Join the GOOGLE +Rubber Room Community

Monday, April 18, 2022

Time To Change the Endless Teacher Evaluations That Make No Impact on Student Outcomes

 


Dear readers,

I think that Peter Greene (see below) has a point that needs to be addressed in New York City by the Department of Education VIPs - the attorneys, mayor, Commissioner, and all others - who have put into place a fraudulent teacher discipline procedure that punishes teachers who receive "developing" or "ineffective" ratings on their formal and/or informal observations.

I have posted my opinion on this blog many times on the way principals, assistant principals, and evaluators (remember PIP+?) rate teachers on observation reports and then charge them with 3020-a and pursue their termination. I have collected most of the decisions made at 3020-a since 2007, and I have found that most arbitrators agree with the assessment/rating given by the school administrators. The ratings are taken as evidence and fact, not opinions. I think this needs to be changed: arbitrators must stop finding ratings credible without looking behind the numbers and at all circumstances in the storyline.

At 3020-a, we at Advocatz always go into the backstory of evaluators - who paid them, is there any animosity between the administrators at the school and the teacher, etc. The evaluator is there to create a paper trail of incompetence for the administrator and must agree with whatever the principal wants the evaluator to write. Teachers/staff often tell me that the evaluator may say "Well done!" at the end of a lesson but never put that in writing or put something completely different in the actual observation report. Often, the evaluator knows that if he/she puts his/her honest opinion in the report (i.e., that the teacher is terrific), then he/she may be fired - it's the evaluator's job at stake or yours. You, the teacher, won't win.

And then along came Charlotte Danielson and her teaching rubric, which was erroneously used as a benchmark for rating teachers, and failed miserably to point to any standard of good teaching, only the whim of an evaluator who wanted a teacher fired. Evidently, Ms. Danielson herself was not happy with the way her rubric is used in New York City. 

The problem is, the NYC Department of Education will not admit to any wrongdoing, and never allows anyone to suggest changes in the way their business is done.

Nonetheless, what needs to be done 

New research shows that endlessly evaluating teachers has made no impact on student outcomes.

BY , Progressive.org,

 

For several decades, everyone from President George H.W. Bush to then-Arkansas Governor Bill Clinton believed it was possible to measure outcomes (“deliverables,” some called them) to separate the educational wheat from the chaff. Teachers would be held accountable for real data—numbers generated by tests.

No Child Left Behind (NCLB) had its bipartisan birth in 2001. Central to the law was not only to collect test scores but to break them down by subgroups. This meant that, at least on paper, Black students or low-income students would not have their struggles hidden in a school average score. 

“Effective” schools would be those that pulled high test scores from all students. NCLB had data and a deadline: By 2014, all schools would have all students scoring above average on the Big Standardized Test. This was, of course, a statistically impossible goal.

The Obama Administration moved past the unachievable goals of NCLB, but held onto the belief in data. Tests would be used to collect data, and that data would be used not only to judge schools but also to evaluate individual teachers. Students across the nation would take Common Core-aligned tests, and then teachers would be judged on the results.

Educators like me were subjected to complicated professional development programs, where we learned that models could predict what students would have scored in some parallel universe with an imaginary “neutral value” teacher.

The plan was, as one presenter told me, like predicting the weather. And if the student scored higher or lower than their hypothetical counterpart, that difference could be chalked up to either the teacher’s credit or blame.

Education reformers were convinced they had unlocked the secret of identifying good and bad teachers and, as a result, revitalizing schools. Once identified, “good” teachers could be moved to struggling schools that needed them, incentive programs could be created to pay the better teachers, and “bad” teachers could be fired.

Position papers like “The Widget Effect” called for a future where teachers could be differentiated by their ability to improve student achievement. Critics of public education complained that teacher evaluations produced too many satisfactory ratings, and a new system of high-stakes testing would “toughen” the evaluation systems.

This new system would mimic the stack ranking of industry giants like Microsoft, where employees were ranked according to desired metrics. The bottom rung of the employee ladder would be dropped, and the company could fire its way to excellence.  


But, as a recently released working paper from the Annenberg Institute at Brown University underscores, the “massive effort to institute high-stakes teacher evaluation systems” had essentially no effect on “student achievement.” Though high-stakes teacher evaluation was supposed to raise test scores all across the nation, student achievement—measured by test scores—barely budged.

“Following the data,” it would seem, has led us far astray. The experiment has failed.

Why? I can think of several reasons.

Reducing the job of teaching to “get students ready to take a single math and reading test” was insulting and demoralizing. It was not what teachers had signed up for. Instead of using professional expertise and autonomy to create instruction to meet the many varied needs of students, teachers were increasingly required to deliver test prep.

The high-stakes testing also had the effect of warping education. Another new working paper shows that by focusing on math and reading, schools actually reduced education in other subject areas.

High-stakes testing also led to upside-down schools—schools that, instead of helping to meet all students’ needs, were now there to serve the need to produce test scores. More fundamentally, however, the very premise of the test-driven system was flawed: The notion that a great teacher is always a great teacher in all schools for all students on every day of their career doesn’t reflect reality

It’s more akin to any other human relations-based role. I might be a great partner for my spouse but not a great partner for every other person on the planet. My “spousal effectiveness” is not a static quality that can be measured like height or hair color. It is the same for teacher effectiveness.

Beyond that flawed premise, the single biggest problem with this data-based system is that it runs on bad data. The tests themselves are a black box. Teachers in many states are forbidden to see the content of the test and, therefore, can never know where their students came up short. Worse yet, since not all educators teach reading or math, many teachers were evaluated based on the scores of students they didn’t teach.

The data that was intended to drive the system came solely from the student tests, and the tests are not good. Questions are often poorly designed; one famous bad set required students to answer questions about a talking pineapple. A poet found herself unable to answer test questions about her own poem; how could students be expected to do any better?

Standardized test results were supposed to provide useful data about student learning, but they are closely tied to the wealth and whiteness of families. To try to turn test scores into more usable data, value-added scores (based on a formula used for farming) were introduced, but these turn out to be random and unpredictable—the data is so bad it has been thrown out by courts.

Still, supporters have argued that higher test scores mean greater success in life. That’s a dubious finding—correlation is not causation, and socioeconomic background is a good predictor of both test scores and life outcomes. Nor does evidence support that raising a test score raises a student’s life outcomes, which would be a far more important insight. It is no wonder that a system running on junk data has produced undesirable results.

In the meantime, private industry giants like Microsoft dropped stack ranking, realizing that it bred a bad workplace atmosphere, stopped innovation, and could not always be trusted.

And yet, there are still education reformers who cling to this high-stakes testing teacher evaluation model.

Current discussions about learning loss due to the pandemic often assume that high-stakes tests must serve as the best measure of what students are missing. Amid our current pandemic disruption, some are still suggesting that high-stakes testing remains our best bet.

But we have tried to guide education with high-stakes standardized testing data for two decades now, and yet we still have little evidence that this approach actually works. It’s the great irony of the modern school reform era—we must follow the data, except when the data shows that our systems don’t work.

English teacher for a few years. Blogger at Curmudgucation.

READ MORE BY PETER GREENE

Sunday, April 17, 2022

UFT Press Releases and Current News From the NYC Department of Education on COVID

 

Mike Mulgrew, UFT President


New York Teacher

UFT reacts to change in vaccine mandate for athletes and performers

 
 Press Releases

On March 24, 2022, Mayor Adams announced that athletes and performers would no longer be subject to a COVID vaccine mandate.

In response, a UFT spokesperson issued the following statement: 

Vaccinations are a critical tool against the spread of COVID, and the city should not create exceptions to its vaccination requirements without compelling reasons. If the rules are going to be suspended, particularly for people with influence, then the UFT and other city unions are ready to discuss how exceptions could be applied to city workers.

Related Topics: Coronavirus
The UFT has compiled answers to our members' most frequently-asked questions about the upcoming 2021-22 school year. We will continue to update this page as more information and guidance is determined

Coronavirus Hub 

The UFT is consolidating all the information about the impact of the COVID-19 outbreak on New York City public schools in our coronavirus info hub.

Rapid tests to be distributed ahead of Spring Break: Both staff and students will receive additional at-home rapid test kits prior to the start of the spring break vacation and are strongly encouraged to use the kits prior to returning to school on Monday, April 25. See the latest Principals Digest excerpt for more information. 

Update on masking for students aged 2-4: Mayor Adams announced that students younger than five and the staff that work with them should continue to wear face masks in school due to a recent uptick in COVID-19 cases across the city, and not go mask-optional on April 4 as previously stated. We will inform you when additional guidance is released.  

Masks will now be optional for K-12 students and staff as of Monday, March 7: Face coverings will become optional, indoors and outdoors, on NYC public school grounds and on school buses, for K-12 students and staff. Pre-K staff and students will be required to remain masked indoors. Anyone who chooses to wear a mask may continue to do so. See the updated DOE guidance on face coverings » and our safety FAQ »

School outdoor mask mandate lifted: As of Monday, Feb. 28, 2022, masks will no longer be required when outside on school grounds. Read the DOE's message to families » and the DOE's guidance on face coverings » 

Increased in-school testing: Beginning Feb. 7, schools will test the larger of either: 10% of the school’s student enrollment in grades 1-12, up to a cap of 250 students; or, 20% of the school’s unvaccinated student population. This change allows schools with a high student vaccination rate to test a larger number of students every week. Up to 10% of staff who wish to be tested will continue to be able to get a PCR test from the testing team each week. See the DOE's guidance on COVID-19 testing in schools »

Spring break arbitration decided: UFT members who worked over spring break of 2020 will receive a vacation day — a brand-new type of day that members have never had before — for each day they worked. Read the arbitrator's decision »  Read our FAQ »

Time off for boosters and children's vaccines: Members may take up to four hours off for travel and receipt of a booster vaccine, or to accompany a child to receive each vaccine dose without a charge to CAR balances. Read the memo »

Guidance from the DOE

Executive Orders and COVID Mandates Will Be Reviewed by the 11th Circuit Court of Appeals in Georgia v Biden


Re-posted from ADVOCATZ.com

I am asked almost every day by an employee of the NYC Department of Education,  "how can I be fired after 2-35 years of stellar work at the Department as a teacher/assistant principal/secretary (or other) simply because I requested a medical/religious exemption from getting the COVID vaccine?"

My answer is, this is a manifest disregard of Constitutional rights and Law, and let's create your fight strategy. I am not an attorney, just a person who loves reading law books, and this is not legal advice, only my opinion.

COVID vaccine mandates are being litigated throughout New York City and the United States, right now. The battle is too much for some, and they resign, retire, or leave their jobs, never to return. I certainly respect their opinion as to what is right for them.

But it saddens me to think that NYC will 'get away with' instigating the biggest and baddest mass termination of terrific, hardworking men and women in history, to please a select few who happen to be politically connected to Big Money. And, in the NYC DOE, as soon as someone writes Human Resources to say he/she cannot get vaccinated, the employee's fingerprints are immediately placed on the Problem Code, a database which is sent to the FBI and New York State Division of Criminal Justice services ("DCJS). 

See the Affirmation of Mallory O. Sullivan

Georgia v Biden: Legality of Executive Mandates Requiring Vaccination Up For Review By the 11th Circuit Court of Appeals


According to various news sources, there are more than 1,000 New York City teachers, firemen/women, policemen/women and other City workers who have been fired because they asked for a medical or religious exemption from getting the COVID vaccine and were denied. These denials come in the form of an email with a box at the bottom with an X next to “denied”. See the SAMS Barry Peek denial. 

But in the U.S., several Courts are putting a hold on vaccine mandates and test-or-mask mandates, as I quote from the analysis below (see also the ORDER December 7, 2021):

“No matter the policy benefits of a well-intended regulation, a court may not enforce it if the agency’s reach exceeds a statute’s grasp. Once before, in the throes of another threat to the country, the executive branch claimed it needed to seize control of the country’s steel mills as a “necessary” measure “to avert a national catastrophe.” Youngstown Sheet & Tube Co. v. Sawyer , 343 U.S. 579, 58272 S.Ct. 86396 L.Ed. 1153 (1952). But that threat, like this one, did not permit the second branch to act without authorization from the first branch. Id. at 588–89, 72 S.Ct. 863. As the Supreme Court recently explained in invalidating an eviction moratorium promulgated by the Center for Disease Control, “our system does not permit agencies to act unlawfully even in pursuit of desirable ends.” Ala. Ass’n of Realtors , 141 S. Ct. at 2490. Shortcuts in furthering preferred policies, even urgent policies, rarely end well, and they always undermine, sometimes permanently, American vertical and horizontal separation of powers, the true mettle of the U.S. Constitution, the true long-term guardian of liberty. ”

and,

” Additionally, requiring compliance with EO 14042 would likely be life altering for many of Plaintiffs’ employees as Plaintiffs would be required to decide whether an employee who refuses to be vaccinated can, in practicality, be reassigned to another office or another task or whether the employee instead must be terminated. “[A]ny abstract `harm’ a stay might cause . . . pales in comparison and importance to the harms the absence of a stay threatens to cause countless individuals and companies.” BST Holdings, 17 F.4th at 618. Accordingly, the Court finds that the balancing of the harms weighs heavily in favor of enjoining the enforcement of EO 14042.

