President Biden's Cannabis Scheduling Directive: Part 3
By Shane Pennington
Nov 9, 2022
The first two posts in this series introduced the formal rulemaking process that will govern the cannabis scheduling review that President Biden requested earlier this month. I covered the applicable law, the mechanics of the administrative process, and the tedious and enraging history of DEA’s use of that process to keep cannabis in schedule I unlawfully for decades. In this third installment in my series of posts on President Biden’s cannabis scheduling directive, I’m going to discuss the end game: What outcomes are possible from the impending administrative process to reconsider cannabis’s schedule I status? What are the implications of those various possible outcomes? And how should we all proceed in light of it all?
In Part I, I argue that descheduling is the outcome most aligned with rational drug policy reform. I also refute the widespread myth that rescheduling to schedules II-V would do more harm than good. While rescheduling is not the best result we can hope for, it certainly wouldn’t be a bad thing. Indeed, I’ll demonstrate that simply transferring cannabis from schedule I to schedule II—the most modest change to cannabis’s status under the CSA possible—would be a historic victory with several attendant benefits for industry, state and federal regulators, and the public. In Part II, I discuss the various barriers to rescheduling and descheduling and identify the strategic steps necessary to overcome them.
What’s happening here is historic. As I’ll explain, though, to make the most of the opportunity, we must navigate an obscure process under unprecedented circumstances. Developing a winning game plan will therefore be tricky business. Indeed, I believe there is one—and only one—strategy that makes descheduling a realistic possibility here, and as I explain below, it requires a carefully coordinated and expertly led effort from public and private stakeholders.
Part I: Identifying the Goal
A. Descheduling > Rescheduling
Many argue that those interested in sensible federal cannabis policy reform should adopt a “deschedule or bust” approach to the impending administrative process. Anything short of that, they say, would be a disaster. I’ll explain why that view and the various arguments commonly marshaled in support of it are wrong. I want to start, though, with something we all seem to agree about: descheduling is far and away the best outcome we could hope for from the administrative process. That is so for many reasons. I discuss three of them here.
First, descheduling is the only result that would remove cannabis from DEA’s purview and the CSA’s regime of federal criminal sanctions. See, e.g., 21 U.S.C. 844(a). The CSA’s prohibited acts and criminal sanction provisions apply uniformly to all controlled substances whether they’re listed on schedule I or schedule V. Take 21 U.S.C. 844(a). With limited exceptions not relevant here, that provision makes it “unlawful for any person knowingly or intentionally to possess a controlled substance ….”—full stop and without regard to its specific scheduling classification. 21 U.S.C. 844(a) (emphasis added). Or consider section 829(a), which addresses controlled substances listed in schedule V—the CSA’s least-restrictive classification—and expressly forbids “distribut[ing] or dispens[ing] [them] other than for a medical purpose.” 21 U.S.C. 829(c). Thus, even if cannabis were transferred to schedule V, virtually every transaction in the adult-use cannabis market today would remain criminal under federal law. The only way this administrative process fixes that core problem is through descheduling.
Second, descheduling is the only outcome that aligns with the reason President Biden gave in his October 6th announcement for requesting an administrative review of cannabis’s scheduling status in the first place. He emphasized that “no one should be in jail just for using or possessing marijuana,” and declared that he was taking “three steps … to end this failed approach”:
As I often said during my campaign for President, no one should be in jail just for using or possessing marijuana. Sending people to prison for possessing marijuana has upended too many lives and incarcerated people for conduct that many states no longer prohibit. Criminal records for marijuana possession have also imposed needless barriers to employment, housing, and educational opportunities. And while white and Black and brown people use marijuana at similar rates, Black and brown people have been arrested, prosecuted, and convicted at disproportionate rates.
Today, I am announcing three steps that I am taking to end this failed approach.
Descheduling is the only outcome consistent with that goal. Look at it this way, had cannabis never been listed as a controlled substance in the first place, the President wouldn’t have had anyone to pardon on October 6th. Leave cannabis on any of the CSA’s schedules, however, and jail time for cannabis possession and use—the essence of “failed approach” that President Biden hopes to “end” through this administrative process—will remain. See 21 U.S.C. 844(a); see also, e.g., id. 829(c) (prohibiting the “distribut[ion] [and] dispens[ation] [of schedule V substances] other than for a medical purpose”).
Of course, the President was quick to add that “even as federal and state regulation of marijuana changes, important limitations on trafficking, marketing, and underage sales should stay in place.” While many claim that qualification proves that President Biden would oppose descheduling, I’m not so sure. For one thing, that’s not what he said. As just explained, his stated goal—“end[ing]” the “failed” federal policy that puts people in jail for cannabis use and possession—can be accomplished only through descheduling. I’m therefore hesitant to read his subsequent qualifying statement in a way that assumes he contradicted himself, especially because there are sensible alternative interpretations that don’t create such a problem. President Biden might, for example, have intended his reference to “important limitations” that “should stay in place” to announce a secondary purpose, i.e. “important limitations … should stay in place” to
announce a secondary purpose, i.e. “important limitations … should stay in place” to the extent they don’t undermine the primary objective of “right[ing] the wrongs” of the federal government’s “failed” approach of criminalizing cannabis possession and use;
reference legal constraints that stop short of the sort of criminal sanctions that could result in jail time; or
limitations that exist under state law.
So, I think we should all take the President at his word when he says he hopes this process will ensure that no one ends up in jail for cannabis use or possession. I fully support that goal. And because descheduling is the only outcome consistent with it, prioritizing it over rescheduling as the primary goal of the administrative process strikes me as a no-brainer.
