RN Collins has written a series of 20 new articles for cannabis law report on 2026 Psychedelics & Legal Issues.

This is the 12 th in a series of 20

Ballot Initiative Governance Risks in Psychedelic Legalization

Cannabis Law Report | Legislative Brief Series

Abstract

The ballot initiative process has served as the primary vehicle for psychedelic legalization in the United States, producing landmark voter-approved measures in Oregon (2020), Colorado (2022), and elsewhere. Yet direct democracy, as a legislative mechanism for complex drug policy reform, presents structural governance risks that differ fundamentally from risks arising in the legislative process. This brief examines those risks — among them, regulatory capture through initiative drafting, local opt-out fragmentation, implementation gaps created by statutory rigidity, legislative rollback dynamics, and voter fatigue effects — drawing on the parallel experience of cannabis ballot initiatives and the early implementation record of psychedelic programs. It concludes with a framework for risk mitigation for future psychedelic initiative drafters and state legislatures charged with implementing voter-approved psychedelic programs.

Keywords: ballot initiative; direct democracy; psychedelic legalization; Oregon Measure 109; Colorado Proposition 122; regulatory governance; initiative rollback; local opt-out; cannabis policy; Measure 110

I. Introduction: The Ballot Box as Policy Vehicle

Since California voters approved Proposition 215 — the nation’s first medical marijuana initiative — in 1996, advocates across the drug policy reform movement have increasingly relied on citizen-initiated ballot measures to bypass legislatures unwilling to enact reform.¹ Of the twenty-four states that have legalized adult-use cannabis, fourteen did so via the ballot box.² Psychedelic reform has followed this same trajectory: Oregon voters approved Measure 109, creating the nation’s first regulated psilocybin services program, in November 2020, and Colorado voters approved Proposition 122, the Natural Medicine Health Act, in November 2022.³

The appeal of direct democracy for psychedelic advocates is straightforward: legislative chambers in most states remain reluctant to act on substances that carry significant federal Schedule I status and residual cultural stigma from the counterculture era. Ballot initiatives allow advocates to appeal directly to voters, bypassing committee chairs, floor leaders, and executive vetoes. The tactic has succeeded: Oregon’s Measure 109 passed with approximately 55.75% of the vote, and Colorado’s Proposition 122 passed with roughly 54%.⁴

But direct democracy carries structural risks that are absent or substantially reduced in the legislative process. These risks are not merely theoretical. The implementation records of Oregon Measure 109, Colorado Proposition 122, and the companion decriminalization measure Oregon Measure 110 — subsequently revised by the legislature — provide a rich empirical foundation for identifying governance pathologies unique to initiative-based drug policy. The November 2024 defeat of Massachusetts Question 4, a comprehensive psychedelic legalization measure, adds a further data point: even when initiatives are carefully designed, inadequate regulatory specificity and poorly targeted scope can sink voter support.⁵

This brief catalogs those risks in six categories: (1) regulatory capture through initiative drafting; (2) statutory rigidity and implementation gaps; (3) local opt-out fragmentation; (4) legislative rollback dynamics; (5) voter fatigue and the “second ballot” problem; and (6) federal preemption exposure. It concludes with a risk mitigation framework for initiative drafters, implementers, and legislatures.

II. Risk Category One: Regulatory Capture Through Initiative Drafting

A. The Structural Problem

Ballot initiatives, by design, encode policy decisions into statutory or constitutional text approved directly by voters. This structure creates a powerful incentive for the interest groups that finance initiative campaigns to embed favorable substantive rules — rules that will govern the resulting industry — directly into the initiative text, long before regulators, legislators, or public health authorities have had a meaningful opportunity to shape those rules.⁶

A 2021 analysis of cannabis legalization through state legislatures found that the ballot initiative process is a particularly vulnerable point for regulatory capture, because industry players can encode pro-profit, anti-public-health rules into law that will persist long after the election, often surviving subsequent legislative amendment attempts.⁷ The same analysis noted that non-governmental health organizations — which lack the political and legal flexibility to fund ballot campaigns — are structurally absent from most initiative campaigns, ceding the drafting field to industry-aligned funders.⁸

In the psychedelic context, this risk manifests somewhat differently than in cannabis, because the leading psychedelic initiative campaigns have been primarily funded by reform-oriented political action committees rather than established commercial operators. Massachusetts Question 4, for example, was funded and overseen by the Washington, D.C.-based New Approach PAC, which had previously backed both Measure 109 and Proposition 122.⁹ The PAC’s approach — consistent across all three measures — favored a facilitated-access, non-medical model that enabled broad market entry by lay facilitators rather than licensed clinicians.

B. The Oregon Model’s Structural Ambiguity

Oregon Measure 109 illustrates the governance consequences of initiative-layer ambiguity about program character. The measure’s chief petitioners — psychotherapists Tom and Sheri Eckert — promoted the initiative publicly as a vehicle for legalizing psilocybin-assisted therapy for mental health conditions. Yet the final initiative text explicitly prohibited the word “therapy” in the context of the program (OAR 333-333-1010(72)) and authorized only “supported adult use” by non-clinician facilitators, with no diagnosis requirement.¹⁰

Bioethicists and practitioners working in Oregon’s program have documented the resulting contradiction: the initiative text “has resulted in a policy that doesn’t allow for any of” the therapeutic use that was publicly promised during the campaign, while the Oregon Health Authority (OHA) and licensed service center operators continue to be publicly described in therapeutic terms.¹¹ The absence of a diagnosis requirement — a deliberate drafting choice designed to broaden the voter coalition — created immediate scope-of-practice confusion about whether facilitators are providing health services and, if so, who governs their practice. More than three years after program launch, fundamental questions about the relationship between OHA rules and professional licensing boards remain unresolved.¹²

C. Colorado’s Bifurcated Agency Structure

Colorado Proposition 122 created a different form of regulatory capture risk through its allocation of authority between two agencies. Under the Natural Medicine Health Act (NMHA), as amended by SB23-290, the Department of Regulatory Agencies (DORA) is responsible for licensing and regulating facilitators, while the Department of Revenue (DOR), through a new Division of Natural Medicine, is responsible for licensing and regulating healing centers and other natural medicine businesses.¹³ This bifurcation — encoded in the voter-approved initiative text — was retained in SB23-290 and cannot be altered without additional legislation.