E. Public Interest

For similar reasons, a stay is firmly in the public interest. From economic uncertainty to workplace strife, the mere specter of [EO 14042] has contributed to untold economic upheaval in recent months” and “the principles at stake when it comes to [EO 14042] are not reducible to dollars and cents.” Id. at 619.”

The solution to this problem is to give everyone a fair hearing where the unvaxxed get to tell their story to a neutral person or panel and are given the right to a job at an equal salary to what they were making before COVID, but in a remote location if vaccination is not possible.

We know that placing an employee away from students is entirely within the NYC DOE’s authority and they do this all the time. See the following cases:



Queens principal booted for fraud will get nice paycheck for 7 years

Susan Edelman, NY POST, April 16, 2022

A Queens principal accused of using fraudulent schemes to boost his school’s graduation rate can never again work with city students — but will get a $1.8 million desk job, The Post has learned.

Khurshid Abdul-Mutakabbir, who was removed as principal of Maspeth High School last July, won’t return to any city school as a principal, according to a settlement of misconduct charges. But he can stay on the Department of Education payroll for another seven years.

Under Abdul-Mutakabbir, Maspeth HS created fake classes, awarded credits to failing students, and fixed grades to push kids out the door, the Special Commissioner of Investigation for city schools found, confirming exposès by The Post.

Instead of trying to terminate Abdul-Mutakabbir, as city investigators recommended, the DOE settled the charges on Jan. 25 by fining him $12,000  – and barring him from working as a principal.

But under the sweetheart deal – which DOE officials kept hidden for months – the disgraced educator, now age 47, will sit in an office until he “irrevocably” retires on Nov. 30, 2029.

See more

and then there is Santiago Taveras, who changed grades and then was ousted as Principal and “promoted” to a Mentorship position. (Just so you know, I like Mr. Taveras, who I asked many years ago to appear as a witness for my client, the Respondent at a 3020-a, and he gave excellent testimony which helped us win the case).

Lucio Celli, who I met in 2008 in the Bronx rubber room reassignment center and who threatened to kill three Federal Judges in 2018, was put into prison and now is on house arrest for two years, yet still receives his salary of $122,424 (according to seethroughny.net) and has never been brought to a 3020-a disciplinary arbitration.

And yet, excellent, hard-working educators both tenured and untenured who requested either medical or religious exemptions from getting the vaccine while the COVID vaccination mandate is in place have been fired because their request for telework from a remote location was deemed an “undue hardship” by the Department?

This is truly detestable and the NYC DOE should be held accountable for the highest amount of damages possible under the Law.

This is Accountability, and the NYC DOE gets an F.

Betsy Combier



betsy.combier@gmail.com
Editor, ADVOCATZ.com
Editor, ADVOCATZ Blog


Eleventh Circuit to Review Nationwide Injunction Against Vaccine Mandate for Federal Contractors

By Camila Laval, J.D. on April 15, 2022 8:31 AM

The coronavirus pandemic may be winding down, but there are still cases pending on the legality of executive mandates requiring vaccination for different types of employees. One of them, Georgia v. Biden, is currently before the 11th Circuit Court of Appeals, and its resolution will determine the fate of President Joe Biden’s executive order mandating vaccination for federal contractors.

Federal Contractor Vaccine Mandate

In September 2021, President Biden issued Executive Order 14042, which mandated the previously created Federal Workforce Task Force to provide guidance about COVID-19 protections for federal contractors and subcontractors.

President Biden claimed that he was acting under the powers vested in him by the Federal Property and Administrative Services Act and the president’s general delegation powers. He argues the objective of the order was to “promote economy and efficiency in procurement by contracting with sources that provide adequate COVID-19 safeguards for their workforce.”

The Task Force issued its guidance two weeks later, which required federal contractors to be fully vaccinated by Jan. 18, 2022.

Georgia District Court Grants a Nationwide Injunction

Soon after the issuance of the vaccination mandate, seven states (along with some of their governors) and several state agencies filed a federal lawsuit in Georgia, seeking an injunction against the executive order. A construction industry trade association that represented contractors who frequently worked on federal contracts was later allowed to join.

The district court held that the plaintiffs were likely to succeed on the merits of their claim that President Biden exceeded the authorization granted by Congress in the Act when issuing the executive order, and thus granted the injunction.

According to the court, the order brought about an enormous expansion in regulatory authority without clear congressional authorization, went beyond addressing administrative and management issues to promote efficiency and economy in the procurement process, and operated as a regulation of public health that was not clearly authorized under the Act.

The court not only granted the injunction but made it nationwide in scope. The judge reasoned that the trade association had members all over the country and that a narrower injunction would have caused more confusion given the extent of the association’s membership and the number of contracts involved.

Arguments on Appeal

The case is currently pending before the 11th Circuit Court of Appeals following an appeal by the U.S. government. The government argues that the spread of COVID-19 would slow down transactions of companies that do business with the government and thus justifies the vaccination mandate. In fact, the government argues, many private businesses have imposed similar vaccination requirements to advance economy and efficiency. The state plaintiffs counter that the Act only grants powers to impose conditions on the government’s general operations rather than on the contractors’.

Regarding the scope of the injunction, the federal government argues that there was no evidence of how many trade-association members faced imminent harm by the order and that the fact that there were similar cases pending in different jurisdictions made it unnecessary for the court to issue such a broad injunction.

The Fate of the Other Mandates

The vaccine mandate for federal contractors is just one of many vaccination requirements imposed by the Biden administration. Last year, the executive branch also ordered employees working for private businesses with more than 100 employees and healthcare workers to be fully vaccinated to keep their jobs. So far, only the healthcare worker mandate has survived judicial scrutiny. Since coronavirus infections have declined, maybe so has the strength of the arguments that the mandates further the economy and efficiency of the procurement process. As is often the case, courts will have the last word.

Stinson - Government Contracting Matters

Georgia v. Biden Places Nationwide Stay on the Government Contractor Vaccination Mandate

December 8, 2021

Breaking news – Following an expedited briefing schedule and hearing, the United States District Court for the Southern District of Georgia, in Georgia v. Biden, No. 21-cv-00163, has issued a nationwide preliminary injunction staying implementation of the government contractor vaccination mandate under Executive Order (EO) 14042. The Court determined that the States of Georgia, Alabama, Idaho, Kansas, South Carolina, Utah, and West Virginia, and the Associated Builders and Contractors, Inc. had standing to bring suit against the vaccination mandate. In its decision the Court held:

even in times of crisis this Court must preserve the rule of law and ensure that all branches of government act within the bounds of their constitutionally granted authorities. Indeed, the United States Supreme Court has recognized that, while the public indisputably “has a strong interest in combating the spread of [COVID-19],” that interest does not permit the government to “act unlawfully even in pursuit of desirable ends.” Ala. Ass’n of Realtors v. HHS, 141 S. Ct. 2485, 2490 (2021) (citing Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 582, 585–86 (1952)). In this case, the Plaintiffs will likely succeed in their claim that the President exceeded the authorization given to him by Congress through the Federal Property and Administrative Services Act when issuing Executive Order 14042.

Finding that the President exceeded his authority granted under the Federal Property and Administrative Services Act (FPASA), the Court declined to address the other potentially viable grounds alleged by Plaintiffs to support the grant of a preliminary injunction, including (1) the Safer Workforce Task Force Guidance and FAR Deviation Clause were issued without complying with the Administrative Procedure Act’s notice and comment requirements and (2) if the FPASA does authorize EO 14042, then both are unconstitutional under the non-delegation doctrine and because they exceed Congress’s authority and intrude on state sovereignty.

This national ruling follows on the heels of last Friday’s preliminary injunction granted on similar grounds by the United States District Court for the Eastern District of Kentucky that was limited to the States of Kentucky, Tennessee and Ohio.

The impact of this ruling cannot be understated as many government contractors are working to persuade their employees to commence vaccination in order to be fully vaccinated by January 18, 2022. Moreover, as the Court noted, the administrative processes required of Plaintiffs to comply with the requirements of the vaccination mandate are “incredibly time-consuming processes… complying with a regulation later held invalid almost always produces the irreparable harm of nonrecoverable compliance.” It further found that:

In contrast, declining to issue a preliminary injunction would force Plaintiffs to comply with the mandate, requiring them to make decisions which would significantly alter their ability to perform federal contract work which is critical to their operations. Indeed, it appears that not granting an injunction could imperil the financial viability of many of ABC’s members. Additionally, requiring compliance with EO 14042 would likely be life altering for many of Plaintiffs’ employees as Plaintiffs would be required to decide whether an employee who refuses to be vaccinated can, in practicality, be reassigned to another office or another task or whether the employee instead must be terminated. “[A]ny abstract ‘harm’ a stay might cause . . . pales in comparison and importance to the harms the absence of a stay threatens to cause countless individuals and companies.”  (Citations omitted).

We will further analyze the impacts and follow up actions of the Administration in light of these rulings.  Stay tuned.

Below is an excellent argument against Big Government overstepping their authority to issue a national vaccine mandate in the case of Mass. Bldg. Trades Council v. U.S. Dep’t of Labor (In re MCP No. 165, Occupational Safety & Health Admin.), 20 F.4th 264 (6th Cir. 2021)

DISSENTING FROM THE DENIAL OF INITIAL HEARING EN BANC

SUTTON, Chief Judge, dissenting from the denial of initial hearing en banc.

When much is sought from a statute, much must be shown. The Secretary of Labor asks a lot of the Occupational Safety and Health Act. He claims authority to issue an emergency rule, scheduled to go into effect on January 4, 2022, that will require roughly 80 million workers to become vaccinated or face a weekly self-financed testing requirement and a daily masking requirement. At the same time, he assumes authority to regulate an area—public health and safety—traditionally regulated by the States. If valid, the rule would nullify all contrary state and local regulations, as the power to regulate nationally is the power to preempt locally.  Such broad assertions of administrative power demand unmistakable legislative support. The federal courts “expect Congress to speak clearly when authorizing an agency to exercise powers of ‘vast economic and political significance’ ” and to use “exceedingly clear language if it wishes to significantly alter the balance between federal and state power.” Ala. Ass’n of Realtors v. Dep’t of Health & Hum. Servs. , ––– U.S. ––––, 141 S. Ct. 2485, 2489210 L.Ed.2d 856 (2021) (quotation omitted).

Congress did not “clearly” grant the Secretary of Labor authority to impose this vaccinate-or-test mandate. First , as a threshold matter, the Occupational Safety and Health Act gives the Secretary power to address only occupational health and safety risks. But it is by no means clear that this authority extends to all hazards that might affect employees at some point during the 16 hours of each weekday and the 48 hours of each weekend when they are not at work, whether the hazard arises from a coronavirus of one sort or another, a virulent flu, traffic safety, air pollution, vandalism, or some other risk to which people are equally exposed at work and outside of work. It is one thing to tell a worker to don a mask at the start of a hazard-filled shift and doff it at the end. It is quite another to tell a worker to vaccinate on the basis of a risk that exists whether he is on the clock or off and that amounts to a medical procedure that cannot be removed at the end of the shift. Confirming the point, the Secretary of Labor has never imposed a vaccine mandate or for that matter a vaccinate-or-test mandate on American workers. The Act does not clearly give the Secretary power to regulate all health risks and all new health hazards, largely through off-site medical procedures, so long as the individual goes to work and may face the hazard in the course of the workday.

