Rationality is not irrational

The Fifth Circuit's decision in Louisiana v. FDA is wrong.

On May 1, 2026, the Fifth Circuit stayed the FDA’s decision to permit women to obtain mifepristone from pharmacies and by mail. According to the Fifth Circuit, the FDA did not think hard enough about safety when it made this decision. The Fifth Circuit’s decision, if it goes into effect, would impose a nationwide requirement that women travel to their health care providers to obtain mifepristone, even when the woman, her doctor, her state legislature, and the FDA all think this purported safety measure is unnecessary.

The Supreme Court has temporarily blocked the Fifth Circuit’s order and is considering whether to keep that block in place while the case proceeds. It should. The Fifth Circuit erred. Louisiana, the plaintiff challenging the FDA’s decision, lacks standing. The FDA’s decision was also legal. It was not arbitrary and capricious for the FDA to conclude, based on an extensive body of evidence and years of experience with mifepristone, that this safety measure is unnecessary.

The Mifepristone Wars

The FDA regulates drug safety. When it approves a drug, it has the authority to impose a “risk evaluation and mitigation strategy”—known as “REMS” for acronym-loving lawyers—when “necessary to ensure the benefits of the drug outweigh the risks of the drug.”

Mifepristone is an FDA-approved drug taken by women to terminate pregnancies. Pre-COVID, the REMS for mifepristone required women to go to their health care providers, in person, to pick up mifepristone, rather than obtaining it at the pharmacy or by mail. They didn’t have to take the mifepristone in the provider’s presence. They just had to physically pick it up there.

In July 2020, during COVID, a federal court enjoined this in-person dispensing requirement. In January 2021, the Supreme Court stayed this injunction and the in-person dispensing requirement sprang back into force. But in April 2021, the FDA said it would stop enforcing the requirement. In December 2021, following a review of the REMS, the FDA concluded that in-person dispensing wasn’t necessary to protect safety and decided that the REMS should be modified.

A modification to the REMS occurs only after the drug manufacturer submits a REMS modification and the FDA approves that modification. And so, in December 2021, the FDA directed the drug manufacturers to submit proposed REMS modifications. They did. The FDA approved the modified REMS in January 2023.

This latest episode of the Mifepristone Wars began in October 2025, when Louisiana sued the FDA. The purpose of this suit was to overturn the FDA’s decision to modify the REMS and force the FDA to restore the in-person dispensing requirement.

After the FDA told the district court that it was reconsidering the REMS, the district court elected to give the FDA the time it needed to complete that review and declined to grant Louisiana any immediate relief. Louisiana took the case to the Fifth Circuit, which issued an order immediately blocking the REMS modification and effectively reimposing the in-person dispensing requirement nationwide.

Protecting safety in order to not protect safety

Louisiana may be standing in this image, but it does not have standing in this case.

To establish standing, Louisiana must show that it is experiencing an injury caused by the FDA’s decision to modify the REMS that would be redressed by an order overturning that decision.

Louisiana offers two theories of standing. Theory #1 goes like this: if it’s possible to obtain mifepristone by mail, then women in Louisiana will order mifepristone by mail from out-of-state providers, which inflicts what Louisiana refers to as a “sovereign injury” because abortion is illegal in Louisiana.

Louisiana’s claim on the merits—the thing that, according to Louisiana, the FDA did wrong—is that the FDA was insufficiently attentive to women’s safety when it jettisoned the in-person dispensing requirement. However, Louisiana does not actually want women to take the extra safety measure of picking up mifepristone from their providers. To the contrary, it wants women not to do this. It bans local hospitals from doing the very thing that, it contends, should be mandatory for safety purposes—in-person dispensing of mifepristone. Women could in principle drive to a state like Illinois to try to obtain mifepristone, but Louisiana is hoping that women will find this too burdensome and will instead carry the pregnancy to term.