The practical consequence is a divided regulatory infrastructure for what is functionally a unified service transaction: a facilitator working in a healing center simultaneously holds a DORA license and works within a DOR-licensed business. Rules governing facilitator conduct that conflict with healing center operations require inter-agency coordination that the initiative text did not structurally provide for. Colorado’s legislature addressed some coordination gaps in SB24-198, the Regulated Natural Medicine Implementation Act of 2024, but the fundamental two-agency structure persists.¹⁴

III. Risk Category Two: Statutory Rigidity and Implementation Gaps

A. The Amendment Problem

Ballot initiatives — particularly those enacted as statutory text rather than constitutional amendments — can ordinarily be amended by subsequent legislatures. However, in several states, initiatives are protected for a period following passage (typically two years) from legislative amendment without voter approval, or can only be amended by a supermajority. Colorado’s NMHA, enacted as a citizen-initiated constitutional amendment, carries stronger protection from legislative alteration than a typical statute, creating a more rigid framework in areas where post-implementation evidence has revealed gaps.¹⁵

The more pervasive rigidity problem arises not from amendment restrictions but from the initiative process itself. Initiative text is necessarily drafted without the benefit of agency expertise, stakeholder comment periods, or the legislative committee process. Key implementation details — which substances qualify, how facilitator training is structured, what safety protocols govern adverse events, how banking and tax liability interact with program participation — are often left to post-enactment rulemaking. When the initiative text constrains rulemaking authority or fails to adequately specify it, agencies face gaps they lack statutory authority to fill.

B. Oregon’s Implementation Gap Record

Oregon Measure 109 provides the most detailed record of initiative-generated implementation gaps. Among the most significant: the initiative required OHA to establish a framework, but the two-year rulemaking period (2021-2023) was insufficient to resolve questions about liability insurance, banking access, and integration of the program with existing professional licensing frameworks. As of program launch in 2023, facilitators lacked access to standard liability insurance — a market gap that persists because underwriters will not cover services that remain federally illegal — and service centers face ongoing banking restrictions under the same federal scheduling that applies to cannabis operators.¹⁶

The initiative’s residency requirement — which required majority ownership of psilocybin businesses by Oregon residents — was widely regarded as likely unconstitutional under the dormant Commerce Clause from the moment of enactment. Rather than prompting a judicial challenge, the requirement was left in place for its two-year term and then sunsetted as scheduled on January 1, 2025, without formal resolution of the constitutional question.¹⁷ The result was a two-year period of regulatory uncertainty for non-resident investors who nonetheless structured around the requirement using options and redemption features — a pattern that initiative drafters could have avoided by consulting Commerce Clause precedent.

The initiative also required OHA to establish a data collection and reporting framework, but failed to specify the data elements or reporting timeline with sufficient precision. Oregon’s legislature subsequently passed SB 303 (2023) to impose additional data collection requirements, illustrating how initiatives routinely generate a second wave of remedial legislation that drafters did not anticipate.¹⁸

C. Massachusetts Question 4: A Cautionary Drafting Case

Massachusetts Question 4, which voters rejected in November 2024 by a margin of approximately 56.9% to 43.1%, illustrates how implementation gap concerns can doom an initiative before it takes effect.¹⁹ The measure proposed decriminalizing five psychedelic compounds — psilocybin, psilocyn, dimethyltryptamine, mescaline, and ibogaine — for adults twenty-one and older, establishing supervised access at licensed facilities and creating a five-member commission modeled on the state’s Cannabis Control Commission to develop implementing regulations.²⁰

Two statutory provisions contributed heavily to the defeat. First, the measure authorized home cultivation of psychedelic plants and fungi in a twelve-by-twelve-foot area, a provision that campaign supporters for Massachusetts for Mental Health Options later acknowledged likely drove the measure’s rejection by voters who otherwise supported supervised therapeutic access.²¹ Second, the measure included ibogaine — a substance with known cardiac risks at therapeutic doses — in the same regulatory framework as psilocybin, without meaningful differentiation of safety protocols.²² The Coalition for Safe Communities, representing organizations comprising more than 60,000 physicians and psychiatrists nationwide, characterized the measure as “poorly written and misguided.”²³

The Massachusetts defeat reveals a core governance risk of the initiative format: binary yes/no votes cannot accommodate the nuance that complex regulatory frameworks require. Voters who supported supervised psilocybin access but opposed home cultivation and the ibogaine provision had no mechanism to vote for the parts they supported while rejecting the problematic elements. The all-or-nothing structure of direct democracy, combined with inadequate regulatory specificity, produced a result that satisfied neither reform advocates nor public health experts.

IV. Risk Category Three: Local Opt-Out Fragmentation

A. The Structural Incentive for Opt-Out Provisions

Both Oregon Measure 109 and Colorado Proposition 122 incorporated local opt-out or opt-in provisions, though they structured those provisions very differently. Oregon adopted an automatic opt-in model: all cities and counties were automatically included in the psilocybin program unless local voters elected to opt out through a referendum placed on a general election ballot. Colorado adopted a partial preemption model: local jurisdictions may regulate the “time, place and manner” of healing center operations but may not prohibit healing centers entirely, unlike the Oregon model where full prohibition by local vote was expressly authorized.²⁴

Sam Chapman, the Healing Advocacy Fund executive director who served as campaign manager for Measure 109, acknowledged that the opt-out provision was critical to building sufficient voter support for the statewide measure: without the option for localities to exit, rural and conservative communities that would have voted against the measure statewide might have swung the outcome.²⁵ The opt-out provision thus served an electoral function — broadening the coalition — at the cost of creating a structural incentive for post-enactment fragmentation.