Third and finally, compared to rescheduling, descheduling would do more to close the “cannabis policy gap,” a term used to describe the dramatic and ever-widening gulf between the cannabis policies of the states that have legalized/decriminalized cannabis on the one hand and those of the federal government on the other. By removing cannabis from DEA oversight and the criminal penalties associated with federal controlled-substance status, descheduling would bring state and federal cannabis policy into alignment to a greater degree than they have been in decades.
That would be a very good thing on several levels. As Justice Thomas recently lamented, the cannabis policy gap has created a “contradictory and unstable state of affairs [that] strains basic principles of federalism and conceals traps for the unwary.” Standing Akimbo v. United States, 141 S.Ct. 2236, 2237 (Mem.) (Thomas, J., statement respecting denial of certiorari). He went on to describe the problem in greater detail:
This disjuncture between the Government’s recent laissez-faire policies on marijuana and the actual operation of specific laws is not limited to the tax context. Many marijuana-related businesses operate entirely in cash because federal law prohibits certain financial institutions from knowingly accepting deposits from or providing other bank services to businesses that violate federal law. Black & Galeazzi, Cannabis Banking: Proceed With Caution, American Bar Assn., Feb. 6, 2020. Cash-based operations are understandably enticing to burglars and robbers. But, if marijuana-related businesses, in recognition of this, hire armed guards for protection, the owners and the guards might run afoul of a federal law that imposes harsh penalties for using a firearm in furtherance of a “drug trafficking crime.” 18 U.S.C. § 924(c)(1)(A). A marijuana user similarly can find himself a federal felon if he just possesses a firearm. § 922(g)(3). Or petitioners and similar businesses may find themselves on the wrong side of a civil suit under the Racketeer Influenced and Corrupt Organizations Act. See, e.g., Safe Streets Alliance v. Hickenlooper, 859 F.3d 865, 876–877 (CA10 2017) (permitting such a suit to proceed).
I could go on.
Id. at 2238.
In short, closing the gap promotes the rule of law and the interests of all of us who depend on it for basic health, safety, and welfare.
B. Rescheduling Would Have Positive—Not Cataclysmal—Consequences.
Many argue that descheduling isn’t just the optimal outcome, but the only one that wouldn’t bring immediate devastation to the cannabis industry and everything reform advocates hold dear. Proponents of this view often claim that rescheduling would mean subjecting cannabis to “the prescription drug model.” They’re wrong. As I’ll explain, rescheduling cannabis to any of the CSA’s other four schedules would be a historic event with concrete benefits for cannabis stakeholders, regulators, and the public at large. Arguments to the contrary rest on fallacious reasoning and assumptions that can’t withstand the slightest scrutiny.
To assess whether and to what extent a transfer to another of the CSA’s schedules would be a good or bad thing, you must first understand the basic requirements associated with each. The chart below covers the territory.
As you read through the following discussion of the various benefits that would flow from transferring cannabis to the CSA’s other four schedules, you’ll notice that as thorough as the chart above is, it doesn’t capture everything. Where a relevant detail is not reflected in the chart, I’ve tried to provide other authority so you can verify my claims. I’ll start with the benefits of rescheduling cannabis to schedule II before discussing the benefits of schedules III, IV, and V.
Benefits of rescheduling to schedule II. There are two big differences between schedules I and II (a point Matt Zorn fleshed out in fascinating detail in his killer On Drugs post on PCP). First, doctors may prescribe schedule II drugs under certain circumstances—something they can’t do with schedule I substances. Second, and this is the point I’ll focus on here: those seeking to research schedule I substances face significantly more red-tape than those seeking to research schedule II substances. In testimony before the House Subcommittee on Health, Committee on Energy and Commerce, Dr. Nora Volkow, the Director of the National Institute on Drug Abuse (NIDA), recently described the extraordinary challenges researchers face when seeking to study schedule I substances, contrasting them with the dramatically more relaxed regulations that apply to substances in schedules II-V:
Even experienced researchers have reported that obtaining a new Schedule I registration [from DEA under 21 U.S.C. 823(f)], adding new substances to an existing registration, or getting approval for research protocol changes is time-consuming. Unlike for Schedule II through V substances, new and amended Schedule I applications are referred by the DEA to the HHS for a review of the protocol and a determination of the qualifications and competency of the investigator. This review is often in addition to other reviews of the proposed research and investigator, such as the federal grant review process, the FDA Investigational New Drug (IND) application review process, and Institutional Review Board and Institutional Animal Care and Use Committee reviews. Establishing the security infrastructure needed to conduct Schedule I research can be expensive and may need to be duplicated for each registrant working within a single research department. Researchers have also reported that there is a lack of clarity in some of the registration requirements and variability in their interpretation, which complicates and adds time to the process. For example, researchers report inconsistency in the guidance they have received on whether one individual can work under the registration of another, whether separate registrations are needed for each of an investigator’s research sites within the same campus, whether a manufacturing registration is needed to create final dosage formulations for research purposes, among other issues. Researchers have reported that sometimes these challenges impact Schedule I research and deter or prevent scientists from pursuing this critical work.
A paper published by the National Library of Medicine in 2017, examined the problem of schedule I research barriers in the cannabis context specifically. After describing “[t]he substantial layers of bureaucracy that emerge from cannabis’s Schedule I categorization,” the authors explain that cannabis’s schedule I classification has stifled “research on the health effects of cannabis and cannabinoids . . . in the United States, leaving patients, health care professionals, and policymakers without the evidence they need to make sound decisions regarding the use of cannabis and cannabinoids.” Their conclusion is ominous (and obviously right): “This lack of evidence-based information on the health effects of cannabis and cannabinoids poses a public health risk.”