Ironically, under the REMS statute, the FDA must periodically evaluate whether the elements of the REMS “assure safe use of the drug” and “are not unduly burdensome on patient access to the drug.” In a normal world, a plaintiff trying to force the FDA to impose a more stringent REMS requirement would be saying things like “the requirement will assure safe use of the drug and won’t be unduly burdensome.” But Louisiana is saying the exact opposite. It’s saying it doesn’t want “safe use of the drug” and is hoping it will be “unduly burdensome” for Louisiana women to drive to a different state to get the drug. Sure, there are lots of women in other states where abortion is legal who will be forced to take safety measures they don’t want or need, but from Louisiana’s perspective, that’s just collateral damage. In the Fifth Circuit, this argument wins.

Consider the situation from the perspective of a woman living in a state where abortion is legal. She wants to obtain mifepristone by mail or at the pharmacy, but, if the Fifth Circuit’s order goes into effect, she can’t. Instead, purportedly for purposes of protecting safety, she has to drive to a further-away building. But she doesn’t want this extra safety measure. Her doctor doesn’t recommend this extra safety measure. Her state—which, Dobbs claims, now gets to decide abortion policy—doesn’t support this extra safety measure. The FDA doesn’t think this extra safety measure is necessary. Yet she’s forced to take this extra safety measure, because Louisiana convinced the Fifth Circuit to mandate the safety measure nationwide so that Louisiana’s citizens also don’t take the safety measure.

Article III’s “case or controversy” requirement exists to guard against precisely this sort of vexatious litigation. In FDA v. Alliance for Hippocratic Medicine, a group of pro-life doctors brought a lawsuit seeking to ban all women in America from using mifepristone. Their theory of standing was that if mifepristone were banned, they wouldn’t face the risk of encountering women who had taken mifepristone in the emergency room. The Supreme Court rejected this theory. The Court explained that “FDA has not required the plaintiffs to do anything or to refrain from doing anything.” It observed that “the law has never permitted doctors to challenge the government’s loosening of general public safety requirements simply because more individuals might then show up at emergency rooms or in doctors’ offices with follow-on injuries.” If the plaintiffs’ theory were correct, the Supreme Court explained, absurd consequences would follow: “Police officers,” for example, “could sue to challenge a government decision to legalize certain activities that are associated with increased crime.”

That’s this case. The FDA’s decision to modify the REMS doesn’t require Louisiana to do, or not do, anything. Instead, Louisiana is suing the FDA to stiffen safety restrictions for purposes of decreasing crime in Louisiana. Under Alliance for Hippocratic Medicine, it cannot do this.

But, you might retort, Louisiana is a state, not a private citizen, so the Supreme Court’s reasoning in Alliance for Hippocratic Medicine shouldn’t apply. Consider, then, United States v. Texas. Texas and Louisiana sued the Department of Homeland Security, alleging that the Department was illegally refusing to enforce statutes requiring it to arrest and detain certain aliens. The plaintiffs’ theory of standing was that these aliens, if not arrested, would impose burdens on the states—for instance, they might commit crimes in Texas and Louisiana, forcing the states to spend money incarcerating them.

The Supreme Court rejected this argument. As the Court explained, “when the Executive Branch elects not to arrest or prosecute, it does not exercise coercive power over an individual’s liberty or property, and thus does not infringe upon interests that courts often are called upon to protect.” The Court continued: “the absence of coercive power over the plaintiff makes a difference: When a plaintiff’s asserted injury arises from the government’s allegedly unlawful regulation (or lack of regulation) of someone else, much more is needed to establish standing.”

This case involves the Executive’s failure to impose enhanced safety requirements, as opposed to the Executive’s failure to arrest and prosecute people, but the same principle applies. According to Louisiana, the FDA injured it by imposing overly lax safety requirements on one set of third parties that are not in Louisiana (drug manufacturers), which in turn makes it easier for another set of third parties who are also not in Louisiana (doctors) to send mifepristone to a third set of third parties, this time in Louisiana (women seeking abortions). That’s not a cognizable injury. Louisiana can’t force the FDA to impose more stringent nationwide regulations on drugs for the purpose of making it harder for Louisianans to obtain the drug from out-of-state doctors.