B. The Oregon Fragmentation Outcome

The electoral compromise generated by the opt-out provision materialized precisely as anticipated. In the November 2022 general election — held two years after Measure 109’s passage and immediately before the program launch date — voters in 23 of Oregon’s 36 counties and 111 of its 241 cities elected to opt out of the psilocybin program either permanently or through a two-year moratorium.²⁶ As a result, by program launch in January 2023, psilocybin services were legally available in only a minority of Oregon’s geographic area, concentrated along the Interstate 5 corridor from Portland to Eugene, with limited availability in Deschutes and Jefferson counties.²⁷

The geographic consequences compound over time. In the November 2024 general election, seventeen additional Oregon jurisdictions placed opt-out measures on the ballot; sixteen of the seventeen passed, including several communities that had previously approved two-year moratoriums now voting for permanent prohibition.²⁸ Four cities — spanning Portland suburbs and rural coastal communities — approved new bans that were not previously in place. The net result is that as of early 2025, the counties where psilocybin services are permitted include Benton, Clatsop, Columbia, Deschutes, Hood River, Jackson, Lane, Lincoln, Multnomah, Wasco, Washington, and Yamhill — a geographic footprint that leaves large sections of eastern and southern Oregon, as well as rural western Oregon, without accessible services.²⁹

The access equity implications are significant. Patients in rural communities with PTSD, terminal illness anxiety, or treatment-resistant depression who might benefit from psilocybin services must travel, in some cases hundreds of miles, to reach a licensed service center — a burden that falls disproportionately on those with the fewest resources and the greatest mobility limitations, including disabled and homebound patients.³⁰

C. Colorado’s Preemption Contrast

Colorado’s structural choice to preempt local prohibition rather than authorize it produced a meaningfully different geographic distribution of healing center access. While localities retain authority to regulate time, place, and manner, no Colorado municipality or county has lawful authority to prohibit healing centers entirely.³¹ This approach trades a narrower electoral coalition — Colorado’s Proposition 122 passed with a thinner margin of approximately 54% versus Oregon’s 55.75% — for a more uniform geographic implementation footprint.³²

The Colorado Lawyer has characterized the local preemption provision as establishing that local governments “may not prohibit” natural medicine business operations and “may not adopt local ordinances that are unreasonable or conflict with state law regarding licensing and operation of natural medicine businesses.”³³ This stronger statewide framework reduces the fragmentation risk that has characterized Oregon’s program, at the cost of greater political resistance from localities that oppose the program.

V. Risk Category Four: Legislative Rollback Dynamics

A. The Measure 110 Precedent

The most consequential governance risk associated with ballot-initiative drug policy in Oregon is not fragmentation but legislative rollback: the statutory reversal of voter-approved policy by the legislature. Oregon Measure 110, approved by voters in November 2020 alongside Measure 109, decriminalized possession of small amounts of all controlled substances and directed hundreds of millions of dollars from marijuana tax revenue into addiction treatment services — with at least $265 million in grants distributed through 2024.³⁴˒⁴⁸ The Oregon Judicial Department’s own records confirm that the measure went into effect on February 1, 2021, and that possession of controlled substances (PCS) offenses were re-criminalized by House Bill 4002, effective September 1, 2024.³⁵

The process through which Measure 110 was revised illustrates the governance vulnerability of ballot initiatives to subsequent legislative reversal. HB 4002 passed the Oregon House 51-7 and the Senate 21-8, with bipartisan support.³⁶ The Oregon legislature’s rollback was preceded by sustained public criticism linking Measure 110 to Oregon’s fentanyl crisis and rising homelessness — criticism that public health researchers and the Drug Policy Alliance disputed on evidentiary grounds, arguing that national fentanyl supply dynamics rather than decriminalization policy were responsible for the overdose trends cited.³⁷ Regardless of the underlying empirical dispute, the political dynamic was clear: a voter-approved initiative was substantially revised by the legislature within four years of enactment.

The Measure 110 precedent is not perfectly analogous to psychedelic service programs. Measure 110 was decriminalization-only, with no commercial regulatory infrastructure and no licensed industry constituency invested in its continuation. Oregon Measure 109, by contrast, has generated a network of licensed service centers, facilitators, manufacturers, and training programs with direct financial interests in the program’s survival. This stakeholder entrenchment creates stronger political resistance to outright legislative repeal than existed for Measure 110. Nevertheless, partial rollback — through budget cuts to OPS compliance staffing, licensing fee increases that force service center closures, or statutory amendments that narrow the client eligibility base — remains a live risk even where full repeal is politically foreclosed. Legislative advocacy by licensed operators and their professional associations will be the primary check on this form of incremental erosion.

B. The Rollback Risk for Psychedelic Programs

The Measure 110 precedent is directly relevant to psychedelic programs. Oregon Measure 109 — a more complex regulatory program than Measure 110 — will face sustained legislative and political pressure as implementation challenges accumulate. The Linn County Commissioner who described the state’s marijuana rollout as “botched” and voiced skepticism about psilocybin implementation represents a predictable constituency that will use program difficulties to advocate for curtailment or repeal.³⁸

Legislative rollback of psychedelic measures is both legally available and politically predictable. Where initiatives are enacted as ordinary statutes rather than constitutional amendments, they may be amended or repealed by majority vote of the legislature (subject, in many states, to a waiting period). The Health Affairs brief on ballot initiatives notes that state legislatures amended or repealed more than one in five voter-approved initiatives passed since 2010, with the pattern accelerating in recent years.³⁹

Constitutional amendment initiatives — like Colorado’s Proposition 122, which amended the state constitution — are substantially more resistant to legislative rollback, because amendment of the constitution typically requires voter approval. This greater durability comes with a cost: constitutional provisions are also harder to improve or adapt as implementation experience accumulates.