If you ask me, blocking science that “patients, health care professionals, and policymakers” urgently need to “make sound decisions” on issues so central to public health and safety isn’t just bad policy—it’s immoral. That might be an exaggeration if DEA could hide behind claimed fealty to the CSA’s “closed system of drug regulation.” Here, though, “the Federal Government’s current approach to marijuana bears little resemblance to the watertight nationwide prohibition that a closely divided Court found necessary to justify the Government’s blanket prohibition in Raich.” Standing Akimbo, 141 S. Ct. at 2238. So DEA doesn’t even have that fig leaf to hide behind.
But wait. It gets worse. In his October 6th announcement, President Biden noted the patent absurdity of keeping cannabis in schedule I while fentanyl—the nightmare synthetic opioid fueling the overdose epidemic that is mowing down over 100,000 Americans a year right now—continues to wreak unimaginable havoc from schedule II. He’s right about that. As I’ve explained before, however, the fentanyl-cannabis comparison points up an even deeper problem: that cannabis’s unlawful schedule I classification may be exacerbating the opioid and overdose epidemic:
The Administration’s defenders say that addressing the overdose and addiction epidemics is simply a higher priority at the moment. With hundreds of thousands of Americans dying of overdoses each year and millions more addicted to powerful opioids, they argue, you’ll just have to excuse the President for not having gotten around to cannabis and psychedelics reform.
That argument may seem compelling, but only if you assume that cannabis and/or psychedelics aren’t part of the solution. Trouble is, there is powerful evidence that they are.
Don’t believe me? Fair enough. Perhaps you’ll believe NIDA Director, Dr. Nora Volkow, and former NIH Director and recently named “Science Advisor to the President,” Dr. Francis S. Collins. In a 2017 article published in one of the most prestigious peer-reviewed medical journals on the planet, they recognized that “[t]here is strong evidence of the efficacy of cannabinoids, including tetrahydrocannabinol (THC), in treating pain” and concluded that “[m]edications that target the endocannabinoid system … could provide a powerful new tool” for “chronic pain management.” And studies, including clinical trials, investigating the use of psychedelics for opioid use disorder are already underway.
Yet the Wall Street Journal recently quoted one lawmaker who opposes cannabis reform because, in his words, “[w]e have enough problems with gateway drugs the way that it is right now.” While he’s unquestionably right that America has a problem with gateway drugs, it seems he is laboring under the long-ago-discredited view that cannabis is a gateway drug. In fact, however, as Drs. Collins and Volkow recognized in 2017, America’s gateway drugs are pharmaceutical opiods. With millions of Americans struggling with opioid use disorder and thousands overdosing on fentanyl every year, that much should be obvious to anyone paying attention. Their more significant insight, however, is that there is powerful evidence that cannabis and (I would submit) psychedelics can be America’s “exit drugs.”
Put simply, the seriousness of America’s overdose and addiction epidemic makes the need for cannabis and psychedelics reform more—not less—urgent.
Rescheduling cannabis to schedule II would end this DEA-imposed dark age on cannabis science once and for all. That’s unquestionably a good thing. So while I acknowledge that descheduling would be even better, I remain convinced that rescheduling to schedule II would qualify as a historic victory for rational federal drug policy reform.
Before I move on, it’s worth mentioning that schedule II would have a few other benefits, including:
theoretically opening the door to federally legal cannabis prescriptions and likely making some physicians less reticent to consider cannabis’s therapeutic potential, see 21 U.S.C. 829(a);
theoretically opening the door to access to cannabis medicines under the federal Right to Try law, see 21 U.S.C. 360bbb-0a;
resolving the 280E tax problem for companies whose business would no longer qualify as “trafficking” in a schedule I or II substance under state or federal law, see 21 U.S.C. 280E;
potentially influencing lawmakers in jurisdictions that have maintained especially restrictive cannabis laws to consider loosening them to some extent;
encouraging regulators and researchers to entertain the idea of occasionally thinking outside the “molecule/compound” model of drug development that currently dominates drug development and policy in the U.S.; and
taking an incremental but significant step toward normalizing cannabis under federal law.
Next, I’ll quickly highlight some of the key benefits that would attend a transfer to schedules III, IV, and V. Then I’ll address the naysayers who view rescheduling as a doomsday scenario to be avoided at all costs.
Rescheduling to schedule III. Transferring cannabis to schedule III would have all the benefits of schedule II plus:
lighter criminal penalties for cannabis-related violations of the CSA;
eliminating the 280E tax problem altogether;
relaxing various other regulatory burdens like quota requirements, security regulations, etc. (see the Vodra chart above for more detail); and
do a bit more to normalize cannabis along the lines mentioned in the bullet-point list of additional benefits of schedule II.
Rescheduling to schedule IV. Schedule IV would have all the benefits of schedules II and III plus slightly lighter criminal penalties for trafficking, relaxing other technical compliance obligations slightly (see chart), and doing a bit more still to normalize cannabis along the lines already discussed.
Rescheduling to schedule V. Schedule V would have all the benefits of schedules II, III, and IV plus permitting cannabis to be distributed and dispensed without a prescription under certain circumstances, slightly lighter criminal sanctions for trafficking, relaxing other compliance obligations slightly (see chart), and doing a bit more still to normalize cannabis along the lines already discussed.