The last case in the trilogy is Diamond Alternative Energy, LLC v. EPA. EPA approved California regulations requiring automakers to manufacture fewer liquid-fuel-powered vehicles. Fuel producers sued EPA over that approval. The Supreme Court held that the producers had standing. The whole purpose of California’s regulations was to reduce gasoline consumption. EPA put its stamp of approval on those regulations. Fuel producers are harmed by regulations designed to reduce fuel consumption. As the Supreme Court put it, the fuel producers had standing because they were an “object of the California regulations because the regulations explicitly seek to restrict the use of gasoline and other liquid fuels in automobiles.” That reasoning doesn’t help Louisiana here, though. Louisiana isn’t the “object” of the FDA’s action. The FDA isn’t restricting Louisiana from doing anything. It’s merely declining to impose nationwide safety restrictions.

Targeting and not targeting Louisiana simultaneously

Why, then, does the Fifth Circuit conclude that Louisiana has standing?

The opening stanza of the Fifth Circuit’s decision goes like this:

In Dobbs v. Jackson Women’s Health Organization, 597 U.S. 215 (2022), the Supreme Court returned the regulation of abortion to the states. In response, the Biden Administration directed federal agencies to “expand access to . . . medication abortion.” Exec. Order No. 14076, 87 Fed. Reg. 42053 (July 8, 2022). The next year, the Food and Drug Administration (FDA) formally altered its safety guidelines for the abortion drug mifepristone.

And then Executive Order No. 14076 makes another appearance:

An avowed purpose of the 2023 REMS was to expand access to medication abortion. See Exec. Order No. 14076, 87 Fed. Reg. 42053 (July 8, 2022) (in the wake of Roe’s overruling, directing federal agencies to “expand access to . . . medication abortion” by “protect[ing] healthcare service delivery and promot[ing] access to . . . abortion”).

Apparently based on this 2022 Executive Order, the Fifth Circuit makes the leap that:

[E]nsuring out-of-state medical providers could prescribe mifepristone to women in states that restrict abortion was a goal of the regulation.

The Fifth Circuit then reiterates:

The policy does not merely “increase[] crime or disorder, or impose[] indirect compliance costs for state law enforcement.” Washington v. FDA, 108 F.4th 1163, 1177 (9th Cir. 2024). Rather, the 2023 REMS sanctions and facilitates conduct with the express purpose of undermining Louisiana’s legal restrictions on abortion.

In these passages, the Fifth Circuit is suggesting that the FDA issued the 2023 REMS for the “express purpose” of helping women evade pro-life laws. The two citations to Executive Order No. 14076 are meant to suggest that the FDA implemented the 2023 REMS pursuant to that Executive Order for purposes of expanding access to abortion post-Dobbs. The Fifth Circuit is implying that the 2023 REMS targeted Louisiana and other pro-life states, so that the Fifth Circuit can claim, under Diamond Alternative Energy, that Louisiana has standing to challenge it.

The Fifth Circuit’s implication is incorrect. The REMS was not issued as part of a post-Dobbs effort to evade anti-abortion laws. Executive Order No. 14076 directs the Secretary of Health and Human Services to submit a report to the President identifying “potential actions” to “expand access to abortion care, including medication abortion.” The Secretary duly submitted a report to the President which the current Administration has annoyingly deleted from HHS’s website. However, a press release summarizing the report survives online, and it does not mention the REMS for mifepristone.

Why not? Because the FDA had already decided to change the REMS in 2021, before Dobbs, following its COVID-based decision not to enforce the in-person dispensing requirement. The post-Dobbs approval of the REMS in 2023 was the ministerial approval of a change that the FDA had already announced well before Dobbs was decided.

It’s not just me who is saying this. This is the explicit premise of a different part of the Fifth Circuit’s decision. Louisiana failed to administratively exhaust its challenge to the 2023 REMS—it went straight to court. The Fifth Circuit excuses this failure on the theory that because the FDA had already decided to get rid of the in-person dispensing requirement in 2021—pre-Dobbs—the 2023 REMS was a foregone conclusion, so there was no need for Louisiana to administratively challenge it. The Fifth Circuit says that the “FDA’s denials of citizen petitions” (which occurred in 2021) demonstrates that the FDA was “committed to implementing these changes” and there is “no reason to think that, today, FDA would administratively stay the 2023 REMS, which formalized the 2021 decision.”