Second , even apart from the workplace-anchored scope of the Act, the Secretary of Labor’s power to issue “emergency temporary standards” does not justify the first vaccinate-or-test mandate in federal labor law history. This emergency power extends only to “necessary” measures, namely measures indispensable or essential to address a “grave” danger in the workplace. But this set of preconditions does not apply (1) when the key population group at risk from COVID-19—the elderly—in the main no longer works, (2) when members of the working-age population at risk—the unvaccinated—have chosen for themselves to accept the risk and any risk is not grave for most individuals in the group, and (3) when the remaining group—the vaccinated—does not face a grave risk by the Secretary’s own admission, even if they work with unvaccinated individuals. Countless lesser and more focused measures were available to the Secretary: targeting certain industries susceptible to high risk, focusing on protections for workers most vulnerable to the virus, and varying any requirements to account for the wide range of settings in which people work. A blunt national vaccine mandate for 80 million workers with little regard to the relevant employment circumstances—well-spaced or not, together or apart, high risk individuals or not, indoors or mainly outdoors—was not necessary under the Act, and Congress did not clearly say otherwise.

Third , the setting of these requirements—authority to set “emergency temporary standards” without complying with the notice-and-comment process—confirms the narrowness of this authority and its inapplicability here. Start with “emergency.” The Secretary does not invoke this power based on a sudden revelation that the virus presents a serious health risk. How could he? He relies on something else—the increased availability of vaccines.  That development, however, does not heighten health risks; it alleviates them—and it’s hardly a new development anyway. What, moreover, is “temporary” about a vaccination ? A reluctant or coerced vaccination cannot be undone if the Secretary changes course during the notice-and-comment process or if the proposed rule exceeds the Secretary’s authority. All of the Secretary’s emergency decrees to date, even the ones invalidated by the courts, have involved truly temporary measures to protect workers from certain hazards at work until the notice-and-comment process ends. Ready access to free vaccinations may not have quelled the pandemic as quickly as the Secretary, or any of us, would like. But that reality does not justify, much less justify clearly, a sudden invocation of an emergency medical power at roughly the two-year anniversary of the pandemic merely because the Secretary determines that not enough Americans are vaccinated.

For my part, the resolution of this conflict between existing law and the Secretary’s proposed policy is not particularly hard. What makes the case difficult are the ongoing challenges of the pandemic and the health-and-safety benefits of obtaining vaccinations. The challenges presented by the pandemic are serious, no one can deny. The record confirms what common experience shows—”that the public has a strong interest in combating the spread” of a virus that has prematurely ended over three-quarters of a million American lives. Ala. Ass’n of Realtors , 141 S. Ct. at 2490. The record also shows the utility of vaccinations. The medical studies to date show that vaccinated individuals face fewer risks of getting the virus and, for those who still suffer breakthrough infections, fewer risks of serious symptoms or death. It is the rare federal judge, indeed the rare employee in the third branch, I suspect, who has not gotten the message.

But the issue here is not that simple. No matter the policy benefits of a well-intended regulation, a court may not enforce it if the agency’s reach exceeds a statute’s grasp. Once before, in the throes of another threat to the country, the executive branch claimed it needed to seize control of the country’s steel mills as a “necessary” measure “to avert a national catastrophe.” Youngstown Sheet & Tube Co. v. Sawyer , 343 U.S. 579, 58272 S.Ct. 86396 L.Ed. 1153 (1952). But that threat, like this one, did not permit the second branch to act without authorization from the first branch. Id. at 588–89, 72 S.Ct. 863. As the Supreme Court recently explained in invalidating an eviction moratorium promulgated by the Center for Disease Control, “our system does not permit agencies to act unlawfully even in pursuit of desirable ends.” Ala. Ass’n of Realtors , 141 S. Ct. at 2490. Shortcuts in furthering preferred policies, even urgent policies, rarely end well, and they always undermine, sometimes permanently, American vertical and horizontal separation of powers, the true mettle of the U.S. Constitution, the true long-term guardian of liberty.

For these reasons and those elaborated below, the challengers are likely to prevail on the merits when it comes to their petitions targeting the emergency rule. That reality together with the other stay factors show that the emergency rule should remain stayed. Nken v. Holder , 556 U.S. 418, 434129 S.Ct. 1749173 L.Ed.2d 550 (2009) ; Ala. Ass’n of Realtors , 141 S. Ct. at 2490.

I.

Congress passed the Occupational Safety and Health Act in 1970. Pub. L. No. 91-596, 84 Stat. 1590. With the Act, Congress created an agency to administer the statute—the Occupational Safety and Health  Administration, called OSHA for short—which sits within the Department of Labor. From the outset, the Act was designed to ensure “safe and healthful working conditions” for employees. 29 U.S.C. § 651(b). The Act empowers the Secretary of Labor, through OSHA, to create health and safety regulations for workplaces across the country. Id. § 655(b). Before such regulations go into effect, they must withstand a rigorous process. The Secretary must provide notice of any proposed regulation and give 30 days for any affected entity to submit data or offer comment about the costs, benefits, feasibility, legality, or any other reason for rejecting, adopting, or modifying the proposed rule. Id. § 655(b)(2). Those who object to the rule may request a public hearing. Id. § 655(b)(3). Within 60 days of the end of the period for submitting comments or the completion of a requested hearing, the Secretary must publish a rule or decide not to issue one. Id. § 655(b)(4). Still more process is called for if the proposed rule involves, as this one allegedly does, “toxic materials or harmful physical agents,” in which case its development must be “based upon research, demonstrations, experiments, and such other information as may be appropriate.” Id. § 655(b)(5).

An exception exists. The Act allows the Secretary to create an “emergency temporary standard” without undergoing all of these notice-and-comment requirements. Id. § 655(c). To allow an “emergency” regulation to go into immediate effect, the Secretary must show (1) that “employees are exposed to grave danger from exposure to substances or agents determined to be toxic or physically harmful or from new hazards,” and (2) that the “emergency standard is necessary to protect employees from such danger.” Id. § 655(c)(1).

Since 1970, the Secretary of Labor has used these emergency powers infrequently—and never to require a medical procedure. Over more than a half-century, the agency has used this power just nine times before this year. BST Holdings, L.L.C. v. Occupational Safety & Health Admin. , 17 F.4th 604, 609 (5th Cir. 2021). Six of these standards were challenged in court. Id. Just one was allowed to go into effect. Id. ; see also 79 Fed. Reg. 61,384, 61,419 (Oct. 10, 2014) (noting that “OSHA has not successfully adopted an emergency temporary standard for over thirty years”). In a more recent exercise of this power, which a court has not yet addressed, the Secretary issued an emergency regulation in June 2021, which imposed requirements on the healthcare industry to reduce transmission of COVID-19, mainly protective clothing and physical distancing. 86 Fed. Reg. 32,376 (June 21, 2021). The emergency rule did not require workers to get vaccinated or subject themselves to uncompensated weekly tests.

At issue is OSHA’s November 5 emergency standard, entitled ” COVID-19 Vaccination and Testing; Emergency Temporary Standard.” 86 Fed. Reg. 61,402 (Nov. 5, 2021). It applies to employers with 100 or more employees, what comes to roughly 80 million employees nationwide. Id. at 61,467. And it contains a narrow exemption for employees who “work[ ] remotely 100 percent of the time” or who “perform their work exclusively outdoors.” Id. at 61,419, 61,467.

The emergency rule also applies to the 26 States in the country that administer their own state OSHA Plans, which means that those States must enforce the vaccinate-or-test mandate against any covered public employees and private businesses in their jurisdiction. Id. at 61,462. Although Congress did not require state and local governments to adhere to the Act, see 29 U.S.C. §§ 652(5)654(a)(2), it used its spending power to encourage States to  accept federal funding—up to 50% of the total cost of each state plan—in return for adopting an OSHA-approved state plan, id. § 672(g). Under the Act, state plans must be at least as effective as the federal standards required by the Secretary. Id. § 667(b), (c)(2).

Under the emergency rule, the employer must verify “the vaccination status of each employee,” “maintain a record of each employee’s vaccination status,” and “preserve acceptable proof of vaccination.” 86 Fed. Reg. at 61,552. For employees who opt not to get vaccinated, the employer must require a test every seven days, one that neither the Federal Government nor the employers must pay for and one that the employees may not take without the supervision of an authorized person. Id. at 61,530, 61,532, 61,551, 61,553. Unvaccinated employees who do not comply must be “removed from the workplace.” Id. at 61,532. Unvaccinated employees must wear masks at work with few exceptions. Id. at 61,553. The testing and masking requirements do not apply to vaccinated employees. Id. Employers who violate the Act face penalties imposed by OSHA: up to $13,653 for each violation and up to $136,532 for each willful violation. 29 C.F.R. § 1903.15(d).

Several companies, organizations, individuals, and 27 States filed challenges to the emergency rule, raising a variety of claims in the various courts of appeals. On November 12, in one of those cases, the U.S. Court of Appeals for the Fifth Circuit stayed the vaccinate-or-test mandate. BST Holdings , 17 F.4th at 619. After our circuit was selected to handle the petitions for review on a consolidated basis, we received two sets of pertinent motions: a motion by the Secretary of Labor to vacate the Fifth Circuit’s stay order, see 28 U.S.C. § 2112(a)(4), and requests by various parties to grant initial hearing en banc.

II.

A few words are in order about the en banc motions in front of us—requests by roughly 59 parties that the full Court hear this case at the outset. At one level, granting the motion makes considerable sense. This is an extraordinary case, suitable for an extraordinary procedure. Given the unusual setting of these consolidated cases—a statutory delegation of authority over countless appeals to one regional court of appeals, 28 U.S.C. § 2112(a) —there is something to be said for putting all hands on deck, particularly when it comes to handling the stay motion, which could turn out to be the key decision point in all of these petitions for review. If the stay motion is the main event in a case about the legitimacy of a six-month emergency rule that ends on May 5, 2022, little opportunity for traditional en banc review will exist at the back end of the case. All of this explains why we favor granting the motion. But at another level, it makes little difference that our Court has divided 8-8 on whether to grant the en banc motion. We likely will not be the final decisionmakers in this case, given the prospect of review by the U.S. Supreme Court. And the existence of the en banc motion gives the judges of our Court the option to offer their perspectives on the stay motion, in opinions concurring in the denial of initial hearing en banc or dissenting from it.

III.

In evaluating a stay motion, we ask four questions: Which side is likely to prevail on the merits? What are the costs to the challengers of allowing the emergency rule to go into effect? What are the costs to the Secretary of Labor and others of barring the emergency rule from going into effect? What does the public interest favor? Nken , 556 U.S. at 434129 S.Ct. 1749. In this  instance, as in many others, we focus primarily on the likelihood-of-success inquiry. See Ala. Ass’n of Realtors , 141 S. Ct. at 2490.

IV.

The challengers should prevail for two main reasons. A clear-statement rule applies to this wide-ranging and unprecedented assertion of administrative power, and the Secretary of Labor has failed to show that Congress clearly delegated this authority to him.

A.

Today’s emergency rule is not an everyday exercise of federal power. The Secretary claims authority to require 80 million Americans—in virtually every type of American business there is—to obtain a COVID-19 vaccine or, in the alternative, to undertake a weekly COVID-19 test and wear a mask throughout each workday. Because the Federal Government pays for the vaccine but not the weekly test, it is fair to say that the Secretary is prioritizing the vaccine mandate over the test-and-mask mandate, if not coercing vaccinations. See 86 Fed. Reg. at 61,434 (acknowledging that the emergency rule “is designed to strongly encourage vaccination”). Further pressure on employees comes from other features of the rule: (1) Employers must provide time off for employees to get vaccinated and to recover from any side effects, id. at 61,457, while the rule does not require them to do so for employees who must undergo weekly tests, even if that requires considerable travel in rural areas, see id. at 61,484 ; (2) the agency normally requires employers to compensate employees for occupational safety gear and required testing but not in this instance, compare 29 C.F.R. § 1910.132(h)with 86 Fed. Reg. at 61,407 & n.2; and (3) employers can escape many of the administrative burdens of administering the rule if they require their employees to get vaccinated, 86 Fed. Reg. at 61,437. Either way, whether treated as a vaccine mandate or a vaccinate-or-test mandate, the Secretary must answer mandates of his own if he wishes to regulate large swaths of Americans with respect to substantial public policy, medical, and economic matters customarily regulated by the States.