C. The Remedial Legislation Dynamic

Even short of full rollback, ballot-initiative drug programs routinely generate a cycle of remedial legislation that alters the program in ways the initiative’s drafters and voters did not anticipate. Colorado’s SB23-290, the implementation bill for Proposition 122 passed in 2023, significantly amended the voter-approved constitutional framework; SB24-198 made additional changes in 2024.⁴⁰ Oregon’s legislature passed SB 303 in 2023 to add data collection requirements. These legislative additions are often necessary and beneficial — rulemaking gaps must be filled — but they create a governance dynamic in which the elected legislature, rather than the voter-approved initiative, increasingly defines the program’s actual operational parameters. Over time, the initiative may serve primarily as a constitutional floor rather than an operational blueprint.

VI. Risk Category Five: Voter Fatigue and the “Second Ballot” Problem

A. The Sequential Election Dynamic

Oregon Measure 109 embedded a “second ballot” mechanism directly into the initiative text: local jurisdictions were authorized to place opt-out measures on the 2022 general election ballot, requiring voters across the state to revisit psilocybin policy just two years after approving the original measure. This sequential election dynamic is not unique to psilocybin — it tracks the pattern that followed cannabis legalization in Oregon, where dozens of localities placed opt-out measures on ballots following the 2014 recreational marijuana vote.⁴¹

The second-ballot dynamic creates voter fatigue and political attrition: voters who supported the original measure may fail to engage with local opt-out elections, particularly in off-year or lower-turnout elections, ceding the field to organized opposition. Organized opposition groups, who tend toward higher engagement in elections specifically targeting their issue, are structurally advantaged in second-ballot elections compared to diffuse voter majorities.

B. The 2024 Pattern

The November 2024 elections confirmed this dynamic. In Oregon, sixteen of seventeen psilocybin opt-out measures on local ballots passed.⁴² In Massachusetts, voters rejected Question 4 — the third psychedelic ballot initiative in four years nationally — amid what analysts described as voter skepticism about psychedelic legalization specifically and drug liberalization generally in the context of the ongoing fentanyl crisis.⁴³ All three states with recreational marijuana measures on the 2024 ballot — Florida, North Dakota, and South Dakota — rejected legalization.⁴⁴

The 2024 pattern suggests that psychedelic ballot initiative campaigns face structural headwinds from what might be characterized as post-fentanyl voter risk aversion: electorates that supported marijuana and early psychedelic reforms are now more cautious about further drug policy liberalization, particularly in the absence of robust medical and regulatory frameworks that separate therapeutic psychedelic use from recreational drug culture associations.

C. Implications for Future Initiative Design

The voter fatigue problem has direct implications for psychedelic initiative design. Initiatives that include broad personal decriminalization provisions — like Colorado’s Proposition 122 — may appeal to a broader voter coalition on initial passage but generate greater ongoing political resistance, as opponents can characterize decriminalization as enabling recreational use. Massachusetts Question 4 was specifically critiqued for including home cultivation and possession provisions that its proponents later acknowledged alienated swing voters. More targeted initiatives that limit scope to supervised therapeutic access — without decriminalization riders or home cultivation provisions — may produce more durable voter support over time, even if they pass with narrower initial margins.

VII. Risk Category Six: Federal Preemption Exposure

A. The Structural Federal Tension

All state psychedelic programs operate in tension with federal Schedule I classification of psilocybin, psilocin, DMT, ibogaine, and mescaline under the Controlled Substances Act (21 U.S.C. § 812). Unlike cannabis, where nearly three decades of state legalization have established a de facto federal enforcement tolerance codified in annual appropriations riders, psychedelic programs lack this established federal accommodation. The DEA has not issued enforcement guidance analogous to the Cole Memorandum⁴⁹ for psychedelics, and the Rohrabacher-Blumenauer amendment’s protections apply only to state medical cannabis programs.⁴⁵

Ballot initiatives, by their nature, cannot modify federal law. They can create state-law authority for activities that remain federally prohibited, but they cannot provide the facilitators, service centers, or participants who engage in those activities with immunity from federal prosecution. Every psilocybin service center currently operating in Oregon, and every healing center licensed in Colorado, engages in activity that remains illegal under federal law. State initiative text cannot change that fact.

B. Preemption Risk Profiles

Colorado’s SB23-290 includes a savings clause acknowledging federal enforcement authority (C.R.S. § 12-170-111(k)), and the Colorado Lawyer analysis notes that “Colorado is the second state in the union to legalize the use of natural medicines” but that the program remains subject to federal law.⁴⁶ Oregon Measure 109 similarly contains a provision acknowledging Schedule I status (ORS 475A.215(7)). These savings clauses reduce but do not eliminate preemption risk.

The Tufts University Center for State Policy (cSPA) report on Massachusetts Question 4 explicitly concluded that psychedelic programs “could put Massachusetts at risk of a future federal backlash,” citing the initiative’s federal preemption vulnerability as a material governance risk that weighed against voter approval.⁴⁷ This risk calculus will intensify if the federal government shifts enforcement priorities or if psychedelic rescheduling proposals stall, leaving a growing number of state programs exposed to federal preemption claims with no congressional protection.

C. Designing Preemption Buffers Into Initiatives

Future psychedelic initiatives can reduce preemption exposure through several structural design choices. First, expressly limiting program scope to activities that align most closely with existing DEA registration frameworks — cultivation by registered manufacturers, administration at licensed facilities with medical oversight — reduces the gap between state-authorized and federally-prohibited activity. Second, including FDA trigger provisions that automatically align state scheduling with any federal rescheduling decision reduces the duration of the conflict period. Third, adopting a medical oversight model — requiring physician involvement or psychiatric screening — aligns the program with the kind of right-to-try and compassionate access frameworks that have historically attracted federal deference, even without formal DEA exemption.