The Fifth Circuit’s exhaustion analysis is also wrong, but let’s set that to the side. In the span of a single judicial opinion, the Fifth Circuit is simultaneously saying:

1. Louisiana didn’t have to exhaust because the FDA had already decided to change the REMS before Dobbs.

2. Louisiana has standing because the “express purpose” of the REMS was to implement a post-Dobbs Executive Order.

I want to return to the Fifth Circuit’s conclusion that “the 2023 REMS sanctions and facilitates conduct with the express purpose of undermining Louisiana’s legal restrictions on abortion.” There’s a whole genre of litigation against the Trump Administration in which litigants challenge government actions while arguing that the Administration had an “express purpose” that didn’t appear on the face of the orders announcing those actions. Typically the litigants gamely cobble together statements President Trump made over the years and ask the court to infer that the actions were rooted in illicit motives. Versions of this argument failed during Trump I in challenges to the DACA repeal and the travel ban, which were alleged to be motivated by anti-Latino and anti-Muslim animus, respectively.

Here, though, the Fifth Circuit skips over the step of cobbling together the evidence of the government’s purported “express purpose.” It just declares that, verily, the 2023 REMS had “the express purpose of undermining Louisiana’s legal restrictions on abortion.” The REMS doesn’t say this, the Administration never said this, the FDA directed the REMS to be changed when Louisiana’s abortion ban wasn’t even operative … whatever, nothing matters. The Fifth Circuit has declared it to be Law that the Biden Administration had the “express purpose” of targeting Louisiana, and so now all Americans are rewarded with safety restrictions they don’t need or want.

A single penny

Let me turn to Louisiana’s second theory of standing: financial injury. Louisiana claims that women who receive mifepristone by mail might have complications and wind up in an emergency room in Louisiana, causing Louisiana to expend Medicaid funds. According to Louisiana, in 2025, two women in Louisiana required emergency care based on mifepristone use, leading to $92,000 in Medicaid costs. Therefore, the argument goes, if all women in America are banned from obtaining mifepristone from the pharmacy or by mail, Louisiana will save money.

This $92,000 figure comes from the declaration of Kathleen Willis, who is the Associate Medical Director for Louisiana Medicaid in the Louisiana Department of Health. It turns out that, according to Dr. Willis’s declaration, $80,000 out of the $92,000 corresponds to the cost of caring for a baby that was born alive and that was allegedly harmed by the anonymous mother’s ingestion of mifepristone that she purportedly obtained by mail. (The other $12,000 was spent on the mother’s care and the care of one other anonymous woman.) I have some questions about whether Dr. Willis can establish personal knowledge of her assertions, but even assuming these events occurred as stated, Louisiana cannot establish standing based on these expenditures.

As the Supreme Court put it in United States v. Texas: “[I]n our system of dual federal and state sovereignty, federal policies frequently generate indirect effects on state revenues or state spending. And when a State asserts, for example, that a federal law has produced only those kinds of indirect effects, the State’s claim for standing can become more attenuated.” To put it more bluntly: If “the federal government has done something that might indirectly cause a state to spend money” was a sufficient basis for standing, then Article III would be dead. Any government decision to regulate or not regulate something might lead to something bad happening to someone, causing the state to spend at least one penny addressing that bad thing. If that were enough to establish standing, then states could sue the federal government in every case. Article III exists to ensure that litigants—including states—cannot file a lawsuit every time they disagree with something the government did.

The standing requirement is not a technicality. It is fundamental to the separation of powers. Justice Scalia’s classic article, “The Doctrine of Standing as an Essential Element of the Separation of Powers”—where he bemoans “the increasingly frequent administrative law cases in which the plaintiff is complaining of an agency’s unlawful failure to impose a requirement or prohibition upon someone else”—expresses the point far better than I ever could. Compressing the point into a couple of sentences: judges are unelected officials with extraordinary power to overturn the will of the political branches. That power is modulated by the requirement that courts must resolve bona fide cases or controversies. Allowing ideologically motivated litigants to use the judiciary to ram through regulations desired by neither the regulator nor the regulated inverts that constitutional design, converting judicial review from a check on governmental power into a mechanism for manufacturing it.

