapse into the old rule of before work begins. I understand your point now, I think. And this goes to the realities of how mining is done. So I think if you read the preamble of the 2017 rule, it's clear that before work begins meant something like, or does mean something a little, with more distance in time. So maybe an examiner, if you have an underground mine, maybe a competent person goes underground in a separate vehicle and conducts the examination and then other miners come in. Or maybe, as Imcha explained, a person at the end of the previous shift in mines that operate around the clock could conduct the examination before the next shift comes in. This is different because it does permit, for example, miners to travel together on the same vehicle, and maybe it seems absurd that a competent person would then inspect the working place before other miners are exposed to hazards, but that is what Imcha is envisioning in this case. So it may seem like an academic distinction, but in practice it really does work. It makes sense. Yeah, I mean, I understand how it could work. The problem is all the details seem to be in the sub-regulatory statements that don't have the force and effect of law. And as your friends point out, the point of the safety rule is to have a codified floor. I think I would respectfully, I guess I can't really disagree because the standard does say promptly, and Imcha did not say before miners are exposed in the standard, that's true. But I also think that this goes to an important point, which is the one or the hazard against standard can't be considered in a vacuum. This standard works, or the as work begins exception works in tandem with the promptly corrected hazards, record-keeping exception as well, for purposes of 101A9. And I think Secretary's net effects determinations are ultimately what the court is reviewing, rather than looking at each piece in isolation. The question is, on the whole, is Imcha's determination, which it made, arbitrary and capricious, or did Imcha give adequate reasons for its ultimate determination that this standard, as a whole, satisfies 101A9? Let me ask you to envision this. What would be the difference if the inspection is done before a shift begins work, and there's a danger that is as obvious as flammable rags near too close to something, or the fact that some equipment has to be so many feet away? If the inspector goes in first and sees those flammable rags, as opposed to bringing in the miners, that's something that has to be fixed immediately, right? I think that would depend on the circumstances. I'm just thinking of cases I've sat on before, where the safety standards are very, very minute. Right. Yes, certainly. And that may well be an imminent danger, which is a separate provision of the standard itself. All right. So the difference between an inspector seeing these rags flammable near something they shouldn't be near, when the miners haven't begun work, and he says, get somebody in here, get those rags out of here, as opposed to the miners with him, and he's got to stop his inspection. I'm just thinking not only of safety, but just of efficiency, because I would think with something like that, and you can imagine other hazards that have to be remedied immediately before he continues with his inspection. I think that's true, that there are some hazards that have to be fixed immediately, imminent dangers, and those are addressed in subsection A2 of the new standard, which states that shall be brought to the immediate attention of the operator, and all persons are withdrawn. So if there is a severe hazard that would immediately have to be addressed, that might be an imminent danger, then there is a separate procedure under the regulation, or under the standard, that does take that into account. But promptly corrected hazard, but again, the standard requires the competent person to notify miners about a hazard. So let's say instead of flammable rags, there was a hose that posed a tripping hazard. It could either be picked up, corrected immediately so that miners aren't exposed to it, or miners could be notified about it. The competent person can either give a verbal warning to the miners, could post a warning sign that states the nature of the hazard. So these are the procedures that enable the inspector to get through the working area and to identify these hazards. And I also think, Judge Henderson, your point about the minute nature of safety standards, because the workplace examination standard is important, but it is not the only protection that MSHA has in place for miners. There are numerous, numerous overlapping safety standards that also impose separate obligations on operators to keep these hazards from existing in the first place, or to correct them when they find them. And so I don't think it makes sense to take this standard in a vacuum either, and assume that the only way that an operator or miners may know about hazards, or the only reason that they're obligated to correct hazards is this standard. That's simply not the case. And so miners are not just protected by this, but by the panoply of other MSHA standards that do regulate equipment, that require prompt maintenance of equipment, require it to be maintained in good condition, and even some standards that require every shift miners to examine equipment that they're going to be using during that shift. So in reality, there are all these overlapping protections that also keep miners safe. And we haven't talked much about the promptly corrected hazards exception, but I did want to make just one point about the idea that it reduces safety because it reduces miners' ability to identify trends. The 2017 rule makes it clear that the benefit of identifying trends was primarily for mine operators' benefit, and was not the primary safety focus of the rule. It is obviously safety-related. Being able to identify recurrent or significant hazards is certainly safety-promoting. But the true core safety for miners' provisions were requiring the examination to be done at a specified point in the shift, requiring notification of miners, and requiring prompt correction of hazards. That's really the meat that protects miners in this case. And even if identifying trends were a significant safety aspect, again, it is important but not the central one, MSHA explained in the