D. Unresolved Access Challenges: Homebound Patients and Tribal Jurisdiction

Two unresolved access issues cut across the opt-out framework. First, Oregon’s in-person service requirement has been challenged in a federal lawsuit as discriminatory against homebound and disabled patients who cannot travel to licensed service centers — an access gap that the opt-out fragmentation analysis compounds rather than creates.⁵⁰ Second, psychedelic retreat operators claiming tribal sovereign immunity have begun operating outside state licensing frameworks in Oregon and elsewhere, creating a parallel unregulated market whose safety profile is unknown and whose legal status remains contested. Neither issue is resolvable through state initiative design alone, but both should be addressed in enabling legislation and rulemaking frameworks that accompany future initiatives.⁵¹

VIII. Risk Mitigation Framework

The governance risks cataloged above are not arguments against psychedelic ballot initiatives as a policy vehicle. They are arguments for designing those initiatives with a clear-eyed understanding of the structural pathologies that direct democracy creates. The following framework identifies mitigation strategies keyed to each risk category.

On regulatory capture: Initiative drafters should engage public health agencies, professional licensing boards, and academic medical centers as co-drafters, not merely endorsers. Provisions encoding substantive regulatory standards — particularly those governing who may facilitate sessions, what training is required, and how adverse events are handled — should be grounded in evidence-based public health frameworks rather than market-access imperatives. Oregon’s Measure 109 illustrates the cost of inadequate engagement: the prohibition on the word “therapy” in OAR 333-333-1010(72), while preserving a commercial facilitated-access model, produced the scope-of-practice confusion that practitioners and bioethicists have documented since program launch. The tradeoff is real: broader co-drafting processes slow campaign timelines and require consensus among parties with divergent interests, potentially narrowing the voter coalition the initiative needs to pass.

On statutory rigidity: Initiatives should distinguish clearly between “floor” provisions — substantive rights and prohibitions that voters are directly enacting — and “framework” provisions — structural authorities and timelines that agencies and legislatures will implement. Rulemaking authority should be drafted broadly, with specific statutory mandates limited to elements that genuinely require constitutional-level entrenchment. Oregon’s two-year residency requirement, broadly understood to be likely unconstitutional from the moment of enactment but embedded as operative statutory text, generated two years of market distortion before sunsetting on January 1, 2025 — a correctable drafting error that persisted because the initiative format created friction around pre-sunset amendment. The limitation of broad delegation is that it reduces the voter-approved character of the program’s substantive rules, potentially diminishing the democratic mandate advocates cite when opposing legislative rollback.

On local opt-out fragmentation: Drafters should carefully model the geographic equity implications of opt-out provisions before committing them to initiative text. Where opt-out authority is politically necessary for passage, initiatives should include provisions limiting the duration of local bans, requiring affirmative re-approval, or mandating that the state provide access infrastructure in underserved areas. Oregon’s experience — in which 23 of 36 counties and 111 of 241 cities opted out in 2022, followed by sixteen of seventeen additional opt-outs in 2024 — provides a precise empirical baseline for modeling the geographic concentration an opt-out mechanism will produce in a politically heterogeneous state. The fundamental tradeoff, as campaign manager Sam Chapman acknowledged, is that the opt-out provision was essential to building sufficient statewide support for Measure 109: eliminating it might have defeated the measure entirely.

On tribal sovereignty: A distinct fragmentation risk not addressed by standard local opt-out analysis involves psychedelic retreat operations claiming tribal sovereignty as a basis for operating outside state licensing frameworks. Several such operations have commenced in Oregon and other states, arguing that tribal sovereign immunity exempts them from state psilocybin service regulations. Future initiatives should include an explicit provision clarifying the relationship between the state program and tribal sovereign immunity claims — either by expressly incorporating tribal consultation and co-regulatory frameworks (as Colorado did through C.R.S. § 12-170-107), or by preserving tribal sovereign rights while establishing minimum safety standards that apply to all psilocybin services administered to non-tribal members within the state’s jurisdiction.See supra notes 50–51.

On legislative rollback: Psychedelic initiatives should, where possible, be structured as constitutional amendments rather than ordinary statutes in states where constitutional amendment is a practicable vehicle. Where statutory form is required, initiatives should include legislative amendment provisions requiring supermajority approval for substantive rollback. Colorado’s Proposition 122, enacted as a constitutional amendment, demonstrates the durability advantage of constitutional entrenchment: three subsequent legislative enactments — SB23-290, SB24-198, and HB 25-1063 — have refined but not reversed the program, in part because full repeal would require voter approval. The tradeoff is that constitutional provisions are harder to improve as implementation experience accumulates: gaps or errors in initiative text that would be correctable through ordinary legislation require a return to voters at the constitutional level.

On voter fatigue: Initiatives should be designed for durability — avoiding provisions that predictably alienate swing voters in favor of focused therapeutic access frameworks that are more defensible in subsequent ballot contests. Massachusetts Question 4 provides the clearest recent evidence: proponents later acknowledged that the home cultivation provision — authorizing growth of psychedelic plants and fungi in a twelve-by-twelve-foot area — was likely the decisive factor in the measure’s 56.9% to 43.1% rejection, despite broader voter support for supervised therapeutic access. The tradeoff of focused scope is that it may under-deliver for advocates who want decriminalization and personal use provisions included alongside therapeutic access, generating coalition tensions that can fragment the advocacy infrastructure needed to defend the program against future rollback attempts.

On federal preemption: Initiatives should include FDA trigger provisions and savings clauses, should expressly align with right-to-try and compassionate access frameworks where possible, and should avoid provisions that create the greatest gap between state-authorized and federally-prohibited activity. Both Oregon (ORS 475A.215(7)) and Colorado (C.R.S. § 12-170-111(k)) demonstrate that savings clauses are a necessary but insufficient preemption buffer: they acknowledge federal authority without providing immunity from it. The more meaningful preemption reduction comes through substantive design — program architectures that limit scope to medically supervised administration at licensed facilities create a smaller conflict with existing DEA registration frameworks than broad personal decriminalization provisions. The tradeoff is that more medicalized designs may be less accessible to clients who cannot afford clinical services or who do not qualify under restrictive diagnostic criteria.