Finally, I have to add a word about Louisiana’s co-plaintiff, Rosalie Markezich. The complaint alleges that, in 2023, Ms. Markezich “felt coerced by her boyfriend” to take mifepristone that was ordered by mail. (Louisiana’s filings refer to her as “Rosalie,” as though she is a child. I will stick with Ms. Markezich.) Past injury is not enough for Ms. Markezich to establish standing, however; to establish standing to obtain forward-looking relief, she has to show a sufficiently imminent threat of future harm. Ms. Markezich’s theory is that she might become pregnant via some hypothetical future boyfriend who will, like boyfriend #1, pressure her to take mifepristone ordered by mail. To ward off this threat and (in the complaint’s words) “protect her future unborn children,” she feels she is entitled to a court order banning all women in America, including her, from obtaining mifepristone by mail, which will in turn allow her to resist the hypothetical future boyfriend and carry this hypothetical unplanned pregnancy to term. (The complaint puts it this way: “Without a requirement for an in-person office visit to prevent coercion, Rosalie could be placed in the same position for future pregnancies … Rosalie has a strong interest in reinstating the in-person dispensing requirement to prevent future coercion.”)

The Fifth Circuit states demurely that it “need not consider whether Markezich also has standing.” You need not look any further than this argument to see why Article III is a good thing.

You can just do things

Turning to the merits, the Fifth Circuit concludes that the 2023 REMS was unlawful.¹ The court leans heavily on a prior Fifth Circuit decision that was ultimately reversed for lack of jurisdiction. I’ve written prior blog posts about why that prior decision was also wrong, but the issues are worth a fresh look.

The Fifth Circuit notes that the current FDA declined to defend the prior FDA’s reasoning. There is a deeply rooted American tradition of administrations conceding that the prior administration from the opposite party acted irrationally—as far back as 1797, President Adams’ FDA Director threw President Washington’s FDA Director under the bus over the approval of foxglove for dropsy—so this does not mean much.

The Fifth Circuit doesn’t suggest that the FDA violated any statute or regulation. Instead, the Fifth Circuit concludes that the safety-obsessed dorks at the FDA acted irrationally in concluding that eliminating the in-person dispensing requirement is safe.

Note that this was an in-person dispensing requirement. There’s no requirement that the woman take the drug at the hospital (most women take it at home). The requirement was that the woman drive to the health care provider, physically pick it up, and bring it home, rather than picking up a prescription from the doctor and going to the pharmacy or getting the drug in the mail. To my knowledge, there’s no other drug subject to this pick-it-up-at-the-hospital-but-take-it-at-home regime.

The FDA concludes that, as with other FDA-approved drugs that can be taken at home, it isn’t necessary to force women to drive to their health care providers to pick it up. The Fifth Circuit says that the FDA acted irrationally in concluding this wasn’t necessary.

Why?

A bit of background is needed. Before 2016, the FDA required that prescribers report all serious adverse events associated with mifepristone. In 2016, the FDA decided that, going forward, prescribers had to report only fatal events. The approving official reasoned: “After 15 years of reporting serious adverse events, the safety profile of Mifeprex is essentially unchanged. Therefore, I agree that reporting of labeled serious adverse events other than deaths can be collected in the periodic safety update reports and annual reports to the Agency.”

In 2021, the FDA rejected a challenge to this reporting change. Importantly, the FDA pointed out that although prescribers don’t have to report non-fatal serious adverse events, the manufacturers still do when they become aware of such events: “We also note that the reporting changes to the Prescriber Agreement Form as part of our 2016 approval do not change the adverse event reporting requirements for the Applicants.” It acknowledged that “there is always a possibility with any drug that some adverse events are not being reported,” but nonetheless concluded that, based on an assessment of “15 years of adverse event reports,” “certain ongoing additional reporting requirements under the Mifeprex REMS, such as hospitalization and blood transfusions, were not warranted.”

Importantly, virtually no drugs are subject to the requirement that prescribers report all serious adverse events. In 2016, the FDA said that after 15 years of experience, it was comfortable treating mifepristone the same way that it treats the vast majority of other drugs.

This is rational behavior by the agency. More reporting requirements are not always better. An agency is allowed to say that, after 15 years of experience, it has enough information about mifepristone’s safety profile that it does not need more paperwork.