preamble, acknowledged that, yes, this is a benefit. We did in 2017 identify this as a benefit. MSHA still thinks this is a benefit, but determined overall that incentivizing operators to correct more hazards promptly so that there will be fewer hazards, that allowing operators to conduct examinations at a time that may take into account change or that may eliminate the risk of change conditions, taken together, all of these things as a whole ultimately satisfy Section 10189's requirement that the standard be at least as protective as the one it replaced. I also just want quickly to make a note about remand. This wasn't briefed extensively or in any depth, but the unions did argue that the rule should be vacated rather than remanded. So if the Court is concerned that MSHA did not adequately explain its safety determinations or that MSHA did not adequately explain its determinations, that was not the kind of fundamental deficiency. It was not so abrupt a departure from prior practice without explanation that would justify vacating the rule as a whole. If we do vacate, what is your position on what would then apply, the 2017 or the 1979? The 1979 standards were the last one that were in effect. I think what would go into effect would depend on the scope of the Court's order, whether the Court vacated only pieces of the standard or the standard itself. I think that is a more complicated question that I think I'd prefer to answer after seeing an order. Why wouldn't there in the preamble, the 2017 stayed until June 2nd of 2018, and then this new rule came along? If we vacate that new rule, then why wouldn't that resurrect the 2017 that was only stayed until June 2nd? No, I think that's right, and I may have misspoken. The 1979 standards were briefly reinstated until the 2017 standards came back into effect. So I think if the Court did vacate, then that rule would come into effect again as well. But, again, I do just want to emphasize that the MSHA at least — then is it your position that the 2017 rule amendments become effective? It is. All right. So just one more question on remedy. Do you — we have two different provisions we've been talking about. Do you view them as severable? Well, I mean, let me just put my cards on the table. Suppose I think that the before work versus as work, as work begins provision is problematic, but the record-keeping change is not problematic. Would we just vacate the one, or would we vacate the entire amendment? My short answer to the question is, yes, they're severable, because the question for severability is whether the agency would have adopted any one part in the absence of the other. That may be a question that is more complicated and might benefit from fuller briefing, but the short answer to that question is yes. Based on what? And I suppose, Judge Rogers, this goes into why it's a little complicated. Well, you have a reason, presumably, for your answer. Yeah, the reason for my answer is that while the two — well, let me give you some context. So our argument is that, overall, the net effects of this rule justify compliance or establish compliance with Section 101A9. So MSHA does think about them in those terms as being intertwined. That being said, there's no indication in the agency record that MSHA would not have undertaken or not have adopted one in the absence of the other. MSHA analyzed them. And there's no indication to the contrary, is there? That's — yes, I suppose that's accurate and may go to a question of the necessity of fuller briefing. I mean, I had read the explanation for the record-keeping change, which is greater incentives to fix hazards, and you don't want to overwhelm the records with unhelpful information. I had thought that stands on its own bottom and indicates non-severability. I'm sorry, indicates severability. But when you started talking about net effects and the rule working together, it seems to cut the other way. I agree that those things are our intention and may benefit from fuller briefing, or if nothing else, perhaps even a remand for MSHA to explain whether it views them as severable. All right. Thank you. Does Ms. Cohn have any time for me? All right. Why don't you take two minutes? I'd like to touch briefly on a couple of points to wrap up today. The first is the record-keeping portion of the rule that you were just discussing with MSHA counsel. This safety change is just as significant in the petitioner's minds as the inspection timing change was. When operators decline to record all hazards, decline to record those that are promptly corrected, it is much more difficult for them to recognize patterns of recurring hazards in the mines and to take the steps that are necessary to interrupt those patterns. Recurring small hazards not only endanger miners on a repeat regular basis, but they can indicate that more serious hazards are likely to develop. The operators need records of all of these hazards so they can take appropriate action, and just as importantly, miners' representatives need access to complete records so that they can hold the operators accountable for taking the right steps to correct these issues. Now in 2017, MSHA made the reasoned fact-based finding that recording all hazards was the only way for operators to adequately protect their workforce, and that reducing record-keeping burdens would not incentivize operators to correct issues more quickly. In 2018, MSHA reversed course without any factual findings, new data, studies, explanation of any kind, saying precisely the opposite, that now in 2018, allowing operators to refrain from recording all hazards would, in fact, incentivize quicker corrections. This, just like the unsupported change to inspection timing, is not a factual finding. It's not based on any real analysis. It's not entitled to any deference. It's just MSHA's wishful thinking about how this might operate under the best possible circumstances. But MSHA has to live and die by the text of the rule as it's written. Miners certainly have to live and die by that text, and we need to look at what are the minimum requirements here, what does this rule at bottom require operators to do and not do, and the fact that MSHA made precisely the opposite safety determinations in 2017. Those are the ones to which the agency must be held. I see that my time has expired. Thank you, Your Honors.