IX. Conclusion

The ballot initiative has been the primary vehicle for psychedelic legalization in the United States, and it is likely to remain so for the foreseeable future in the approximately twenty-one states that permit citizen-initiated legislation or constitutional amendments. But the initiative process, applied to complex regulatory programs governing Schedule I substances, generates predictable governance pathologies: regulatory capture through drafting, implementation gaps, local fragmentation, legislative rollback, voter fatigue, and federal preemption exposure.

Oregon’s experience — including both the geographic fragmentation of Measure 109 and the substantial legislative revision of Measure 110 — and Massachusetts’s rejection of Question 4, collectively provide the most detailed empirical record available of these pathologies in the psychedelic and related drug policy context. Colorado’s Proposition 122, by adopting local preemption and constitutional entrenchment, has made different tradeoffs that have produced a structurally more durable program, though one that was approved by a narrower initial voter margin and remains subject to the same federal preemption exposure as Oregon’s.

Future psychedelic initiative campaigns should treat these precedents as design inputs, not post-hoc critiques. The goal of expanding therapeutic access to psilocybin and related substances is best served by initiatives that are structurally designed for durability — and by state legislatures that, rather than repeating the implementation failures that contributed to Oregon’s decriminalization challenges, treat voter-approved psychedelic programs as mandates requiring affirmative and sustained institutional support.

Endnotes

¹ Moritz College of Law, Drug Enforcement and Policy Center, Drugs on the Ballot, https://moritzlaw.osu.edu/faculty-and-research/drug-enforcement-and-policy-center/research-and-grants/drugs-on-the-ballot (noting that California Proposition 215 in 1996 initiated the ballot initiative trajectory for drug policy reform; as of November 2024, twenty-four states have legalized adult-use cannabis).

² Id. (fourteen of twenty-four states that have legalized adult-use cannabis did so via citizen-initiated ballot measure).

³ Ballotpedia, Oregon Measure 109, Psilocybin Mushroom Services Program Initiative (2020), https://ballotpedia.org/Oregon_Measure_109,Psilocybin_Mushroom_Services_Program_Initiative(2020); Ballotpedia, Colorado Proposition 122, Decriminalization and Regulated Access Program for Certain Psychedelic Plants and Fungi Initiative (2022), https://ballotpedia.org/Colorado_Proposition_122,Decriminalization_and_Regulated_Access_Program_for_Certain_Psychedelic_Plants_and_Fungi_Initiative(2022).

⁴ Oregon Capital Chronicle, Oregon Local Ballot Measures Seek Voter Input on Psilocybin Mushrooms (Nov. 4, 2024), https://oregoncapitalchronicle.com/2024/11/04/oregon-local-ballot-measures-seek-voter-input-on-psilocybin-mushrooms-wind-energy-star-voting/ (Measure 109 passed with 55.75% of the vote); KUNC, Colorado Legalized Certain Psychedelics Last Year. Now Lawmakers Have to Regulate Them (Apr. 18, 2023), https://www.kunc.org/news/2023-04-18/colorado-legalized-certain-psychedelics-last-year-now-lawmakers-have-to-regulate-them (Proposition 122 approved with about 54% of the vote).

⁵ Boston Herald, Massachusetts Question 4: Voters Reject Legalizing Psychedelics (Nov. 6, 2024), https://www.bostonherald.com/2024/11/06/massachusetts-question-4-voters-reject-legalizing-psychedelics/ (Question 4 failed 56.9% to 43.1%).

⁶ Chelsea L. Shover & Keith Humphreys, Six Policy Lessons Relevant to Cannabis Legalization, 45 Am. J. Drug Alcohol Abuse 698, 702 (2019), https://pmc.ncbi.nlm.nih.gov/articles/PMC6745015/ (identifying regulatory capture as a structural risk in cannabis legalization generally, including the vulnerability of ballot initiative campaigns to industry-aligned funding).

⁷ Catharine Bresnahan & Scott Burris, Cannabis Legalization in State Legislatures: Public Health Opportunity and Risk, 47 Am. J.L. & Med. 231, 246 (2021), https://pmc.ncbi.nlm.nih.gov/articles/PMC8351589/ (finding that the ballot initiative process is a particularly vulnerable point for regulatory capture because industry players can encode pro-profit, anti-public-health rules into law that persist long after the election, often surviving subsequent legislative amendment attempts).

⁸ Id. (legislative legalization may allow health groups to influence law construction without committing funding to a ballot initiative, explaining their structural absence from initiative campaigns and the resulting cession of the drafting field to industry-aligned funders).

⁹ Psychedelic Week, Massachusetts Psychedelic Ballot Question 4 Fails to Legalize Psilocybin Therapy (Nov. 6, 2024), https://www.psychedelicweek.com/p/massachusetts-psilocybin-therapy-mushroom-ayahuasca-iboga-mdma-maps-ballot-question-4-failed (the campaign to pass Question 4 launched July 3, 2023, as Massachusetts for Mental Health Options, funded and overseen by the Washington, D.C.-based New Approach PAC, which had previously funded Measure 109 and Proposition 122).

¹⁰ Harvard Law, Petrie-Flom Center, When the Promises of a Policy Do Not Meet the Reality of Its Practice: Ethical Issues Within Oregon’s Measure 109 (Feb. 12, 2024), https://petrieflom.law.harvard.edu/2024/02/12/when-the-promises-of-a-policy-do-not-meet-the-reality-of-its-practice-ethical-issues-within-oregons-measure-109/ (the use of the term “therapy” in the context of Measure 109 is specifically outlawed under OAR 333-333-1010(72); the measure authorizes only “supported adult use”).

¹¹ Id. (the initiative’s creators’ stated intent to make psilocybin therapy accessible has resulted in a policy that does not allow for any of it, yet OHA and service centers continue to be publicly described in therapeutic terms).

¹² Id. (questions about how multiple licensures apply to scope of practice and whether OHA or professional licensing boards govern facilitator conduct remain unresolved years after program launch).