In my view, solving the 280E tax problem and finally removing the ridiculous barriers to cannabis research are huge, and the symbolic and historical significance of removing cannabis from schedule I should not be underestimated. Thus, while I acknowledge that descheduling is far and away the best outcome possible, I’m genuinely mystified that so many smart commentators continue to insist that we should prefer the schedule I status quo to any rescheduling scenario. I’ve studied their arguments carefully. I’ve even argued the point at length with several of them. As I explain next, having heard them out with an open mind, I remain thoroughly unpersuaded.
C. The Various Arguments That Rescheduling Would Be Worse Than Schedule I Are Nonsensical.
Why do so many people claim that keeping cannabis locked away in schedule I—the most restrictive and harshly criminalized federal classification possible—is somehow preferable to moving cannabis to a less-restrictive and less-harshly-criminalized schedule? I’ll let them explain it to you in their own words.
One cannabis attorney says that rescheduling would result in FDA regulation of cannabis and state cannabis laws being struck down under the Dormant Commerce Clause:
[R]escheduling is not necessarily good news for the industry, either. Rescheduling could result in FDA regulation of products that would all but require cannabis companies to operate like pharmaceutical companies.
If cannabis is descheduled or rescheduled, existing state regulatory schemes that feature or include local protectionist measures would likely fall as substantial burden on interstate commerce.
Another attorney argues that moving cannabis to schedule II would mean regulating cannabis under a pharmaceutical model that would doom the existing cannabis industry:
If President Biden’s proposal to move marijuana to Schedule 2 were implemented by his Administration, then the existing group of state-licensed, state-regulated cannabis businesses would continue to remain federally illegal, and only federally-regulated pharmaceutical companies could produce and sell marijuana, only in the form of FDA-approved drugs, and only to consumers who had a medical prescription for the drugs from a licensed physician. This could drastically change the current state of the US cannabis industry and would put many existing cannabis companies out of business if the federal laws were enforced.
Another story quotes a “cannabis law expert and former chair of the State Bar of Michigan’s Cannabis Law Section,” who argues that rescheduling to schedule II would spell the end for state-level adult-use frameworks:
“If we leave marijuana on the controlled substances list, we could never get over the hurdle of allowing the recreational use of marijuana in the United States,” Hendricks said. “Changing the scheduling of marijuana doesn’t change the framework for adult use. All of these states that have taken action, like Michigan, to create a framework for the regulation of marijuana, all of those probably go out of the window [if marijuana is classified as Schedule 2].”
Others, whose identities I will not disclose (since they haven’t made these statements publicly as far as I’m aware), have made similar arguments to me personally. One friend recently argued, for instance, that “rescheduling cannabis would likely create a prescription drug model, setting the industry on a path to mandated clinical trials and FDA premarket review.” That, they claimed, “would likely cause more problems than it would solve for state-legal businesses,” and it would “devastat[e] … smaller businesses and [the efforts of] criminal justice reform advocates.”
The case against rescheduling thus boils down to three basic arguments:
Rescheduling would subject the cannabis industry to FDA regulations designed for the pharmaceutical industry. Complying with these regulations would be impossible for most cannabis companies and would quickly drive the rest out of business.
In the wake of rescheduling, courts would declare the various state laws designed for (and with the help of) the cannabis industry preempted or unconstitutional under the Dormant Commerce Clause. Chaos and upheaval would ensue as cannabis companies attempted to adjust to a burdensome and unfamiliar legal framework on the fly.
Rescheduling would only legitimize FDA-approved medicines dispensed by doctor's prescription. If adult-use cannabis remains criminally prohibited at the federal level post-rescheduling, it will remain subject to a categorical federal criminal ban forever.
From here on, I’ll refer to these arguments (and every variant of them I’ve come across so far) collectively as the “Pro-Cannabis Case for Schedule I.” I’ll explain why they’re woefully misguided shortly. Before I do, though, I want to be very clear about what I’m not saying. I’m not saying that threats of crippling FDA regulation and enforcement, the Dormant Commerce Clause, and preemption—or, as I like to call, “The Three Horsemen the Legalization Apocalypse”—aren’t real threats to the cannabis industry. They are. What I am saying is that (1) cannabis’s schedule I classification isn’t doing a thing to protect anybody from them, and (2) descheduling wouldn’t either (or at least not much).
Let me explain. Proponents of the Pro-Cannabis Case for Schedule I do not—and cannot—deny that rescheduling has all the tangible benefits I went over above. Nor can they deny that some of those benefits—like reducing criminal penalties for cannabis-related crimes and removing obstacles to urgently needed cannabis research—would do a lot of good in terms of promoting social justice and public health and safety. Yet they fear the Three Horsemen of the Legalization Apocalypse so much that they insist that those benefits aren’t even worth pursuing if achieving them means transferring cannabis from schedule I to schedule II, III, IV, or V.
That line of reasoning assumes that cannabis’s schedule I status is somehow protecting the cannabis industry from the Three Horsemen’s wrath. Otherwise, why the trembling fear that rescheduling would unleash their fury, right? Thing is, though, FDA’s jurisdiction over a substance doesn’t end at schedule I. In fact, it doesn’t depend on scheduling status at all. It turns entirely on whether the substance (or product or whatever) qualifies as a “drug,” “medical device,” “food,” "dietary supplement,” “cosmetic,” “tobacco product,” etc. under the surprisingly broad definitions of those terms under the Federal Food Drug and Cosmetic Act (“FDCA”) and FDA’s regulations implementing that statute.