In December 2021, as noted above, the FDA also concludes that the in-person dispensing requirement isn’t necessary anymore.

In reaching this conclusion, the FDA analyzes, among other things, adverse events between January 2020 and September 2021. Remember that the in-person dispensing requirement wasn’t enforced from July 2020 (when the district court issued its injunction) until January 2021 (when the Supreme Court stayed that injunction) and again starting in April 2021 (the date the FDA decided to stop enforcing the requirement). This creates a natural experiment: for part of that period (January 2020 to July 2020 and January 2021 to April 2021), the in-person dispensing requirement was in effect, and for part of that period (July 2020 to January 2021 and April 2021 to September 2021), it wasn’t.

The FDA analyzes all reported adverse events during that period and also conducts a separate analysis of adverse events from 2016 to 2021. After the FDA goes on and on for several paragraphs, it concludes:

Based on this review, we conclude that there does not appear to be a difference in adverse events between periods when the in-person dispensing requirement was being enforced and periods when the in-person dispensing requirement was not being enforced. This suggests that mifepristone may be safely used without an in-person dispensing requirement.

This reasoning isn’t good enough for the Fifth Circuit. According to the Fifth Circuit, the FDA acted arbitrarily and capriciously because it gave “dispositive weight” to the lack of adverse-event data in the reporting system, even though the FDA had previously eliminated the requirement that prescribers report that information. Hence, holds the Fifth Circuit, it’s “unreasonable for an agency to eliminate a reporting requirement for a thing and then use the resulting absence of data to support its decision.”

This is wrong, but please don’t take my word for it. The FDA’s analysis of this issue appears at pages 61 to 64 of this PDF (pages 21-24 of the December 2021 decision).² Read it for yourself. Then, ask yourself whether it is fair to say that the FDA acted arbitrarily and capriciously because it “eliminated a reporting requirement for a thing and then used the resulting absence of data to support its decision.”

The FDA exhaustively analyzed all data available to it and concluded it was safe to make a change to the REMS. Yes, five years earlier, it decided to stop collecting certain information from prescribers and start treating mifepristone like it treats virtually every other drug. But it is OK for an agency to take an action—even, heaven forbid, a deregulatory action—without having the maximum theoretical amount of information available to it.

It is not “arbitrary and capricious” for an agency to say:

1. We will not impose reporting requirements that don’t exist for virtually all other drugs and that we think are unnecessary.

2. We are willing to make safety decisions based on the information we have, even though we could, hypothetically, have imposed more stringent reporting requirements.

The Fifth Circuit seems to think that once the FDA relaxes a reporting requirement, safety requirements should be fossilized forever. It’s arbitrary and capricious, says the Fifth Circuit, for the FDA to say “based on the information we have, we think X is safe” because it could have had more safety information! This is not how rational people make decisions.

Also, we’re not talking here about a new dosage or administration regimen. We’re talking about a change to the physical place in which a woman picks up the drug. The FDA is allowed to make a predictive judgment on this based on the information it has.

Bunkers in Idaho

The Fifth Circuit’s second basis for overturning the FDA’s decision is that it “relied on various literature relating to remote prescription of mifepristone—despite FDA’s admission that the literature did not affirmatively support its position.” It quotes the FDA’s statement that the literature was “not adequate on [its] own to establish the safety of the model of dispensing mifepristone by mail,” and characterizes the FDA’s decision as a “textbook example of arbitrary and capricious agency action.”

Once again, I encourage you to read the FDA’s decision and decide for yourself whether it belongs in the arbitrary-and-capricious textbook. The relevant passage appears on page 79 of the PDF (page 39 of the December 2021 decision).

The FDA concludes that based on lots of different sources of information, including the published literature and other information, it’s safe to modify the REMS. It explains that the published literature isn’t on its own sufficient to establish safety, but it’s consistent with all of the other information the FDA reviewed, and so all of the information, taken together—including lots of data the FDA analyzed in excruciating detail which wasn’t in the published literature—establishes that the in-person dispensing requirement could be eliminated:

Based on the literature identified by our review, dispensing mifepristone by mail from the clinic or from a mail order pharmacy does not appear to jeopardize the efficacy of medical abortion. The studies we reviewed are not adequate on their own to establish the safety of the model of dispensing mifepristone by mail, although the safety and efficacy outcomes reported in these studies remain within the ranges described in mifepristone labeling except for increased numbers of ED/urgent care visits and hospitalizations.

