UPDATE: State Judge Holds Use of Race in Ohio Medical Marijuana Licensing Unconstitutional
A state trial judge has just held that Ohio’s use of racial preferences in the award of state medical marijuana cultivation licenses violated the Equal Protection Clause of the Fourteenth Amendment (and the Ohio state constitution). The full decision in PharmaCann Ohio, LLC v. Ohio Department of Commerce can be found at Pharmacann Ohio v. Ohio Dep’t Commerce Summary Judgment Ruling.
The decision is notable because it is the first time that a court has ruled on the constitutionality of measures aimed at reducing race disparities in marijuana licensing. (The book discusses those disparities, the steps different states have taken to address them, and the legal issues raised by them on pages 454 n.1 and 520-525.)
In this post, I’ll provide some additional background on the PharmaCann case; digest the court’s decision invalidating Ohio’s use of racial preferences; and make three observations on the ramifications of the decision.
First, some background. I wrote about the facts of the case in a prior post:
“Ohio incorporated racial preferences into its own medical marijuana licensing system in 2016. The relevant provision of the Ohio statute (RC section 3796.09(C)) requires the state licensing board (the Department of Commerce) to
‘issue not less than fifteen per cent of . . . licenses to entities that are owned and controlled by . . . . [persons who] are members of one of the following economically disadvantaged groups: Blacks or African Americans, American Indians, Hispanics or Latinos, and Asians.’
In 2017, the Department received applications from 36 firms that met the state’s minimum qualifying criteria. The Department then scored and ranked those applications based on their business, operations, quality assurance, security, and financial plans. . . . In November 2017, it proceeded to award 12 provisional cultivation licenses. The winners included the top 10 scoring firms, but also applicants ranked 14th [(Parma Wellness)] and 23rd [(Harvest Grows)], respectively.
See Cases to Watch: Applicant Sues to Block Ohio from Using Racial Preferences in the Award of Commercial Marijuana Licenses (Dec. 15, 2017).
Two of the “losers” in the licensing process, PharmaCann Ohio (originally ranked 12th) and Greenleaf Gardens (11th), subsequently sued the DOC claiming, inter alia, that RC section 3796.09(C) violates the Equal Protection Clause.
Although the case still bears its name, PharmaCann is no longer a party to the suit, due to an interesting twist in the case. Namely, PharmaCann’s claims were mooted in May 2018 when the DOC awarded it a new license. The DOC acknowledged it had made errors in scoring applications, and that absent those errors, PharmaCann would have ranked 8th. But rather than revoke one of the 10 licenses it had awarded to other non-minority applicants, the DOC decided to create an additional (13th) license for PharmaCann. One might wonder why the DOC didn’t simply award Greenleaf an additional license to moot its claims as well (and thereby make this suit go away), but there would be several problems with that strategy. Among other things, creating a 14th license would have put the DOC out of compliance with section 3796.09(C). That section calls for the agency to award “at least fifteen percent” of licenses to minority owned and operated applicants, but 2/14 is only 14.3 percent.
So the case continued with Greenleaf as the lone plaintiff. On November 16, 2018, the judge granted Greenleaf’s motion for summary judgment on the grounds that RC section 3796.09(C) violated the Equal Protection Clause and the Ohio Constitution.
In so doing, the judge (correctly, I think) held that strict scrutiny applies to RC section 3796.09(C). (PharmaCann, p. 5) Although the law ostensibly targets “economically disadvantaged groups”, it defines that term entirely by race – making this a race-based classification. That means the state had to show the law was narrowly tailored to serve a compelling government interest – a tall task. Indeed, the judge found that the state failed both prongs of the strict scrutiny test.
First, the judge held that the state had failed to demonstrate a compelling interest in using racial preferences to award marijuana licenses. The judge acknowledged that the state’s asserted interest – “redressing past and present effects of racial discrimination . . . where the State itself was involved” – could be compelling, for constitutional purposes. (PharmaCann, p. 7) However, he was not convinced by the state’s evidence that there was discrimination (past or present) against all members of designated class (i.e., all the different racial groups included in the definition of “economically disadvantaged groups”) in the relevant market (i.e., the Ohio legal marijuana industry.)