The same goes for the other preemption and the Dormant Commerce Clause, which emanate from the Supremacy Clause and the Interstate Commerce Clause of the U.S. Constitution, respectively. Let me assure you, those constitutional doctrines—indeed, every constitutional doctrine—apply to cannabis today while it’s in schedule I and will continue to apply to cannabis whether it’s rescheduled, descheduled, excommunicated, or canonized.
Naturally, this raises the question: If cannabis’s schedule I status isn’t keeping the Three Horsemen at bay, then what is? Many things. First, both Congress and the Executive Branch appear to be pushing for less—not more—cannabis enforcement at the federal level over time as a general matter. The President’s October 6th announcement, FDA’s passivity toward CBD even post-descheduling, and Congress’s passage of cannabis spending riders and ongoing focus on cannabis legalization proposals are just some of the proof in that pudding.
Second, federal enforcement consumes a lot of agency time and money. That money has to come from somewhere. In the FDA context, it’s pharma user fees and appropriations (your tax dollars). In other contexts, Congress offers the states access to federal funds conditioned on their cooperation in federal programs. In other words, the states often do the federal government’s dirty work (enforcement) in exchange for the money they need to meet their own responsibilities to the public they serve.
None of that is in place here, though. Appropriations require legislation. Conditional spending does, too. And good luck convincing the states that have legalized (i.e. the ones whose cooperation the federal government needs) to go along with a federal program that would threaten to end the flow of all those state-level cannabis tax dollars! So even if FDA had the inclination to ramp up enforcement in the cannabis space, the only way it could get the resources to do so would be by stealing from itself—i.e. taking dollars away from another agency project/priority. And make no mistake, FDA would have to rob itself blind to fund uniform enforcement against a $13 billion nationwide industry that it has barely engaged with for the past 50 years or so.
Nor is it clear that the prospect of shouldering the new and nebulous responsibility of aggressive cannabis enforcement would be an especially thrilling one for FDA. It sure hasn’t been in any particular hurry to ramp up enforcement in the CBD space since the 2018 Farm Bill descheduled hemp, for example.
Part II: The Path Forward
So far, I’ve demonstrated that (1) descheduling is far and away the best possible outcome we could hope for from the impending administrative process and that (2) rescheduling would also have benefits that are definitely worth pursuing. So what are the barriers and obstacles to descheduling, and how do we overcome them? I answer those questions next.
A. Overview of Formal “on the Record” Rulemaking Process
In part 2 of this series, I went over the formal rulemaking process that, under 21 U.S.C. 811(a)-(c), governs scheduling, descheduling, and rescheduling actions before DEA except when control of the substance is required by one of the treaties identified in 21 U.S.C. 811(d)(1). For the moment, I’m going to assume that DEA will use the formal rulemaking process outlined in section 811(a)-(c). Here’s a quick review of how that process works to refresh your memory:
DEA gathers the necessary data and requests a scientific and medical evaluation and scheduling recommendation from FDA.
FDA’s Controlled Substances Staff (“CSS”) prepares the evaluation and recommendation
CSS’s draft goes to the Office of the Director of FDA’s Center for Drug Evaluation Research (“CDER”) for review.
Once CDER’s review is complete, the draft goes to FDA’s Office of the Commissioner.
Once the Commissioner’s review is complete, the draft goes to the Office of the Assistant Secretary for Health (“OASH”) at HHS.
Once the OASH signs off on the evaluation and recommendation, they’re forwarded along to DEA.
DEA then reviews the relevant evidence and HHS’s recommendation and evaluation and assesses whether there is “substantial evidence” that the substance should be scheduled, rescheduled, or descheduled. 21 U.S.C. 811(b). If there is, DEA “shall initiate proceedings,” accordingly. Id. Otherwise, DEA will enter a final order denying the scheduling request.
If DEA denies the scheduling request at step 8, it publishes its final order in the Federal Register and places its and HHS’s analyses under 811(b) in a public rulemaking docket. See, e.g., this example.
If DEA concludes there is substantial evidence to initiate proceedings to schedule, reschedule, or deschedule the substance at step 8, it drafts a notice of proposed rulemaking (“NPRM”) and publishes it in the Federal Register. The NPRM notifies the public that DEA will accept comments on the proposed rule for a 30-day period. It also explains that “interested persons” have 30 days to file a request for a hearing or waiver of participation under 21 CFR 1308.44 and in accordance with 21 CFR 1316.45, 1316.47, 1316.48, or 1316.49.
After the public comment period and any hearings requested at step 10, DEA considers all comments, HHS’s scientific and medical evaluation and accompanying recommendation, DEA’s own analysis, and the record of any hearing compiled by the ALJ, and issues a final rule explaining the agency’s findings that the substance does or does not meet the CSA’s requirements for inclusion in any schedule. If it does, DEA promulgates a rule to that effect (unless one already exists). If it does not, DEA promulgates a rule removing the substance from the schedules as necessary.
Once DEA issues a final rule, any interested person aggrieved by DEA’s decision has 30 days to seek judicial review. See 21 U.S.C. 877.
This process can take a very long time—9 years on average. See Washington v. Barr, 925 F.3d 109, 120 (2d Cir. 2019). Even relatively simple proceedings take almost that long. Consider Samidorphan. Until last year, Samidorphan was a schedule II controlled substance. See Schedules of Controlled Substances: Removal of Samidorphan From Control, 86 Fed. Reg. 20284-01 (Apr. 19, 2021). On April 14, 2014, however, DEA received a petition to deschedule the drug. Id. DEA eventually issued a final rule doing just that, but it took it over seven years to do so even though (1) it received only two comments and no requests for hearing in response to the NPRM (at step 9 above). I’ve broken out how long the various steps in the process took in Samidorphan’s case:
April 14, 2014: DEA receives petition to reschedule.