¹³ Vicente LLP, The Ultimate Guide to Colorado’s Natural Medicine Health Act (SB23-290 Psychedelics Law) (updated Aug. 2025), https://vicentellp.com/insights/ultimate-guide-to-sb23290-colorado-natural-medicine-psychedelics-regulation-and-legalization-bill/ (DORA oversees facilitator licensing; DOR through a new Division of Natural Medicine oversees healing center licensing; the two-agency structure derives from Proposition 122 text and was incorporated into SB23-290).

¹⁴ Colorado League of Municipalities, Natural Medicine, https://www.cml.org/home/topics-key-issues/natural-medicine (identifying SB24-198, Regulated Natural Medicine Implementation, as the 2024 legislative measure addressing implementation gaps); Vicente LLP, supra note 13.

¹⁵ Colorado Secretary of State, Initiative 58 Final Text, https://www.sos.state.co.us/pubs/elections/Initiatives/titleBoard/filings/2021-2022/58Final.pdf (Proposition 122 amended the Colorado Constitution; constitutional provisions require voter approval to amend, providing significantly stronger durability than an ordinary statute).

¹⁶ Harvard Law, Petrie-Flom Center, supra note 10 (lack of access to liability insurance for facilitators, and tax and banking issues for businesses affiliated with a federally illegal Schedule I substance, identified as serious unresolved issues following program launch).

¹⁷ Harris Sliwoski LLP, Oregon Psilocybin: State of the State (2024), https://harris-sliwoski.com/psychlawblog/oregon-psilocybin-state-of-the-state-2024/ (as of January 1, 2025, the two-year Oregon residency requirement for majority ownership in OPS businesses sunsetted; it was “likely unconstitutional from the start” but no challenges arose and the requirement ran its course).

¹⁸ Id. (Oregon’s legislature passed SB 303 in 2023 to impose additional data collection requirements not specified in the original initiative text, illustrating the remedial legislation dynamic).

¹⁹ Boston Herald, supra note 5 (56.9% voted no, 43.1% voted yes after 88% of votes counted).

²⁰ WBUR, Mass. Voters Oppose Legalizing Natural Psychedelics, as Question 4 Fails (Nov. 6, 2024), https://www.wbur.org/news/2024/11/06/massachusetts-psilocybin-drugs-mushrooms-legalization-voter-results (measure proposed decriminalizing psilocybin, psilocyn, DMT, mescaline, and ibogaine; establishing supervised access at licensed facilities; creating a five-person regulatory commission).

²¹ Boston.com, Voters Rejected Question 4. Here’s Why Readers Are ‘Disappointed’ (Nov. 7, 2024), https://www.boston.com/community/readers-say/2024/11/07/why-readers-disagree-with-question-4-results/ (Massachusetts for Mental Health Options acknowledged that the home-grown aspect of the proposal likely led to its rejection; multiple voters confirmed the home cultivation provision was the deciding factor in their no vote).

²² Psychedelic Week, supra note 9 (the Tufts cSPA report noted that ibogaine “can cause acute cardiac problems” and questioned whether the initiative adequately differentiated safety protocols for ibogaine relative to psilocybin).

²³ Boston.com, Massachusetts Voters Reject Legalizing Psychedelics (Nov. 6, 2024), https://www.boston.com/news/politics/2024/11/06/question-4-voters-reject-legalization-of-some-psychedelic-substances/ (the Coalition for Safe Communities, which opposed the measure, said its coalition included organizations representing over 60,000 physicians and psychiatrists nationwide and characterized the measure as “poorly written and misguided”).

²⁴ Colorado Lawyer, Colorado’s Bold Move to Legalize Psychedelics (2024), https://cl.cobar.org/features/colorados-bold-move-to-legalize-psychedelics/ (a local jurisdiction may regulate time, place and manner but “may not prohibit” operation of natural medicine businesses or adopt ordinances that are unreasonable or conflict with state law); Harris Sliwoski LLP, Oregon Psilocybin: Which Cities and Counties Are In? (2022), https://harris-sliwoski.com/psychlawblog/oregon-psilocybin-which-cities-and-counties-are-in/ (in Oregon, automatic opt-in applies unless local voters elect to opt out through a referendum in a statewide general election).

²⁵ Oregon Capital Chronicle, Many Oregonians Will Have to Vote Again on Psilocybin (Aug. 8, 2022), https://oregoncapitalchronicle.com/2022/08/08/many-oregon-voters-will-have-to-decide-again-on-a-psilocybin-program/ (Chapman stated that the opt-out provision was essential to voter support; without local exit options, rural and conservative communities likely would have shifted the statewide outcome).

²⁶ Oregon Capital Insider, Many Cities, Counties Opt Out of Oregon Psilocybin Program (Nov. 28, 2022), https://oregoncapitalinsider.com/2022/11/28/many-cities-counties-opt-out-of-oregon-psilocybin-program/ (in 23 of Oregon’s 36 counties and 111 of its 241 cities, voters opted out of the state program or imposed two-year pauses in the November 2022 election).

²⁷ Harris Sliwoski LLP, Oregon Psilocybin: State of the State (2024), supra note 17 (service centers and related economic activity concentrated along the I-5 corridor from Portland to Eugene, with limited pockets in Jefferson and Deschutes counties).

²⁸ OPB, Psilocybin Bans Were on the Ballot in 17 Places in Oregon. They Passed in All But One (Nov. 7, 2024), https://www.opb.org/article/2024/11/07/psilocybin-ban-mushroom-hallucinogenic-ballot-politics-elections-oregon/ (sixteen of seventeen 2024 opt-out measures passed; a dozen communities converted two-year moratoriums into permanent bans).

²⁹ Oregon Health Authority, Oregon Psilocybin Services, https://www.oregon.gov/oha/PH/PREVENTIONWELLNESS/Pages/psilocybin-services.aspx (primary source for permitted counties).

³⁰ Oregon Capital Chronicle, supra note 25 (Chapman expressed concern that if large sections of eastern Oregon ban the program, residents would have to travel hundreds of miles, incurring substantial costs beyond treatment fees); Harris Sliwoski LLP, State of the State (2024), supra note 17 (the bigger access issue is lack of services for disabled and homebound patients who cannot travel).