…

Despite the limitations of the studies we reviewed, we conclude that overall, the outcomes of these studies are not inconsistent with our conclusion that, based on the 1st year REMS assessment report and postmarketing safety data, mifepristone will remain safe, and efficacy will be maintained if the in-person dispensing requirement is removed from the Mifepristone REMS Program.

Based on the REMS assessment data, FAERS data from the time period when the in-person dispensing requirement was not being enforced, our review of the literature, and information provided by advocacy groups, individuals, the Applicants, and the plaintiffs in the Chelius v. Becerra litigation, we conclude that mifepristone will remain safe and effective for medical abortion if the in-person dispensing requirement is removed, provided all the other requirements of the REMS are met, and pharmacy certification is added as described below.

The Fifth Circuit plucks one sentence from this analysis— “The studies we reviewed are not adequate on their own to establish the safety of the model of dispensing mifepristone by mail”—and says, aha! The FDA relied on inadequate information! Textbook example of arbitrary and capricious agency action!

I don’t know if I even have to say this. But sometimes, one thing, on its own, isn’t an adequate basis to establish a proposition, but several pieces of evidence taken together, including that thing, establish a proposition. If someone comes in with wet socks, that’s not, on its own, an adequate basis to establish that it’s raining. Perhaps the person just jumped in a puddle. But wet socks plus loud thunder sounds plus the smell of wet asphalt from the window plus a rain icon on one’s phone, taken together, can establish, with a high degree of certainty, that it’s raining.

There’s an irony here because for years, regulatory agencies in general, and the FDA in particular, have suffered from criticism that they are too obsessed with safety and don’t understand tradeoffs. They constantly demand more reporting without acknowledging that there are costs to always demanding more reporting. They unreasonably demand that the published literature be pristine. They refuse to acknowledge the reality that it’s sometimes OK to make decisions based on imperfect information.

There’s a whole literature on why agencies act this way. Maybe it’s because the type of person who becomes a regulator is an Ayn Rand villain who delights in preventing People of Action from doing things. Maybe it’s because agencies like exerting power and the only way to do that is to say no. Maybe it’s because harms from overly lax safety are more visible than harms from excessive safety. It is not uncommon for applicants to become so fed up with the FDA’s neuroses that they move to a bunker in Idaho surrounded by barbed wire.

Whatever the reason that agencies act that way, here the FDA didn’t act that way. It correctly recognized that sometimes it’s unnecessary to demand more data; that it’s OK to do things even if you have limited data; and that sometimes you have to consider lots of pieces of evidence taken together. These shouldn’t be controversial analytical moves. They are the ordinary tools of rational judgment. Yet the Fifth Circuit held that the FDA’s actions were arbitrary and capricious! In effect it held that the FDA acted irrationally because it acted rationally.

***

You might be thinking that I am taking the Fifth Circuit’s legal analysis too seriously. Perhaps, you might say, this case is just a one-off application of the Fifth Circuit’s Pro-Life Always Wins Doctrine and the court will return to a rational approach to administrative law in cases that are less ideologically charged. Obsessing over the details of the Fifth Circuit’s reasoning, you might say, is pointless because the court will never actually follow that reasoning in future cases.

That may be true. And if that is true, it is bad: courts should not selectively apply fake administrative-law principles for purposes of delivering victory to their favored litigants.

Still, the premise of our legal system is that courts resolve cases via neutral principles. Formally speaking, the administrative-law principles applied by the Fifth Circuit in this case are supposed to apply in every case. And so I still think lawyers should take the court’s reasoning seriously and point out its flaws, pointless though it may be.

1. Technically “likely unlawful,” because this was a stay order, but close enough.

2. Although the Fifth Circuit’s decision cites the FDA’s April 2021 non-enforcement decision, the relevant decision is the FDA’s December 2021 decision to direct a change to the REMS.