April 24, 2015: DEA forwards data to HHS and requests evaluation and recommendation.
January 9, 2020: DEA receives HHS’s evaluation and recommendation. In this case, HHS recommended that DEA “not control” the substance. See 21 U.S.C. 811(b).
December 20, 2020: Even though HHS’s recommendation of “no control” was binding on DEA under section 811(b), DEA initiated proceedings and published an NPRM in the Federal Register soliciting comments and hearing requests/waivers. See NPRM, 85 Fed. Reg. 79450-01 (Dec. 20, 2020).
April 19, 2021: After receiving no hearing requests and only two comments in response to the NPRM (both of which supported DEA’s proposed rule by the way), DEA issued a final rule explaining the basis for its conclusion that Samidorphan doesn’t meet the requirements for inclusion on any of the CSA’s schedules. 86 Fed. Reg. 20284-01 (Apr. 19, 2021).
As you can see, even in a relatively simple case like Samidorphan’s, the formal rulemaking process we’re about to witness is arduous and very, very slow. In a complex case—one that involves multiple hearings, thousands of public comments, and one or more hotly contested issues of first impression, for example—the delay is obscene. It took DEA 22 years to issue a final decision denying NORML’s 1972 petition to reschedule cannabis, for example.
I’m fully aware of the fact that the President has urged DEA and FDA to proceed through this process “expeditiously,” and I'm glad he did. But that's kind of like asking xenon-124, a radioactive noble gas with a half-life a trillion times longer than the age of the universe, to finish up ASAP. I'm sure DEA and FDA will do their best, but I'm expecting the process to take years, not months. And once DEA issues a final decision, I'm sure an interested person (more likely several) will seek judicial review. So, I'd settle in and try to get comfortable if I were you. We've got a long, long way to go.
B. The Only Path to Descheduling
(i) The Only Path to Descheduling
As I see it, descheduling is out of the question unless HHS recommends “no control” at step 6 above. Here’s why.
Start with the fact that HHS’s scheduling recommendation is binding on DEA if it recommends “no control.” See 21 U.S.C. 811(b). Otherwise (meaning if HHS recommends schedule I, II, III, IV, or V), DEA can overrule HHS’s ultimate recommendation and place cannabis in the schedule it thinks best as long as it accepts HHS’s determinations of “scientific and medical matters.” See id. To recap:
If HHS recommends descheduling at step 6 above, DEA has no choice under the statute but to deschedule. Id.
If HHS recommends anything other than descheduling, DEA can do what it wants as long as it accepts HHS’s view of scientific and medical issues in the process. Id.
Look, I’ve scrutinized literally everything DEA has ever said about cannabis publicly. I’ve also FOIA’d records that weren’t publicly available and pondered them into bleary-eyed oblivion. I’ve even fought the agency on these issues at the administrative level and in federal courts across the country over and over and over again. After all that, I feel like I’m in about as good a position as anybody other than Anne Milgram to make an educated guess regarding the likelihood of DEA overruling an HHS recommendation that cannabis be placed in schedule I, II, III, IV, or V, and deschedule cannabis anyway. And I’d say the chances are very, very low.
But let’s say I’m wrong about that. Let’s assume for purposes of argument that HHS recommends schedule II, and Administrator Milgram directs her staff at DEA to deschedule anyway. Were that to happen, the staff would have their work cut out for them. Why? Because DEA’s staff would then have the unenviable task of drafting a final decision explaining why the agency was descheduling cannabis in the face of an HHS recommendation to the contrary.
To withstand judicial review under the APA, that decision would have to
acknowledge that for the last fifty years, it has insisted that cannabis must remain in schedule I;
acknowledge that federal courts have repeatedly blessed its reasoning and conclusions for doing so;
provide a reasoned explanation for changing course now that addresses the mountain of uniformly contrary evidence and precedent it has painstakingly collected and curated over the last half century; and
explain why HHS’s recommendation to the contrary is also wrong.
Oh, and one more thing: In crafting its explanation for why HHS’s recommendation is wrong, DEA would not be permitted to disagree with HHS about a single thing having to do with science or medicine. That means that in crafting DEA’s rationale for unilaterally doing a 180 on perhaps the most entrenched policy in the agency’s history, the poor agency staffer holding the pen would have to focus exclusively on law-enforcement issues (abuse and diversion considerations).
For all these reasons, I’m convinced that descheduling will effectively be out of the question here unless HHS recommends “no control” under section 811(b). That means that anybody and everybody who agrees with me that descheduling is the optimal outcome here should be doing everything conceivably possible to convince HHS (really CSS, CDER, the FDA Commissioner, and the OASH) to deschedule. And because HHS is required to send its recommendation to DEA at an early stage of the process—and before public comments or requests for hearing are even permitted—we should have started our “no control” lobbying effort weeks ago.
Still, better late than never right? So let’s proceed to the next issue: What will CSS, CDER, the FDA Commissioner, and the OASH need to craft a descheduling recommendation that is as safe from judicial review as possible?
(ii) Helping HHS Chalk a Win for the Biden Administration
Given that President Biden got this crazy ball rolling for the express purpose of ensuring that nobody goes to jail for cannabis use or possession, HHS should be willing to take proposed rationales for descheduling—the only result that is consistent with the President’s goal—seriously. The question is what’s the simplest and easiest way for HHS to recommend descheduling without creating a serious risk of being overturned on judicial review? Let’s break it down.