In the first place, the federal courts “expect Congress to speak clearly when authorizing an agency to exercise powers” over large numbers of Americans with respect to contested public policy choices of vast significance. Ala. Ass’n of Realtors , 141 S. Ct. at 2489. Skeptical of mismatches between invocations of power by agencies and the statutes that purport to delegate that power, the federal courts require broad assertions of policymaking authority to be premised on direct and specific congressional delegations of that power. Congress must “speak clearly if it wishes to assign to an agency decisions of vast economic and political significance.” Util. Air Regul. Grp. v. EPA , 573 U.S. 302, 324, 134 S.Ct. 2427, 189 L.Ed.2d 372 (2014) (quotation omitted); see Whitman v. Am. Trucking Ass’ns , 531 U.S. 457, 468121 S.Ct. 903149 L.Ed.2d 1 (2001). What Justice Stevens said in 1980 in rejecting the “Benzene rule,” designed by OSHA to protect American workers from cancer, applies with equal force to today’s rule: “In the absence of a clear mandate in the Act, it is unreasonable to assume that Congress intended to give the Secretary the unprecedented power over American industry that would result from the Government’s view” of the statute. Indus. Union Dep’t, AFL-CIO v. Am. Petroleum Inst. , 448 U.S. 607, 645100 S.Ct. 284465 L.Ed.2d 1010 (1980) (plurality opinion). Notably, OSHA initially attempted to issue the Benzene Rule as an emergency rule, but it abandoned that approach  in favor of notice-and-comment rulemaking after the Fifth Circuit stayed the rule. Id. at 623, 100 S.Ct. 2844.

government intrusion on matters of individual liberty and because, in OSHA’s case, the required medical procedures do not comfortably map onto workplace-specific protective remedies. See Cong. Rsch. Serv., Mandatory Vaccinations : Precedent and Current Laws 9 (May 21, 2014); see also 86 Fed. Reg. at 61,436. If OSHA “claims to discover in a long-extant statute an unheralded power to regulate a significant portion of the American economy,” it should not be surprised if courts “greet its announcement with a measure of skepticism.” Util. Air Regul. Grp. , 573 U.S. at 324, 134 S.Ct. 2427 (quotation omitted). As with the eviction moratorium created by the federal Center for Disease Control and invalidated by the Supreme Court, today’s “claim of expansive authority” under this provision “is unprecedented.” Ala. Ass’n of Realtors , 141 S. Ct. at 2489 ; see Tiger Lily, LLC v. U.S. Dep’t of Hous. & Urb. Dev. , 5 F.4th 666, 668 (6th Cir. 2021). If federal courts have been skeptical when a medically based agency (the CDC) issues broad mandates with respect to housing, they should be equally skeptical when a workplace agency (OSHA) issues broad mandates with respect to medical procedures.

In the second place, the States, not the Federal Government, are the traditional source of authority over safety, health, and public welfare. In the context of a vast attempt to assume these police powers by the Federal Government, Congress must speak unequivocally. Whether it is seizing authority to regulate “the landlord-tenant relationship,” Ala. Ass’n of Realtors , 141 S. Ct. at 2489, to regulate private property, U.S. Forest Serv. v. Cowpasture River Pres. Ass’n , ––– U.S. ––––, 140 S. Ct. 1837, 1849–50207 L.Ed.2d 186 (2020), to enact run-of-the-mine criminal laws, Jones v. United States , 529 U.S. 848, 858120 S.Ct. 1904146 L.Ed.2d 902 (2000), to enact out-of-the-ordinary criminal laws, Bond v. United States , 572 U.S. 844, 848134 S.Ct. 2077189 L.Ed.2d 1 (2014), or to regulate the retirement age of state court judges, Gregory v. Ashcroft , 501 U.S. 452, 460–61111 S.Ct. 2395115 L.Ed.2d 410 (1991), Congress must “enact exceedingly clear language if it wishes to significantly alter the balance between federal and state power,” Ala. Ass’n of Realtors , 141 S. Ct. at 2489 (quotation omitted).

In applying this federalism clear-statement canon, it’s worth remembering that the only Supreme Court cases that permitted a government to impose a vaccination mandate on individuals arose from the States, not the National Government. Jacobson v. Massachusetts , 197 U.S. 1125 S.Ct. 35849 L.Ed. 643 (1905) ; Zucht v. King , 260 U.S. 17443 S.Ct. 2467 L.Ed. 194 (1922). In upholding a vaccination requirement against a substantive due process challenge, the U.S. Supreme Court reasoned that “[t]he safety and the health of the people of [a state] are, in the first instance, for that [state] to guard and protect” and “are matters that do not ordinarily concern the national government.” Jacobson , 197 U.S. at 3825 S.Ct. 358. It’s worth remembering that the power of a federal agency to regulate is the power to preempt—to nullify the sovereign power of the States in the area—which explains why 27 States oppose the emergency rule. And it’s worth remembering that, if one casually accepts congressional authority to regulate in this area, that recalibration of power comes with easy-to-overlook risks. It would mean that another administration could destroy the trial-and-error benefits of federalism in a different direction, say  by adopting a federal law that banned state and local governments from issuing all kinds of health-protective orders: stay-at-home orders, mask mandates, vaccine mandates, and many other measures besides. The power to give with preemptive national regulation includes the power to take away.

B.

In passing the Occupational Safety and Health Act, Congress did not clearly give the Secretary authority to require workers to undertake a medical procedure like a vaccine or a medical test, whether under his general authority to regulate “employees” in the workplace or under his specific authority to issue “emergency temporary standards.”

1.

The Occupational Safety and Health Act covers only workplace-specific hazards and permits only workplace-specific safety measures . As a threshold matter, the Act is designed to protect “employees” from dangers that arise directly out of the workplace and addresses only workplace conditions, as the title of the Act suggests (the “Occupational Safety and Health Act”) and as the rest of the Act confirms. The language of the Act covers dangers arising out of work, say a chemical used to make a plastic product or the heat generated at a steel foundry, not any risk facing the country and every citizen in it. Any other approach would facilitate a breathtaking expansion of the Secretary of Labor’s power. Whatever the health and safety challenges of today (air pollution, violent crime, obesity, a virulent flu, all manner of communicable diseases) or tomorrow (the impact of using the internet on mental health), the Secretary does not have emergency authority to regulate them all simply because most Americans who face such endemic risks also have jobs and simply because they face those same risks on the clock. By going to work each day, American workers do not transform these other risks into “hazards” or “grave dangers” to which “employees are exposed.” The Secretary’s authority to regulate workplace safety is simply too “indirect[ ]” to cover this nearly horizonless assertion of power. Ala. Ass’n of Realtors , 141 S. Ct. at 2488.

A comparison between the Secretary’s emergency proposal (a vaccinate-or-test mandate) and the kinds of requirements he has previously imposed on various industries during the pandemic (a mask mandate) illustrates the problem. Accept for now that, under some circumstances and in some places, the Secretary could impose a mask mandate. That would be a workplace requirement at least. It is one thing for the Secretary to require masks to minimize dangers to which “employees are exposed” during the workday and at the workplace. It is quite another to make an across-the-board judgment that the employee is “strongly encouraged”—emphasis on strongly—to undertake a medical procedure (a vaccination ) that cannot be undone at the end of the workday.

Whether it’s the Act as a whole or the narrow exception for emergency rulemaking, they both apply, in the words of the D.C. Circuit, only to dangers arising out of “work or work-related activities,” Oil, Chem. & Atomic Workers Int’l Union v. Am. Cyanamid Co. , 741 F.2d 444, 449 (D.C. Cir. 1984), not all hazards working people may face in their daily lives. That explains why the D.C. Circuit found another medical procedure—the sterilization of women who otherwise would encounter chemicals at work dangerous to the unborn—to be beyond the Act’s scope. Id. ; see also Steel Joist Inst. v. Occupational Safety & Health Admin. , 287 F.3d 1165, 1167 (D.C. Cir. 2002) (noting that “the Act  authorizes OSHA to regulate only the employer’s conduct at the worksite”). “[F]or coverage under the Act to be properly extended to a particular area,” seconds the Eleventh Circuit, “the conditions to be regulated must fairly be considered working conditions, the safety and health hazards to be remedied occupational , and the injuries to be avoided work-related .” Frank Diehl Farms v. Sec’y of Lab. , 696 F.2d 1325, 1332 (11th Cir. 1983).

Other provisions of the Occupational Safety and Health Act reinforce the message. The Act, it is true, refers to “hazards,” “substances,” and “agents,” terms that read in isolation might suggest that the Secretary could regulate any hazardous substance or agent. But context illuminates meaning. Throughout the Act, it speaks to hazards facing employees in work-specific contexts and to occupational risks faced due to work:

• The Act’s preamble says it is designed “to assure … safe and healthful working conditions,” 29 U.S.C. § 651(b), and to avoid “personal injuries and illnesses arising out of work situations,” id. § 651(a).

• A provision says that the Act applies “to employment performed in a workplace” and “to working conditions of employees.” Id. § 653(a), (b).

• A provision tells the Secretary to make rules “for developing information regarding the causes and prevention of occupational accidents and illnesses,” id. § 657(c)(1), or “work-related deaths, injuries and illnesses,” id. § 657(c)(2).

The agency’s regulations reflect this understanding too. In general, OSHA requires employers to compensate employees for protective gear and tests needed for work safety. 29 C.F.R. § 1910.132(h). An exception exists for costs that are not specific to the workplace, say sunscreen or steel-reinforced boots. Id. § 1910.132(h)(2), (4)(iii). In this instance, the Secretary’s decision not to require employers to pay for employees’ weekly COVID-19 tests depletes his claim that this emergency rule arises from a work-focused, as opposed to society-focused, imperative. See 86 Fed. Reg. at 61,437. The Secretary conceded that, while OSHA usually requires employers to bear such costs “in order to remove barriers to employee participation,” the agency has not done so here in order to “strongly encourage” vaccination. Id. at 61,407.

OSHA also requires employers to give their employees and the agency access to “relevant exposure and medical records” to identify, handle, and prevent “occupational disease.” 29 C.F.R. § 1910.1020(a). The agency requires employers to keep records that “monitor[ ] the amount of a toxic substance or harmful physical agent to which the employee is or has been exposed.” Id. § 1910.1020(e)(2)(i)(A)(1). But these exposure risks do not cover “situations where the employer can demonstrate that the toxic substance or harmful physical agent is not used, handled, stored, generated, or present in the workplace in any manner different from typical non-occupational situations.” Id. § 1910.1020(c)(8). As still another example, the agency has rules about occupational noise exposure, which require employers with affected employees to administer a testing program that determines the employee’s hearing loss. Id. § 1910.95(g). If the hearing loss is determined not to be “work related,” however, the employer does not have to provide assistance. Id. § 1910.95(g)(8). With respect to the recordkeeping requirements, moreover, an employer “must consider an injury or illness to be work-related if an event or exposure in the work environment either caused or contributed to the resulting condition or significantly aggravated a  pre-existing injury or illness.” Id. § 1904.5(a). In OSHA’s rules concerning air contaminants, the rules center on the amount of an employee’s exposure to a substance “during an 8-hour shift.” Id. § 1910.1000(a)–(c).

The agency in the past has understood its authority in this work-anchored way. An examination of the nine “emergency temporary standards” promulgated before 2021, even the five of six that were successfully challenged, reveals only regulations addressing exposures solely because of, not in spite of or in addition to, the workplace. See 36 Fed. Reg. 23,207 (Dec. 7, 1971) (workplace protection from asbestos); 38 Fed. Reg. 17,214 (June 29, 1973) (workplace protection from pesticides); 38 Fed. Reg. 10,929 (May 3, 1973) (workplace protection from carcinogenic substances in “area[s] to which access is restricted and controlled by the employer”); 39 Fed. Reg. 12,342 (Apr. 5, 1974) (workplace protection from vinyl chloride); 41 Fed. Reg. 24,272 (June 15, 1976) (workplace protections for diving operations, while noting that “diving by persons engaged in recreational or sport diving or other diving not in an employment context are beyond the jurisdiction of the Act”); 42 Fed. Reg. 22,516 (May 3, 1977) (workplace protections from benzene); 42 Fed. Reg. 45,536 (Sept. 9, 1977) (workplace protection from manufacturing pesticides); 43 Fed. Reg. 2586 (Jan. 17, 1978) (workplace protection from acrylonitrile); 48 Fed. Reg. 51,086 (Nov. 4, 1983) (workplace protection from asbestos).

All in all, the Secretary might have authority to impose mandates of some sort on doctors and nurses who treat COVID-19 patients or researchers who work with the underlying virus given the workplace “exposure” risks caused by that work. And it might give the Secretary authority to impose workday masking requirements in other settings vulnerable to COVID-19 exposures. But the emergency rule extends well beyond such workplace-specific hazards and workplace-specific remedies.

2.