³¹ Colorado Lawyer, supra note 24 (local jurisdictions may not prohibit healing center operations; only time, place, and manner regulations are permissible).

³² KUNC, supra note 4 (Proposition 122 approved with approximately 54% of the vote, a thinner margin than Oregon’s 55.75%).

³³ Colorado Lawyer, supra note 24.

³⁴ Drug Policy Alliance, Oregon’s Measure 110: What Really Happened (2024), https://drugpolicy.org/resource/oregons-measure-110-what-really-happened/ (Measure 110 approved by Oregon voters in November 2020, making Oregon the first state to decriminalize possession of small amounts of all drugs and directing marijuana tax revenue to addiction services).

³⁵ Oregon Judicial Department, Measure 110 (2024), https://www.courts.oregon.gov/about/Documents/BM110Statistics.pdf (Measure 110 approved November 2020, took effect February 1, 2021; PCS offenses re-criminalized by 2024 HB 4002, effective September 1, 2024).

³⁶ OPB, Oregon’s Drug Decriminalization Experiment Appears Dead (Mar. 2, 2024), https://www.opb.org/article/2024/03/01/oregon-legislature-passes-bill-recriminalizing-drug-possession-sends-to-governor-measure-110/ (Oregon House passed HB 4002 51-7; Oregon Senate passed it 21-8 with bipartisan support).

³⁷ Drug Policy Alliance, supra note 34 (drug possession recriminalized in 2024 after what the Alliance characterized as an intense disinformation campaign scapegoating Measure 110 for overdose trends driven by national fentanyl supply dynamics).

³⁸ Oregon Capital Chronicle, supra note 25 (Linn County Commissioner Nyquist described the state’s cannabis rollout as “botched” and stated he had no confidence the state would manage psilocybin implementation correctly, representing a constituency that will use implementation difficulties to advocate for curtailment).

³⁹ Health Affairs, Direct Democracy and Population Health: Making Health Policy Through State Ballot Initiatives (2024), https://www.healthaffairs.org/do/10.1377/hpb20240821.364598/ (state legislatures amended or repealed more than one in five voter-approved initiatives between 2010 and 2023).

⁴⁰ Vicente LLP, supra note 13 (SB23-290 passed May 2, 2023, signed May 23, 2023, implementing and amending Proposition 122; SB24-198 made additional changes in 2024 as part of ongoing implementation refinement).

⁴¹ Oregon Capital Chronicle, supra note 25 (dozens of Oregon cities and counties banned marijuana production and sales after voters statewide legalized the drug in 2014, anticipating the same pattern that occurred with psilocybin in 2022).

⁴² OPB, supra note 28.

⁴³ U.S. News & World Report, A Growing Number of Oregon Cities Vote to Ban Psychedelic Mushroom Compound Psilocybin (Nov. 22, 2024), https://www.usnews.com/news/best-states/oregon/articles/2024-11-22/a-growing-number-of-oregon-cities-vote-to-ban-psychedelic-mushroom-compound-psilocybin/ (quoting Josh Hardman, founder of Psychedelic Alpha: “Perhaps the fact that the drug policy reform pendulum appears to be swinging back towards prohibition is part of a broader trend toward a preference for ‘law and order’ among American voters”).

⁴⁴ Akerman LLP, Cannabis Reform Ballot Initiatives Take a Hit in 2024 Election (2024), https://www.akerman.com/en/perspectives/cannabis-reform-ballot-initiatives-take-a-hit-in-2024-election.html (all three states with adult-use marijuana measures on the 2024 ballot — Florida, North Dakota, and South Dakota — rejected legalization).

⁴⁵ Consolidated Appropriations Act of 2023, Pub. L. No. 117-328, div. BB, § 531 (Rohrabacher-Blumenauer protections apply only to state medical cannabis programs, not to psychedelic programs).

⁴⁶ Colorado Lawyer, supra note 24 (Colorado is the second state to legalize use of natural medicines; SB23-290 § 12-170-111(k) includes savings clause acknowledging federal enforcement authority).

⁴⁷ Psychedelic Week, supra note 9 (Tufts cSPA concluded that Question 4 “could put Massachusetts at risk of a future federal backlash” because psychedelics remain federally illegal, constituting a governance risk that weighed against voter approval).

⁴⁸ Oregon Capital Chronicle, Oregon House Passes Bill Unwinding Measure 110 to Address Addiction Crisis (Feb. 29, 2024), https://oregoncapitalchronicle.com/2024/02/29/oregon-house-passes-bill-unwinding-measure-110-to-address-addiction-crisis/ (reporting at least $265 million in grants distributed through 2024 from marijuana tax revenues directed to addiction treatment services under Measure 110).

⁴⁹ U.S. Dep’t of Justice, Memorandum for All United States Attorneys: Guidance Regarding Marijuana Enforcement (Aug. 29, 2013) (Cole Memorandum), https://www.justice.gov/iso/opa/resources/3052013829132756857467.pdf, rescinded by U.S. Dep’t of Justice, Memorandum for All United States Attorneys: Marijuana Enforcement (Jan. 4, 2018) (Sessions Memorandum) (the Cole Memorandum established federal enforcement priorities deprioritizing state-compliant cannabis activity; no equivalent guidance has been issued for psychedelic programs).

⁵⁰ Filter Magazine, Lawsuit Seeking Psilocybin Home-Care in Oregon Permitted to Proceed (June 2025), https://filtermag.org/lawsuit-psilocybin-home-care-oregon/ (federal lawsuit challenging Oregon’s in-person service requirement as discriminatory against homebound and disabled patients; illustrating unresolved access and legal boundary questions in Oregon’s program).

⁵¹ C.R.S. § 12-170-107 (Colorado Federally Recognized American Tribes and Indigenous Community Working Group, establishing a co-regulatory framework and mandatory advisory representation as a model for addressing tribal sovereign immunity questions in state psychedelic program design).