To understand what HHS would need to say to justify descheduling, you need to go back to the text of the statute. Start with section 812(b)—the CSA provision that requires various findings to place a substance in one schedule or another. If you study that provision, which I covered in detail in part 2 of this series, you’ll notice that a substance with no abuse potential or a lower abuse potential than substances in schedule V cannot be scheduled. DEA and HHS have acknowledged this reality. See 85 Fed. Reg. 79450-01 (Dec. 20, 2020) (“[T]he finding that samidorphan lacks abuse potential would, irrespective of other findings, permit decontrol of samidorphan prior to or in the absence of an FDA action under 21 U.S.C. 355(c).”). Because that single finding would be enough to justify descheduling, I think it would likely be the easiest and simplest basis for an HHS recommendation to deschedule.
As I also explained in part 2, however, if HHS applied its and DEA’s longstanding standard for assessing relative abuse potential, it would have little choice but to reach the same conclusion it always has: that cannabis has a high potential for abuse. drawing that conclusion should be the focus. That means that the most straightforward path to victory here would be to convince HHS to adopt a new interpretation of “potential for abuse” and then provide the evidence necessary under the 8 factors listed in section 811(c) to support a finding that cannabis has no abuse potential or at least that its abuse potential is lower than the substances listed in schedule V. I went over what I think are decent arguments in support of that effort in part 2, so I won’t repeat them here. Instead, I’ll move on to the other barriers to scheduling.
C. Other Barriers
Let’s take stock. So far I’ve explained that
the only realistic path to descheduling is convincing HHS to recommend “no control” under section 811(b) and
the easiest and simplest path for HHS to recommend no control would be to change its longstanding definition of “potential for abuse.”
There are, however, many additional barriers to descheduling that we must be prepared to navigate. I address the main ones briefly below.
The APA. As I’ve emphasized throughout this series, when an agency changes a long-settled position, courts demand that the agency “display awareness that it is changing position” and “show that there are good reasons for the new policy.” Encino Motorcars, LLC v. Navarro, 579 U.S. 211, 221–22 (2016) (quotes and cites omitted). “[A] reasoned explanation is needed for disregarding facts and circumstances that underlay or were engendered by the prior policy.” Id. (cites and quotes omitted). If agencies fail to provide this sort of self-conscious rationale for a change in position, the new policy will not hold up in court. Id.
HHS is fully aware of this bedrock principle of administrative law, and anybody thinking of urging the agency to change a longstanding policy will need to take great care to assure the agency they’re aware of it, too. Unfortunately, most commentators I’ve come across seem to ignore HHS/FDA’s prior views because they (the commentators) believe them to be “obviously wrong.” If given an audience with the relevant decision-makers at CSS, CDER, the FDA Commissioner’s Office, and/or the OASH, these folks would apparently urge the agency decision-makers to abandon their long-settled policies out of hand are making a grave mistake.
Because FDA/HHS can’t ignore the previous evidence and rationale for keeping cannabis in schedule I, we can’t either. Indeed, the previous evidence and rationale are perhaps the primary hurdle to changing federal policy through this administrative process. The only way HHS is recommending descheduling is if it can support its decision with a mountain of evidence and a thorough and tightly-reasoned explanation that engages with the agency’s past precedent in painstaking detail. I simply cannot emphasize this enough.
The CSA. In describing the path to descheduling here, I’ve focused primarily on the “potential for abuse” standard. As I’ve explained, if HHS is willing to adopt a new test for assessing substances’ relative abuse potential, then it might have a relatively simple path to conclude that cannabis has no abuse potential or at least an abuse potential lower than most substances in schedule V. That would be a substantial departure from past agency precedent, but there’s no way to deschedule cannabis without abandoning one or more longstanding agency policies. Because a finding of no or very low abuse potential is enough to dictate no control under the CSA, I’ve recommended focusing on that rifle-shot path as the core of the plan to convince HHS to reschedule.
While the abuse potential standard is the centerpiece of the plan, for HHS’s recommendation of “no control” to succeed, the agency will also need to fulfill its various other obligations in the process, including
providing an 8-factor evaluation of cannabis as section 811(b)-(c) requires (I discuss the 8-factor analysis more in part 2 of this series);
framing its entire recommendation, including its change of position on the proper standard for “potential for abuse,” around the 8 factors identified in section 811(c);
considering and addressing how its new abuse potential standard would apply in other scheduling contexts with other substances (the agency will, for example, hesitate to adopt any standard that would throw the entire scheduling regime into chaos or open the floodgates of litigation and administrative challenges to every scheduling decision it and DEA have ever made);
The THC Problem. I’ve waited until almost the very end of this series to mention it, but there is a major problem with the President’s directive that no one seems to have noticed. He has requested a review of cannabis’s scheduling status, but cannabis isn’t a controlled substance. I suppose everybody just assumes the President was referring to the schedule I substance “marihuana.” See 21 U.S.C. 812 schedule I(c)(10). Thing is, though, if HHS deschedules marihuana but leaves “Tetrahydrocannabinols” listed in schedule I without explanation, confusion and chaos could easily ensue. I could imagine cannabis’s opponents responding to a final decision descheduling cannabis by insisting that because all cannabis contains schedule I Tetrahydrocannabinols, all cannabis (except, presumably, “hemp,” see 7 U.S.C. 1639o) remains subject to schedule I restrictions for all practical purposes.