Not only is it doubtful that Congress gave the Secretary of Labor clear authority to impose this vaccinate-or-test mandate through the general provisions of the Act, but Congress also failed to do so clearly under the provision for “emergency temporary standards.” In relevant part, the provision for emergency rules says:

The Secretary shall provide, without regard to the requirements of chapter 5 of Title 5, for an emergency temporary standard to take immediate effect upon publication in the Federal Register if he determines (A) that employees are exposed to grave danger from exposure to substances or agents determined to be toxic or physically harmful or from new hazards, and (B) that such emergency standard is necessary to protect employees from such danger.

29 U.S.C. § 655(c)(1).

The statute applies only to “necessary” provisions that address “grave” workplace dangers . The term “necessary” has one of two meanings, either “useful” or “indispensable”/”essential.” Black’s Law Dictionary 928 (5th ed. 1979); American Heritage Dictionary of the English Language 877 (1975). Picking between the options might be difficult if the word appeared alone. But it does not. It appears in the context of a provision dealing with an “emergency” and “grave” danger. Understanding words, like filling in crossword puzzles, works best by attending to context—what is nearby, what is known. Gutierrez v. Ada , 528 U.S. 250, 255120 S.Ct. 740145 L.Ed.2d 747 (2000). Once connected, the reference to “necessary” powers to address “grave” dangers in an “emergency”  clarifies that “necessary” has the narrower meaning. It refers only to indispensable or essential measures, not to whatever the Secretary determines is useful or beneficial.

A comparison to the Secretary’s authority to impose permanent standards confirms this reading. When he puts a rule through notice and comment, the standard need not be “necessary to protect employees,” 29 U.S.C. § 655(c)(1), only “reasonably necessary or appropriate to provide safe or healthful employment,” id. § 652(8); see id. § 655(b). An emergency measure thus must be more than just appropriate; it must be indispensable or essential.

Turn to “grave” dangers, which refer to “serious” workplace dangers. Webster’s Ninth New Collegiate Dictionary 534 (1984). Taken by itself, there is room for debate about the meaning of a serious workplace danger, particularly one that the statute allows the Secretary to “determine[ ]” himself. But the record in this case and the Secretary’s position in describing his rulemaking narrow the range of debate. Whatever a grave or serious workplace danger might mean in the abstract, the Secretary concedes that vaccinated individuals who get the virus do not face that risk, even though they can contract it while going to work with unvaccinated individuals. 86 Fed. Reg. at 61,434. Else, the Secretary would require vaccinated Americans to work at home or stay home altogether.

This interpretation of the statute and the Secretary’s concession make it exceedingly difficult to maintain under any standard of review that the emergency mandate is necessary or indispensable to address a grave danger. One problem arises from a core tenet of administrative law. The Secretary never considered this meaning of the statute—that it requires indispensable or essential measures, not simply useful or beneficial ones—in proposing the emergency rule. It is a staple of administrative law that federal courts may not uphold a rule on a ground never addressed by the agency. SEC v. Chenery Corp. , 318 U.S. 80, 8763 S.Ct. 45487 L.Ed. 626 (1943). The Secretary to date has explained only why he thinks the vaccinate-or-test mandate is beneficial to protect workers and society as a whole. He has not explained why it is the indispensable or essential way to protect workers. We have no authority to uphold a rule as “necessary” when the Secretary has not made that finding himself under the correct interpretation of the law. See 5 U.S.C. § 706(2).

The other problem is that the Secretary cannot satisfy this interpretation of the statute. Consider the many less intrusive, more tailored protective measures that address grave dangers on the Secretary’s own terms. Just as the Secretary targeted the healthcare industry in June 2021 with mask and other protective-gear requirements, he could do the same for industries that face high spreading risks. The record does not show that full vaccination or weekly testing is necessary on top of a tailored mask mandate. The Secretary could focus any requirements on the workers most at risk—those over 65, those with pre-existing conditions most vulnerable to the virus, those who have not already gotten the virus. The Secretary could create exemptions for those least at risk, say cohorts from age 18 to 49, a population range that faces healthcare risks from COVID-19 at roughly the same level as the Secretary’s own assessment of what is not a grave risk, with some slightly above and some slightly below. See 86 Fed. Reg. at 61,434 ; Ctr. for Disease Control, Rates of COVID-19 Cases and Deaths by Vaccination Status, https://covid.cdc.gov/covid-data-tracker/#rates-by-vaccine-status. Or  the Secretary could impose requirements that account for the many environments in which Americans work. Consider the range of possibilities—from the two-person janitorial staff working the night shift, to the consultant who comes into the office a few times a week, to the company that already requires masks (but not weekly tests) and requires significant separation of workers protected by up-to-date ventilation systems, to the firm that rotates workers between telework and in-person to minimize contact. But that is not what the rule does. “Applying to 2 out of 3 private-sector employees in America, in workplaces as diverse as the country itself, the Mandate fails to consider what is perhaps the most salient fact of all: the ongoing threat of COVID-19 is more dangerous to some employees than to other employees.” BST Holdings , 17 F.4th at 615.

In the face of the many less intrusive options available to the Secretary, the idea that a national vaccinate-or-test mandate for 80 million workers is necessary is hard to maintain. And that is true under any standard of review: fresh review of the language of the statute, substantial evidence review, arbitrary or capricious review, or the “harder look” review due emergency rules. Asbestos Info. Ass’n v. Occupational Safety & Health Admin. , 727 F.2d 415, 421 (5th Cir. 1984).

The statute covers only an “emergency” and only “temporary” requirements . In construing statutes, courts frequently look to the context in which they arise—here authority to set “emergency temporary standards” that sidestep the notice-and-comment process. See, e.g. , Bond , 572 U.S. at 861–63134 S.Ct. 2077 ; Johnson v. United States , 559 U.S. 133, 139–40130 S.Ct. 1265176 L.Ed.2d 1 (2010). Whether one looks to the Secretary’s strongly encouraged preference (vaccinate) or discouraged alternative (test and wear a mask), it is difficult to understand how on November 5, 2021, an “emergency” suddenly took hold requiring the imposition of a vaccinate-or-test mandate by January 4, 2022. Start with the mask requirement. As the Secretary well knows, masks are not a new idea. They have been a protective tool from the outset. Given the wide availability of this option since the beginning, the view that this requirement counts as an “emergency” measure, all at a time when fewer people face lethal risks from COVID-19, sucks the concept dry of meaning.

Vaccines are newer, to be sure. But they hardly are a revelation. They have been readily available since last spring, and they alleviate the health risks from the pandemic rather than make them worse. Why now? Why above all immediately impose such a controversial mandate on 80 million workers without undergoing the give and take that comes with the notice-and-comment process—and that usually leads to better rulemaking and always leads to more transparency about the costs and benefits of any new rule for workers and companies. See Azar v. Allina Health Servs. , ––– U.S. ––––, 139 S. Ct. 1804, 1816204 L.Ed.2d 139 (2019). The “more expansive” a rule’s reach, “the greater the necessity for public comment.” Am. Fed’n of Gov’t Emps. v. Block , 655 F.2d 1153, 1156 (D.C. Cir. 1981).

How, moreover, is a vaccine “temporary”? That approach conveys considerable insensitivity to those who, for reasons of their own, are reluctant to roll up their sleeves. By any measure, a vaccine injection is not temporary.

Making the invocation of this emergency temporary power odder still is the nature of the risks presented by COVID-19 today. It is not working men and women in the main who face the most serious risks. It is older men and women, most of whom are retired and who no longer are subject to  the Secretary’s oversight. The key risks to individuals who do work and who remain unvaccinated are to them, not to their vaccinated colleagues. Sure, there have been, and likely will continue to be, breakthrough cases that infect vaccinated individuals, some no doubt facilitated by unvaccinated individuals. But the Secretary agrees that this risk is not serious. During the rulemaking process, he acknowledged that the risk to vaccinated employees of continuing to work with unvaccinated employees is “not” a “grave danger.” 86 Fed. Reg. at 61,434.

That leaves the Secretary with the burden of answering this question: Is it really an emergency to protect retired individuals from a workplace they no longer visit, to protect vaccinated working people from a risk the Secretary does not consider grave, and to protect unvaccinated working people from themselves based on highly personal medical decisions? That is a heavy lift, one that is highly unlikely to withstand any standard of review.

Equally unavailing is the Secretary’s other explanation for the emergency rule. Education, public-health advocacy, and easy-to-obtain free vaccinations, he points out, have not worked as well as or as quickly as the Federal Government hoped—because just 70% or so of Americans have received one shot and just 60% or so of Americans are fully vaccinated. Ctr. for Disease Control, COVID-19 Vaccinations in the United States, https://covid.cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total; see 86 Fed. Reg. at 61,431 –32. The Secretary projects that the “strongly encourage[d]” vaccination option would lead an additional 22.7 million workers to get vaccinated, increasing the vaccination rate in the covered workforce from 62% to 89%. 86 Fed. Reg. at 61,433, 61,472. These estimates as an initial matter lift the veil on the Secretary’s understanding of the rule, revealing that he thinks it will operate much more like a vaccine mandate than a vaccine option. Another problem lurks as well. In the context of new viruses, new variants, and other challenges presented by communicable diseases, there will always be a spectrum of medical developments and innovations, whether it is new types of vaccinations, booster shots, medical treatments, or something else. That ongoing reality does not give one national agency the option of labeling something an “emergency” in perpetuity, immediately imposing a one-size-fits-all-companies solution on the country, preempting all contrary approaches to the matter in our States and cities, and circumventing the notice-and-comment process. “In case of emergency break glass” this is not—unless we wish to sideline the notice-and-comment process and the trial-and-error benefits of American federalism with respect to every future medical innovation concerning COVID-19 for this federal agency and other ones too.

One last point on this score. The statute gives the Secretary authority to issue an emergency rule only for six months. 29 U.S.C. § 655(c)(2)–(3). It does not mention any authority to extend the rule for another six months. To our knowledge, the Secretary has never used this narrow authority to extend an emergency rule for another six months. All of this prompts a question: Does the Secretary expect to finish the notice-and-comment process with respect to this uniquely important and uniquely wide-ranging rule by May 5, 2022, when the emergency rule dissolves? That seems improbable. As our circuit has come to appreciate, this rule affects a lot of industries and a lot of people. Consistent with that reality, the Secretary has already granted one 45-day extension of time, extending the end of the public comment period from December 6, 2021, to  mid-January 2022. The six-month nature of the Secretary’s emergency-rule authority highlights the unusual nature of its exercise today.

In view of this conclusion, we need not address several serious constitutional claims raised by the challengers. Among others, there are at least these three that would need to be addressed before the emergency rule could be enforced. One, does this regulation of non-commercial inactivity—a requirement that the unvaccinated get shots or weekly tests—exceed Congress’s Commerce Clause power? See Nat’l Fed. of Ind. Bus. v. Sebelius , 567 U.S. 519, 550–52132 S.Ct. 2566183 L.Ed.2d 450 (2012) ; infra at 285-86 (Bush, J., dissenting); BST Holdings , 17 F.4th at 619 (Duncan, J., concurring). Two, if we accepted the Secretary’s sweeping reading of the Act—permitting him to regulate any substance, whether unique to work or not, so long as the Secretary finds it dangerous—would that amount to an unconstitutional delegation of power? See Gundy v. United States , ––– U.S. ––––, 139 S. Ct. 2116, 2123204 L.Ed.2d 522 (2019) ; id. at 2135–37 (Gorsuch, J., dissenting). Compare Indus. Union Dep’t , 448 U.S. at 645100 S.Ct. 2844 (plurality opinion) (avoiding this constitutional question by construing the statute to narrow OSHA’s authority), with id. at 687, 100 S.Ct. 2844 (Rehnquist, J., concurring in the judgment) (finding an unconstitutional delegation because “[i]t is difficult to imagine a more obvious example of Congress simply avoiding a choice which was both fundamental for purposes of the statute and yet politically so divisive that the necessary decision or compromise was difficult, if not impossible”). Three, does compelling faith-sensitive employers to administer these mandates violate the Free Exercise Clause or the Religious Freedom Restoration Act by interfering with their employment decisions or religious mission? See Burwell v. Hobby Lobby Stores, Inc. , 573 U.S. 682, 719–20134 S.Ct. 2751189 L.Ed.2d 675 (2014) ; Our Lady of Guadalupe Sch. v. Morrissey-Berru , ––– U.S. ––––, 140 S. Ct. 2049, 2060–61207 L.Ed.2d 870 (2020). Because our interpretation of the relevant statutes avoids these constitutional claims and any others, we need not address them. See United States v. Erpenbeck , 682 F.3d 472, 476 (6th Cir. 2012). By contrast, anyone who takes the view that the Fifth Circuit’s stay should be lifted must come to grips with each of the statutory imperatives, each of the clear statement requirements, and all of the constitutional claims.

C.

The Secretary insists that any ambiguity in the statute favors him, not the challengers. He claims that uncertainty about the meaning of the statute allows him to construe the statute to exercise more power, not less. Resp. Mot. to Dissolve Stay at 17; Chevron, U.S.A., Inc. v. Nat. Res. Def. Council , 467 U.S. 837, 843–44104 S.Ct. 277881 L.Ed.2d 694 (1984). But ambiguity for Chevron purposes comes at the end of the interpretation process, not at the beginning. Id. at 843 n.9, 104 S.Ct. 2778. The clear-statement canons eliminate any power-enhancing uncertainty in the meaning of the statute. With “significant constitutional and federalism questions raised” and a federalism-protecting interpretation of the statute not clearly ruled out, we must accept that interpretation and “reject the request for administrative deference.” Solid Waste Agency of N. Cook Cnty. v. U.S. Army Corps of Eng’rs , 531 U.S. 159, 174121 S.Ct. 675148 L.Ed.2d 576 (2001).

A contrary approach leads to a characterization of administrative law under which significant decisions of the U.S. Supreme Court were one emergency regulation, no notice, no comment, away from  oblivion, indeed from effectively being overruled. If the Secretary is right, the federal office of civil rights suddenly could have construed the ambiguity in the ADEA to cover state court judges. Cf. Gregory , 501 U.S. at 460–61111 S.Ct. 2395. If true, the Department of the Interior suddenly could have construed the ambiguity in the Mineral Leasing Act and National Trails System Act to regulate all manner of private property. Cf. Cowpasture River Pres. Ass’n , 140 S. Ct. at 1848–50. If true, the SEC suddenly could have construed the Securities and Exchange Act to apply outside the United States. Cf. Morrison v. Nat’l Austl. Bank Ltd. , 561 U.S. 247, 272–73130 S.Ct. 2869177 L.Ed.2d 535 (2010). And so on. Chevron has no role to play in this case.

The Secretary counters that he is entitled to issue an emergency rule given new knowledge about the dangers of COVID-19 and the increased risk of infection and transmission due to the Delta variant. But the Delta variant has dominated our country’s COVID-19 statistics since June. See 86 Fed. Reg. at 61,408 –09. Even then, the Secretary found that vaccinated workers do not face a “grave danger” from COVID-19, with or without the existence of Delta. Id. at 61,434.

The Secretary emphasizes that he is regulating the workplace because the virus creates risks for working men and women. But authority to regulate the workplace with protective gear designed to handle on-the-job exposures to substances and tailored to the circumstances of that job is one thing; authority to require medical procedures or tests for two-thirds of American workers, no matter their work circumstances or individual risks, is quite another. This is precisely the kind of broad assertion of administrative power that should be accompanied by clear, direct, and channeled delegations by Congress. It is hard to think of a better example of the need for a clear statement of congressional authority than this one.

The Secretary and some of his supporters claim that regulating infectious diseases through vaccines is not as unusual as the challengers maintain, pointing to a bloodborne pathogen regulation from 1991. See 29 C.F.R. § 1910.1030. But that regulation shows what works and what does not. The 1991 regulation required employers to make the hepatitis B vaccine “available” to employees “who have occupational exposure” to bloodborne pathogens at no cost to the employee and at a reasonable time and place. Id. § 1910.1030(f)(1)(i)–(ii). Consider all of the differences between that regulation and this one. It narrowly targeted “health care workers” for protection “from viruses, particularly those causing Hepatitis B and AIDS, that can be transmitted in the blood of patients.” Am. Dental Ass’n v. Martin , 984 F.2d 823, 824 (7th Cir. 1993). It did not regulate all American businesses, no matter the nature of the industry, product, or service, so long as 100 employees or more work there. It was “[p]romulgated after a protracted notice-and-comment rulemaking proceeding.” Id. It did not sidestep that process. And it appreciated the personal nature of the decision whether to get a vaccine—that a truly voluntary program, in OSHA’s words, would “foster greater employee cooperation and trust in the system.” 56 Fed. Reg. 64,004, 64,155 (Dec. 6, 1991). It did not pressure or coerce unvaccinated employees by imposing significant costs and burdens on them alone. Instead of helping the Secretary’s cause, a comparison between the 1991 rule and the 2021 rule undermines it.

The Secretary relatedly points to a different part of the statute to suggest that Congress contemplated immunization when delegating its authority. In a section  on “Research and Related Activities,” Congress gives the Secretary of Health and Human Services authority to establish programs to examine and test the workplace to “determin[e] the incidence of occupational illnesses.” 29 U.S.C. § 669(a)(5). The authorization comes with this caveat: “Nothing in this or any other provision of this chapter shall be deemed to authorize or require medical examination, immunization, or treatment for those who object thereto on religious grounds, except where such is necessary for the protection of the health or safety of others.” Id. This argument tries to squeeze a lot of power out of a very small statutory tube. It involves a single reference to immunizations, a reference that explains when they are prohibited. It comes from a different part of the statute and concerns the Secretary of Health and Human Services, not OSHA and not the Secretary of Labor. If this is a “clear statement” of congressional authority that OSHA may impose a vaccinate-or-test mandate on the American workforce, we should call it a “nearly silent” rule, not a “clear statement” rule.

What of the Secretary’s claim that he should not be second-guessed for applying the emergency rule just to companies with 100 employees or more? The problem is not second-guessing; it is matching the Secretary’s explanations for this emergency rule with its scope. If the explanation for announcing an emergency rule is the “grave dangers” that American workers face on the job from getting the virus, that risk applies to all companies in which employees work together inside. Nor does it answer the point to say, as the Secretary does, that he was concerned about imposing administrative burdens on smaller companies. Think of how that argument would fare in another context. If the Secretary suddenly realized that exposure to a new chemical created a “grave” danger of cancer, it is difficult to imagine that anyone would permit an emergency rule targeting the problem to apply only to companies with over 100 employees in order to save the other companies money.

What of a related reality—that federal agencies historically have been able to impose drug tests on workers? But, again, those regulations illustrate the permitted and forbidden sides of the line. The Department of Transportation, to be sure, may require employees in a few industries—airlines, railroads, motor carriers, public transit—to take periodic drug tests given the flat-line risks to the public of having impaired pilots, conductors, truckers, or bus drivers. See, e.g. , 49 U.S.C. §§ 45102(a)(1)20140(b)(1)(A)31306(b)(1)(A)5331(b)(1)(A). But that authority, specific to a few industries and clearly delegated by Congress, would not give the Department of Transportation power to require American workers to take a drug test to end the opioid crisis—even if such tests could save up to 100,000 lives a year.

This last question and answer largely take care of the next objection—that the emergency rule is needed to deal with certain types of private employers that have been devastated by virus break-outs. A good example, as the Secretary and many others point out, is the meatpacking industry, where many of the largest spreading events initially occurred. Two responses. As with the special risks facing the transportation industry, Congress and OSHA may wish to focus on special risks facing healthcare workers and the workers in other high-risk industries. But that is not what this rule does. The other response is to note that the industries most at risk happen to be the ones most proactive in addressing the risks of the pandemic. How could an emergency rule be necessary to protect meatpacking workers when, so far as the record shows, that  industry has obtained high vaccination rates on its own? See, e.g. , Am. Pub. Health Ass’n et al. Amicus Br. at 16 (noting that more than 96% of Tyson Food’s 120,000 U.S. workers are vaccinated); 86 Fed. Reg. at 61,435. Just as the Secretary must match his assertion of power with the statute, he must match his exercise of power with explanations in the record that fit the bill.

What of the collective-action problem at the root of this assertion of power? Doesn’t the agency have authority to deal with the external costs created by vaccination decisions—the cost to others created by individuals who choose not to get vaccinated and the cost to society of slowing down efforts to bring the virus to heel? See 86 Fed. Reg. at 61,539 (explaining that vaccination reduces the risks that workers “present to others given the reduced likelihood of transmission”); id. at 61,520 (noting that “[c]urrent efforts to increase the proportion of the U.S. population that is fully vaccinated against COVID-19 are critical to ending the COVID-19 pandemic”). But, as shown, the risk to vaccinated workers from unvaccinated workers is one that the Secretary agrees is not a grave danger. No less significantly, it’s doubtful this federal power sweeps this broadly given the vertical separation of powers embedded in our Constitution. There is a Commerce Clause, yes. It gives Congress broad powers, to be sure. And it helps the Federal Government to resolve some collective-action problems affecting interstate commerce, no doubt. But through it all, it remains a Commerce Clause, not a collective-action clause—and not a clause that grants the national government all of the police powers customarily associated with state governments in order to fix any new societal challenge.

That the Constitution permits the Federal Government to resolve some collective-action problems facing society but not all of them simply confirms that “there are two sides to today’s story.” Tiger Lily , 5 F.4th at 675 (Thapar, J., concurring). On one side, yes, the Federal Government has considerable authority to regulate and sometimes mandate what individuals may do. But the other side reveals many libertarian guarantees of the U.S. Constitution, each empowering individuals to resist national solutions to pedestrian and urgent policy problems alike. Before we rush to lament the reality that American individualism may present obstacles to quelling the pandemic as quickly as we would like, it’s worth keeping in mind that it is a national trait that has done the country some good from time to time. Perhaps indeed Americans’ non-conformist ways have had something to do with American businesses bringing vaccines to market more quickly than any vaccine in history and doing so more quickly than any other country, collectivist or not, has been able to do. See Drew Armstrong, The World’s Most Loathed Industry Gave Us a Vaccine in Record Time , Bloomberg Businessweek (Dec. 23, 2020), https://www.bloomberg.com/news/features/2020-12-23/covid-vaccine-how-big-pharma-saved-the-world-in-2020; Jared S. Hopkins, How Pfizer Delivered a Covid Vaccine in Record Time: Crazy Deadlines, a Pushy CEO , Wall St. J. (Dec. 11, 2020), https://www.wsj.com/articles/how-pfizer-delivered-a-covid-vaccine-in-record-time-crazy-deadlines-a-pushy-ceo-11607740483.

What of the concern that the federal courts should take a low-impact approach to the public policy exigencies created by a crisis like the pandemic? It is a fair question. But it submits to fair answers. One is that, in the absence of a notice-and-comment process, the federal courts are all that’s left. Who else, what else, is there to assess unfounded assertions of emergency powers by a federal agency that will have  irreversible consequences for American workers and companies? The other answer is that overlooking rule-of-law limitations on federal power usually increases—it does not ameliorate—the footprint of the federal courts. It is the rare accretion of power to the President, Congress, or a federal agency that does not eventually take the federal courts along for the ride.

V.

The other stay factors largely favor the challengers as well. Because OSHA’s authority extends only to regulating the workplace, the equities embedded in the stay factors do not extend to the costs to society of having unvaccinated Americans. They extend only to the risks to workers and companies.

From the perspective of the challengers, there are serious irreversible costs if the emergency rule is immediately allowed to go into effect. Start with employees. The vaccinate-or-test mandate has costs for them that cannot be undone. Whether it is an irreversible vaccination, uncompensated testing costs, or a lost job, the affected employees face considerable jeopardy if the federal courts mistakenly allow this rule to go into effect. The same is true of employers, whether one focuses on the estimated $3 billion in compliance costs or the difficulties small companies (with just over 100 workers) will face in competing with smaller companies who can attract workers disinterested in complying with the mandate. From the perspective of the Secretary of Labor and other parties that support the emergency rule, the main risk of staying the rule is to unvaccinated American workers. But as we near the two-year anniversary of the pandemic, it is hard to see why American workers are not allowed to assess the risk-benefit choice of this personal medical decision for themselves. Even if the mandate would have ancillary benefits for Americans who come into contact with unvaccinated workers outside the workday, that consideration is not OSHA’s to regulate. From the perspective of the public interest, it is both wise and beneficial to stick with historical norms—that the default rule in agency rulemaking should be the notice-and-comment process, particularly when a rule imposes highly consequential new regulations on American workers and companies and when the agency has never invoked such a power before. A “lack of historical precedent” tends to be the most “telling indication” that no authority exists. Free Enter. Fund v. Pub. Co. Acct. Oversight Bd. , 561 U.S. 477, 505130 S.Ct. 3138177 L.Ed.2d 706 (2010) (quotation omitted).