You may think this is a silly argument, but trust me when I tell you that doesn’t mean opponents of legalization won’t be able to use it to undermine what would otherwise be a historic victory for cannabis reform. So what to do? I would suggest that HHS go out of its way to address the issue in its recommendation. Specifically, HHS could simply point out that federal appellate courts have already considered and rejected this argument—repeatedly. Anticipating this precise problem, they have held that the “Tetrahydrocannabinols” listed in schedule I are only synthetic Tetrahydrocannabinols. See, e.g., HIA v. DEA, 357 F.3d 1012, 1013 (9th Cir. 2004) (“We have previously held that the definition of “THC” in Schedule I refers only to synthetic THC ….”) (internal cites and quotes omitted). As one Ninth Circuit panel explained, interpreting “Tetrahydrocannabinols” to include synthetic and natural THC would violate basic principles of statutory construction by “render[ing] superfluous the separate listing of marijuana and would nullify the explicit exemption of hemp seed and oil from the coverage of marijuana.” See HIA v. DEA, 333 F.3d 1082, 1090 (9th Cir. 2003).
Treaty Issues. HHS doesn’t really need to worry about the treaties because the CSA delegates that responsibility entirely to the Attorney General (who has delegated it to DEA). See 21 U.S.C. 811(d)(1). That said, advocates of descheduling shouldn’t ignore them. The fact of the matter is that descheduling cannabis would make effectively legalize recreational cannabis use at the federal level and thus put the U.S. out of compliance with the Single Convention on Narcotic Drugs of 1961. This could cause problems in several ways that advocates of descheduling need to be prepared to address:
HHS may hesitate to recommend descheduling when doing so would mean putting the U.S. in manifest non-compliance with a treaty—especially where the statute expressly directs the federal agency with primary implementing authority over the CSA generally and the scheduling process in particular to prioritize treaty compliance above all else in making scheduling decisions. Id.
This scheduling action will almost certainly end up in the interagency review and clearance process that applies to significant agency actions under the Congressional Review Act. That means the State Department will very likely weigh in on any proposed change in cannabis’s scheduling before this thing becomes final. The State Department is, as you might expect, touchy about treaty compliance. Moreover, it is currently having to deal with the mess U.S. non-compliance has caused in the context of ongoing tensions with U.S. treaty partners and signatories to the Single Convention, Russia and China, over the Britney Griner affair and China’s role in synthetic fentanyl trafficking into the U.S. Long story short, I just don’t see how any proposal to change cannabis’s scheduling classification gets finalized without the State Department raising treaty concerns.
- Finally, DEA or cannabis opponents outside the federal government could conceivably use the treaty problem to hijack the process at various stages or get it overturned on judicial review.
Descheduling advocates likely won’t be permitted to simply ignore the treaty problem. Here’s where I would start in formulating a response:
Given that descheduling is the only outcome consistent with the President’s October 6th announcement that he intended to “end” the “failed approach” to cannabis that sent people to prison for cannabis use and possession, agency decision-makers like HHS can’t be held responsible for treaty ramifications here.
And because the President all but overtly embraced descheduling on October 6th, and he speaks for the entire United States, any damage the prospect of descheduling might do in other diplomatic contexts is already done, and nothing HHS or DEA does going forward is going to change that.
Given that (1) the U.S. has been out of compliance with the treaty continuously for over fifty years, as OLC recently acknowledged in a now-public opinion, and (2) the organization responsible for enforcing the treaty’s requirements has repeatedly chastized the U.S. for its non-compliance, keeping cannabis in schedule I wouldn’t make any difference. The U.S. is going to be out of compliance regardless. The prudent thing to do therefore is to acknowledge the reality of U.S. non-compliance, explain that public health and safety require that the U.S. change its cannabis policy, and commit to working with our treaty partners around the world to modernize the international drug control treaties to align with post-War-on-Drugs policies and understandings.
Treaty concerns didn’t stop Congress and DEA from descheduling epidiolex, which DEA had previously acknowledged would violate the treaty.
If all that fails, I’d recommend that HHS simply cite the provision of the Single Convention that says signatory states need not comply with the treaty where doing so is incompatible with their constitutional framework, see Single Convention Art. 23(2), and me and attorney Matt Zorn's forthcoming law review article, demonstrating that section 811(d)(1) violates the private non-delegation doctrine. See Pennington and Zorn, The Controlled Substances Act: An International Private Delegation That Goes Too Far, 100 Wash. U. L. Rev. (forthcoming 2023).
I’ve done my best to convince you that we can’t sit on the sidelines. If we do, a sub-optimal result is virtually guaranteed. So, if you want rational federal cannabis policy reform, it’s time to get in the game. Somebody needs to use the blueprint I’ve provided (or an even better one) to develop the evidence and reasoning HHS will need to deschedule, put it in a digestible form, and get it in the hands of the right people to communicate it to agency decision-makers.
The rest of us need to be ready to jump into the notice and comment process that will eventually happen. And anyone who is an “interested person” should start gearing up for that ALJ hearing now. I know some decent lawyers who might be able to assist. Finally, while I’ve focused primarily on descheduling here, I’ve also argued that rescheduling would have historically significant benefits that we shouldn’t ignore. Any strategy to influence the process should therefore have contingency plans in place to secure the best possible “second-best” result in the event descheduling turns out to be off the table.
Did you miss part 1 on the topic? Read it here.
Did you miss part 2 on the topic? Read it here.
Originally posted on On Drugs, which offers "Cutting Edge Thoughts on Drug Regulation by Matt Zorn and Shane Pennington." Subscribe here.