All of this undermines the Secretary’s view that we should lift the stay issued by the Fifth Circuit. But it leaves unmentioned one other part of the stay calculation—that the Secretary estimated during the rulemaking process that the emergency rule would save 6,500 lives—a point unmentioned until now because it is never easy for judges to deal with. In one sense, it is far better to have the President, Congress, an authorized federal agency, or the States making cost-benefit decisions when American lives are at stake. Who are we to say when an emergency rule should go into effect if the rule would save lives? The only thing that prevents such a job from being unbearable is to appreciate that not every such decision is for us to make.

In this instance, the first answer is that the Secretary has assumed a power he does not have. Even though the CDC’s eviction moratorium was defended on the same ground—that it would save thousands of lives—the Supreme Court refused to allow the agency to enforce it. Ala. Ass’n of Realtors , 141 S. Ct. at 2490. So also when States defended stay-at-home orders that restricted religious services on  the ground that they would save lives. These orders, too, were stayed. Roman Cath. Diocese of Brooklyn v. Cuomo , ––– U.S. ––––, 141 S. Ct. 63, 69208 L.Ed.2d 206 (2020) (per curiam). The second answer is that, once judges go beyond the modest task of determining whether statutes permit agency action, these broader considerations become exceedingly complicated—and well beyond our ken. Even the Secretary’s own actions illustrate this complexity, especially if saving lives is the only consideration. Look back on the many times when a vaccinate-or-test mandate was not pressed by the Secretary: not in June 2021, when he issued the protective-gear orders with respect to the healthcare industry; not in September 2021, when he initiated this rulemaking procedure; and not on November 5, 2021, when he announced this six-month rule and said it would not go into effect until January 4, 2022. Consider too the many Americans still unprotected by this emergency rule: workers in companies with fewer than 100 employees and all customers who visit any American retail store or business. But it would be no more fair to criticize the Secretary of Labor on this ground than it would be to register a similar criticism against the Fifth Circuit for staying the emergency rule. That takes us back to where we started: The Secretary’s emergency rule likely exceeds his authority.

The Court should grant the petition for initial hearing en banc and leave the Fifth Circuit’s stay of the emergency rule in place.

DISSENTING FROM THE DENIAL OF INITIAL HEARING EN BANC

BUSH, Circuit Judge, dissenting from the denial of initial hearing en banc.

This is a case about the Occupational Safety and Health Administration, but it is really a case about power. Specifically, it concerns the attempted exercise of a purported power—to impose a de facto national vaccine mandate upon some eighty-million Americans—that OSHA was never given and that Congress likely could never have given to it. Chief Judge Sutton’s dissent ably explains the former defect, and so I join it in full. I write separately to address the latter.

For a discussion of why I apply this label to OSHA’s standard, see infra pages 286-88.

Whether it uses a clear statement or not, Congress likely has no authority under the Commerce Clause to impose, much less to delegate the imposition of, a de facto national vaccine mandate upon the American public. Such claimed authority runs contrary to the text and structure of the Constitution and historical practice. The regulation of health and safety through compulsory vaccination is a traditional prerogative of the states—not the domain of Congress and certainly not fodder for the diktat of a federal administrative agency. Because we should have granted initial hearing en banc to vindicate the correct understanding of the Constitution and to cabin OSHA to its legitimate role, I respectfully dissent.

I.

This case has a veneer of complexity, so it is useful to start with some first principles of constitutional adjudication. It may seem paradoxical that some of the most effusive guarantees of liberty can be found in the bills of rights of some of the world’s most savage dictatorships. See Antonin Scalia, Foreword: The Importance of Structure in Constitutional Interpretation , 83 N.D. L. Rev. 1417, 1418 (2008). Why do we seem to respect our bill of rights, at least in the main, while other  attempts have faltered the world over? The answer is structure. Id. Our Framers understood that the true bulwark of liberty is not a “parchment guarantee[ ],” but the diffusion of power both horizontally and vertically. Id. (quoting The Federalist No. 51, at 323 (James Madison) (Clinton Rossiter ed., 1961)); see also United States v. Morrison , 529 U.S. 598, 616 n.7, 120 S.Ct. 1740146 L.Ed.2d 658 (2000) (“[T]he Framers crafted the federal system of Government so that the people’s rights would be secured by the division of power.”).

James Madison called our constitutional structure a “double security” for “the rights of the people.” The Federalist No. 51, supra , at 320 (James Madison). Power was first divided by the Constitution “between two distinct governments”—federal and state. Id. And that power was then “subdivided among distinct and separate departments”—legislative, executive, and judicial. Id. Thus, just as each government was “controlled by itself,” the federal and state governments “would control each other.” Id. The “extensive portion of active sovereignty” the Constitution left to the states would prevent our institutions from degenerating “into one consolidated government” and would thereby check the resulting infringement on the people’s liberty. The Federalist No. 45, supra , at 286–87 (James Madison); see also New York v. United States , 505 U.S. 144, 181112 S.Ct. 2408120 L.Ed.2d 120 (1992) (“State sovereignty is not just an end in itself: Rather, federalism secures to citizens the liberties that derive from the diffusion of sovereign power.” (cleaned up)).

The constitutional text bears out that original design. Congress inherited from the Constitutional Convention no roving warrant to legislate on whatever matter it sees fit. Indeed, the Framers directly rejected such sweeping authority. See 2 The Records of the Federal Convention of 1787, at 21–27 (Max Farrand, ed. 1911). That was not because the idea lacked a proponent—Gouverneur Morris took “the controversial position that the federal government should possess the police power.” William Michael Treanor, The Case of the Dishonest Scrivener: Gouverneur Morris and the Creation of the Federalist Constitution , 120 Mich. L. Rev. 1, 28 (2021). But he was alone in that view. “No one else at the … Convention argued that the national government should have the ‘police’ power.” Id. at 29. Rather, Morris’s fellow delegates spoke of it only “as a power of the states.” Id. And so our limited Constitution emerged, carefully enumerating and thus carefully cabining each federal branch’s respective powers. See Nat’l Fed. of Indep. Bus. v. Sebelius , 567 U.S. 519, 534132 S.Ct. 2566183 L.Ed.2d 450 (2012) (“The enumeration of powers is also a limitation of powers[.]”). As a result, when Congress wishes to legislate, it must show “that a constitutional grant of power authorizes each of its actions.” Id. at 535, 132 S.Ct. 2566.

But “[t]he same does not apply to the States, because the Constitution is not the source of their power.” Id. States instead enjoy a residual authority to regulate within their borders—a power that pre-dates the Constitution and does not derive from it. Id. at 535–36, 132 S.Ct. 2566. The Tenth Amendment memorializes that point, clarifying that those “powers not delegated to the United States by the Constitution” are “reserved to the States respectively, or to the people.” U.S. Const. amend. X (emphasis added); see also Joseph Story, Commentaries on the Constitution of the United States 711–12 (Ronald D. Rotunda & John E. Nowak, eds. 1987). The states under our federated system thus enjoy a “general power of governing”—what the Supreme Court has repeatedly termed  their “police power.” Nat’l Fed. of Indep. Bus. , 567 U.S. at 536132 S.Ct. 2566.

Part and parcel of that traditional police power—and thus an authority “reserved to the States”—is the power to regulate public health. U.S. Const. amend. X ; Jacobson v. Massachusetts , 197 U.S. 11, 2525 S.Ct. 35849 L.Ed. 643 (1905). Indeed, the Court has called it a “settled principle[ ]” that states enjoy a police power to promulgate “legislative enactment[s to] protect the public health and the public safety.” Jacobson , 197 U.S. at 2525 S.Ct. 358 ; see also Chicago, B. & Q. Ry. Co. v. Illinois , 200 U.S. 561, 59226 S.Ct. 34150 L.Ed. 596 (1906) (holding that “the police power of a State embraces … regulations designed to promote the public health”); Berman v. Parker , 348 U.S. 26, 3275 S.Ct. 9899 L.Ed. 27 (1954) (describing regulation of “public health” as a “traditional application of the police power”). And in the specific context of compulsory vaccination, the Court has twice confirmed that the propriety of such mandates is a matter vested to the police power of the states. See Jacobson , 197 U.S. at 24–2525 S.Ct. 358 ; Zucht v. King , 260 U.S. 174, 17643 S.Ct. 2467 L.Ed. 194 (1922) (describing it as “within the police power of a State to provide for compulsory vaccination”).

Those holdings notwithstanding, OSHA invokes the Commerce Clause to suggest that it is really the federal government, not the states, that enjoys the authority to mandate vaccination for employees nationwide. Before I explore the constitutional validity of that position, let me first explain why I label the standard a de facto national vaccine mandate for eighty-million Americans. OSHA has not minced words about the purpose and effect of its standard; according to OSHA itself, “[c]overed employers must develop, implement, and enforce a mandatory COVID-19 vaccination policy” for their employees. 86 Fed. Reg. 61,402 (Nov. 5, 2021) (emphases added). Thus, some half of our workforce must either become vaccinated or both (1) “wear a face covering at work in lieu of vaccination” and (2) submit to weekly testing for COVID-19. Id. Neither OSHA nor the employer is required to bear the expense. Id. at 61,532. Rather, it falls on the unvaccinated employee to shoulder the costs of compliance. Id. And if states do not adopt OSHA’s standard or some other plan that is “at least as effective,” they face penalties like the revocation of approval of their State Plans and the associated loss of millions in federal funding. See ” Emergency Temporary Standard,” Occupational Safety and Health Administration, https://www.osha.gov/coronavirus/ets2/faqs (last visited Dec. 14, 2021) (explaining that if a State Plan is not “at least as effective” as OSHA’s emergency rule, consequences include “OSHA’s reconsideration and possible revocation of the State Plan’s final approval status”); see also “What is an OSHA-Approved State Plan?”, id. , https://www.osha.gov/stateplans/faqs (last visited Dec. 14, 2021) (“OSHA approves and monitors all State Plans and provides as much as 50 percent of the funding for each program.”).

So again, what constitutional warrant does OSHA possess for this scheme? The agency appeals to commerce. But the Commerce Clause likely cannot be read to grant such an authority, because it cannot be read to confer a general police power upon the national government. True, the Court has at times read the Clause broadly, stretching its meaning to the edge of plausibility. See, e.g. , Wickard v. Filburn , 317 U.S. 11163 S.Ct. 8287 L.Ed. 122 (1942). Yet the Court has never crossed the Rubicon of declaring a federal police power. Time after time, it has rejected the notion that such a power exists. See United States v. Lopez , 514 U.S. 549, 566115 S.Ct. 1624131 L.Ed.2d 626 (1995) (explaining  that the Constitution “withhold[s] from Congress a plenary police power”); id. at 584, 115 S.Ct. 1624 (Thomas, J., concurring) (“[W]e always have rejected readings of the Commerce Clause and the scope of federal power that would permit Congress to exercise a police power[.]”); Nat’l Fed. of Indep. Bus. , 567 U.S. at 536132 S.Ct. 2566 (“Our cases refer to this general power of governing, possessed by the States but not by the Federal Government, as the ‘police power.’ “). So the Commerce Clause, which generated “no apprehensions” upon its addition to the Constitution, cannot be read to effect a late-breaking revolution in state-federal affairs by granting a federal agency the right to regulate a core area of traditional state concern. The Federalist No. 45, supra , at 290 (James Madison).

What first principles dictate, fresh precedent confirms. The Supreme Court in recent years has squarely rejected a view of the commerce power under which “individuals may be regulated … whenever enough of them are not doing something the Government would have them do.” Nat’l Fed. of Indep. Bus. , 567 U.S. at 553132 S.Ct. 2566 (opinion of Roberts, C.J.); accord id. at 649–60, 132 S.Ct. 2566 (Scalia, Kennedy, Thomas, and Alito, JJ., dissenting). The case I mention involved an individual mandate to coerce those without health insurance to purchase it. Id. Congress claimed the power to regulate the failure to engage in a commercial activity—the buying of insurance—because uninsured persons’ failure to do so had a substantial aggregate effect on interstate commerce. Id. at 554, 132 S.Ct. 2566. Here, by contrast, OSHA claims the power to regulate the failure to engage in a non-commercial activity—the taking of a vaccine—because unvaccinated persons’ failure to do so may affect interstate commerce. OSHA’s theory of the commerce power is thus even more extravagant than what the Supreme Court has already rejected. If Congress cannot solve a perceived commercial problem with a “mandatory purchase,” then how can it possess the authority, much less delegate it, to solve a perceived commercial problem by mandating that Americans engage in a non-commercial activity? Id. at 553, 132 S.Ct. 2566. The answer, of course, is that it likely cannot.

The states arguing in support of the stay put it this way: If Congress does not have the power under the Commerce Clause to force individuals to buy health insurance, could it make an end-run around that rule by telling employers that they cannot retain uninsured employees? And if Congress cannot do so, then why can it tell employers that they cannot retain unvaccinated employees?

Before I turn to history, let me close with a final word on precedent, lest I be misunderstood. Here, I do not question the constitutionality of OSHA itself, or of federal workplace-safety regulations more broadly. But see Cass R. Sunstein, Is OSHA Unconstitutional? , 94 Va. L. Rev. 1407 (2008) (questioning OSHA’s constitutionality on non-delegation grounds). For even accepting that Congress (and thus, perhaps, OSHA) has the power to regulate a workplace hazard that affects interstate commerce, that is not what OSHA has done. OSHA has instead pretextually redefined what is at this point a hazard of life in the United States and throughout the world—COVID-19—as a hazard of the workplace. See, e.g. , 86 Fed. Reg. at 61,545 (misleadingly characterizing COVID-19 as a “workplace hazard”). It engages in this pretext in its attempt to bring a traditional matter of state concern—compulsory vaccination—within the ambit of federal jurisdiction. But caselaw is clear. Neither Congress nor OSHA may pretextually relabel such an area as “commerce” to gain what  is, in effect, a novel police power of the national government. See Morrison , 529 U.S. at 616–18120 S.Ct. 1740 (rejecting the notion that Congress may regulate domestic violence merely because of a purported “effect on interstate commerce”); see also id. at 617–18, 120 S.Ct. 1740 (“The Constitution requires a distinction between what is truly national and what is truly local.”); Lopez , 514 U.S. at 567–68115 S.Ct. 1624 (rejecting Congress’s attempt to relabel firearms near schools a problem of interstate commerce).

II.

Given that OSHA is so disarmed of precedent, one might reasonably have expected it to come into court bearing historical examples of the power it seeks to exercise—the federal imposition of a de facto nationwide vaccine mandate. Yet it has none. To the contrary, the relevant history actually undercuts OSHA’s position. For while Congress has long sought to facilitate safe and effective vaccines, it has never invoked the commerce power to mandate their administration upon the public at large.

To be sure, the federal government has, at one time or another, mandated vaccination for discrete segments of the population, such as for soldiers or members of the foreign service working abroad. George Washington himself ordered that his soldiers in the Continental Army receive variolation against smallpox in the winter of 1777. See Ann M. Becker, Smallpox in Washington’s Army: Strategic Implications of the Disease During the American Revolutionary War , 68 J. of Mil. Hist. 381, 427–28 (2004). But the relevant question is not whether the federal government has the authority to order the vaccination of certain populations in a special relationship with it. What is at stake here is whether Congress has a general police power to mandate vaccination for tens of millions of private citizens with no special relationship to the federal government. History suggests that it has no such power.

In the early years of the Republic, Congress did little to respond to epidemics. In the summer of 1793, for example, yellow fever descended on Philadelphia, then the nation’s capital. See Letter from Thomas Jefferson to Martha Jefferson Randolph (Sept. 8, 1793), Founders Online, https://founders.archives.gov/documents/Jefferson/01-27-02-0060 (last visited Dec. 12, 2021); see also Letter from George Washington to Edmund Randolph (Sept. 30, 1793), Founders Online, https://founders.archives.gov/documents/Washington/05-14-02-0105 (last visited Dec. 12, 2021); James Higgins, “Public Health,” Encyclopedia of Greater Philadelphia, https://philadelphiaencyclopedia.org/archive/public-health/ (last visited Dec. 12, 2021); Mathew Carey, A Short Account of the Malignant Fever, Lately Prevalent in Philadelphia 11 (1794), available at Harv. Univ. Lib. Viewer, https://iiif.lib.harvard.edu/manifests/view/drs:7374219$11i (last visited Dec. 14, 2021) (describing the “destroying scourge, the malignant fever,” that had “crept in among us”). The federal government’s response was primarily to leave town for the countryside. See Letter from Thomas Jefferson to Martha Jefferson Randolph, supra. President Washington chose to work remotely at Mount Vernon; the Secretary of State, Thomas Jefferson, fled to Monticello. Id.

And when it did intervene, it did not impose unilateral mandates upon the states, but instead assisted in a cooperative fashion. See Act of May 27, 1796, 4 Cong. Ch. 31, 1 Stat. 474 (authorizing the President to “aid in the execution of quarantine, and also in the execution of the health-laws of the states” during a yellow-fever epidemic).

There was no vaccine available in the 1790s for yellow fever but, in 1796, Sir Edward Jenner discovered a vastly improved vaccination for smallpox—rather  than use live virus as had the earlier “variolation” process, Jenner used cowpox instead. See Stefan Riedel, Edward Jenner and the History of Smallpox and Vaccination , 18 Baylor U. Med. Ctr. Proceedings 21, 23–24 (2005). That discovery led Congress less than two decades later, in 1813, to enter the vaccine arena. See Tess Lanzarotta & Marco A. Ramos, Mistrust in Medicine: The Rise and Fall of America’s First Vaccine Institute , 108 Am. J. of Pub. Health 741 (2018). In response to an outbreak of smallpox, Congress passed “An Act to Encourage Vaccination,” sometimes called the Vaccine Act of 1813. Id. at 742; see also James Colgrove, Immunity for the People: The Challenge of Achieving High Vaccination Coverage in American History , 122 Pub. Health Rep. 248, 249 (2007).

The Act had three salient features: it created the position of a federal vaccine agent, gave him the authority to curate an unadulterated supply of smallpox vaccine, and gave him a franking privilege to distribute vaccines to those who requested them, free of charge, through the U.S. mail. Id. Noted Maryland physician James Smith served as the nation’s first (and only) vaccine agent for nine years, overseeing “twenty agents nationwide who inoculated around 100,000 people” during his tenure. See Letter from James Smith (of Baltimore) to Thomas Jefferson (Mar. 28, 1818), Founders Online, https://founders.archives.gov/documents/Jefferson/03-12-02-0472 (last visited Dec. 12, 2021). Yet Smith’s role as vaccine agent—and the Vaccine Act itself—came to a tragic end in 1822. See Lanzarotta & Ramos, supra , at 742. Smith accidentally shipped packages of live smallpox (rather than cowpox vaccine) to the town of Tarboro, North Carolina, resulting in ten fatalities. Id. Two months later, President Monroe dismissed Smith from his position and Congress repealed the Act, relinquishing further vaccination efforts to the states. Id.

Public response to the vaccine was strikingly similar to modern attitudes about the COVID vaccine. Many voluntarily took the smallpox vaccine and gave it to their children. See, e.g. , The Diaries of Gouverneur Morris: New York 1799–1816, 777 (Melanie Randolph Miller, ed. 2018); see also Letter from Abigail Adams to John Adams (July 13, 1776), Mass. Hist. Society, https://www.masshist.org/digitaladams/archive/doc?id=L17760713aa (last visited Dec. 14, 2021). But others, like some today, were suspicious of a vaccine. See Cynthia M.A. Geppert & Reid A. Paul, The Shot That Won the Revolutionary War and Is Still Reverberating , Fed. Practitioner 298, 298 (2019), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6654165/pdf/fp-36-07-298.pdf (last visited Dec. 14, 2021). And smallpox was as devastating and transmissible, if not more so, than COVID-19. See “History of Smallpox,” Centers for Disease Control and Prevention, https://www.cdc.gov/smallpox /history/history.html (last visited Dec. 12, 2021). Yet there is no indication that anyone in the 1813 Congress thought the federal government had a general police power to nationally mandate vaccination.

Congressional involvement in vaccination ever since has followed the basic contours of the 1813 regime. Congress has passed many laws to regulate the purity of vaccines, facilitate their distribution with information and funding, and compensate those injured by their administration, but it has apparently never invoked the commerce power to mandate their imposition  upon the general public. See, e.g. , Biologics Control Act, Pub. L. No. 57-244, 32 Stat. 728 (1902) (current version at 42 U.S.C. § 262 (2006) ); see also Pure Food and Drug Act of 1906, Pub. L. No. 59-384, 34 Stat. 768; Virus-Serum-Toxin Act, ch. 145, § 1, 37 Stat. 832 (1913) (current version at 21 U.S.C. §§ 151 – 159 ); Federal Food, Drug, and Cosmetic Act, Pub. L. No. 75-717, 52 Stat. 1040 (1938); Public Health Service Act, Pub. L. No. 78-410, 58 Stat. 682 (1944); Poliomyelitis Vaccination Assistance Act of 1955, Pub. L. No. 277, 69 Stat. 704; National Childhood Vaccine Injury Act, Pub. L. No. 99-660, 100 Stat. 3755 (1986) ; Food and Drug Modernization Act of 1997, Pub. L. No. 105-115, 111 Stat. 2296.

I pause to note a seeming counterexample that is, upon further inspection, no counterexample at all. In 1832, Congress passed the Indian Vaccination Act—a functional vaccine mandate for those tribes selected for smallpox vaccination by federal Indian agents. See J. Diane Pearson, Lewis Cass and the Politics of Disease: The Indian Vaccination Act of 1832 , 18 Wicazso Sa Rev. 9, 12 (2003) (noting that “it was left to the secretary of war to determine which American Indians were vaccinated and when and where they would be vaccinated. American Indians had no input into any of the political or decision-making processes involved with the bill or into implementation of the act.”). The Act’s marketing was beneficent, id. at 10, but its administration was sinister. Indian agents selected for vaccination (1) those tribes scheduled for removal, so that smallpox would not derail the journey, id. at 25, and (2) tribes that were considered valuable trading partners of the United States. Id. at 19–23. By contrast, tribes considered “beyond the pale of civilization” were deliberately excluded from vaccination. Id. at 20. Even if a modern agency were inclined to rely on this poisoned precedent, see Ramos v. Louisiana , ––– U.S. ––––, 140 S. Ct. 1390, 1401 & n.44, 206 L.Ed.2d 583 (2020), it would do nothing to advance an interpretation of the Commerce Clause. Congress regulates Indian tribes’ internal affairs under a supposed “plenary power”—much as a state would regulate its own citizens—rather than under its commerce authority. See Puerto Rico v. Sanchez Valle , 579 U.S. 59, 70136 S.Ct. 1863195 L.Ed.2d 179 (2016).

The Poliomyelitis Vaccine Assistance Act of 1955 provides a good example. Soon after Dr. Jonas Salk developed the first effective polio vaccine in 1955, Congress responded with millions of dollars in “grants to assist states in vaccinating children under 20 and expectant mothers,” with funds “allotted to the states” according to their respective needs. See Otis L. Anderson, The Polio Vaccine Assistance Act of 1955 , 45 Am. J. Pub. Health 1349, 1349 (1955). Yet it was “the states [that had] responsibility for the intrastate distribution of the vaccine through both public agency and normal commercial channels.” Id. (emphasis added); see also 42 U.S.C. § 243(a) (directing the Secretary of Health and Human Services to “assist States … in the prevention and suppression of communicable diseases” and to “cooperate with and aid State and Local authorities.” (emphases added)).

OSHA would turn this history on its head. It proposes not a partnership in which the federal government simply encourages vaccination, but an unfunded mandate in which half our workforce must either become vaccinated or subject itself to regular out-of-pocket testing. See 86 Fed. Reg. at 61,532. If Congress purported to delegate such a sensitive “money or lives” determination to an unaccountable agency, we would have to think hard about the propriety of that delegation. See Indus. Union Dep’t, AFL-CIO v. Am. Petroleum Inst. , 448 U.S. 607, 687100 S.Ct. 284465 L.Ed.2d 1010 (1980) (Rehnquist, J., concurring in the judgment). Yet here there likely existed no authority to delegate.

III.

I have no doubt that the pandemic imperils our society, and I recognize that there is sometimes a “judicial impulse to stay out of the way in times of crisis.”  Roman Catholic Diocese of Brooklyn v. Cuomo , ––– U.S. ––––, 141 S. Ct. 63, 71208 L.Ed.2d 206 (2020) (Gorsuch, J., concurring). But while an “emergency may afford a reason for the exertion of a living power already enjoyed,” it cannot “call into life a power which has never lived.” Wilson v. New , 243 U.S. 332, 34837 S.Ct. 29861 L.Ed. 755 (1917). OSHA claims just such a power—history and precedent notwithstanding. It is surely incumbent on the third branch in these circumstances to check the actions of the “fourth.” And because the full court should have had the opportunity to do so, I respectfully dissent.