The government appeared to offer three types of evidence to demonstrate racial discrimination.
1) A 2013 ACLU study and anecdotes showing racial disparities in arrest rates for marijuana related offenses in Ohio.
The ACLU study is excerpted and discussed in the book on pages 248-249. This study was perhaps the government’s strongest evidence demonstrating racial discrimination concerning marijuana. Nevertheless, the judge found it to be insufficient:
“Defendants included evidence of statistical studies published by the American Civil Liberties Union in 2013. . . . This data, in connection with the vast amount of anecdotal evidence provided by Defendants, shows the legislature considered evidence of racial disparities for African Americans and Latinos regarding arrest rates for crimes related to marijuana. The Court does not find this to be evidence supporting a set aside for economically disadvantaged groups, including not only Blacks or African Americans, Hispanics or Latinos, but also American Indians and Asians, who are not referenced in either the statistical evidence or the anecdotal evidence on arrest rates. Evidence of increased arrest rates for African Americans and Latinos for marijuana generally, is not evidence supporting a finding of discrimination within the medical marijuana industry for Blacks or African Americans, Hispanics or Latinos, American Indians, and Asians.” (PharmaCann, pp. 8-9)
In other words, the judge found this evidence a) did not demonstrate racial discrimination against the entire benefited class, only part of it; and b) did not demonstrate discrimination in the relevant market, i.e., the legal marijuana industry. I think this is a relatively straightforward application of City of Richmond v. J.A. Croson, 488 U.S. 469 (1989), where Supreme Court made similar points about the shortcomings of using evidence of discrimination against one group in one market to support preferential treatment of other groups in other markets.
2) Studies showing discrimination in the award of government contracts.
These studies had been successfully used by Ohio to defend minority set aside programs (called Minority Business Enterprise) for government contracting the state adopted in 1980 and 2003.
But the court found this evidence insufficient for two reasons. For one thing, it suggested that the studies were irrelevant because the legislature had not actually considered them when it adopted RC section 3796.09(C):
“The only evidence clearly considered by the legislature prior to the passage of R.C. §3796.09(C) is marijuana related arrests. . . . There is evidence that legislators may have considered MBE history and specifically requested the inclusion of a provision similar to the MBE program. However, the only evidence provided are a few emails seeking a provision like the MBE program and revising the terms. . . . No testimony shows any statistical or other evidence was considered from the previous studies conducted for the MBE program.” (PharmaCann, pp. 8-9).
And crucially, the judge held that he would ignore – or heavily discount – evidence not actually considered by the legislature when passing the program:
“Some of the evidence Defendants provide may not have been considered by the legislature during their discussion of R.C. §3796.09. . . . [In a related context, c]ourts have reached differing conclusions as to whether post-enactment evidence may be used in a court’s analysis; . . . [one court has] held ‘post-enactment evidence may not be used to demonstrate that the government’s interest in remedying prior discrimination was compelling.’ . . . The Court finds this ruling to be the most persuasive. Even if the Court were to find post-enactment evidence permissible, it would give it weight similar to [another court], which has held, ‘the main focus . . . must be the legislative findings and informational backdrop which was available to the state legislature prior to the enactment.’ (PharmaCann, pp. 7-8) (citations omitted)
The judge also suggested that the government contracting studies – like the marijuana arrest studies – were not relevant to discrimination in the legal marijuana industry:
“[E]ven if the Court could find this evidence was considered by the legislature in support of R.C. §3796.09(C), the materials . . . pertain to government procurement contracts only. The law requires that evidence considered by the legislature must be directly related to discrimination in that particular industry.” (PharmaCann, p. 11)
Indeed, the judge went so far as to suggest that the state could not (yet) prove there was discrimination in the legal marijuana industry because that industry is entirely new: “[S]uch newness necessarily demonstrates that there is no history of discrimination in this particular industry, i.e. legal cultivation of medical marijuana.” (PharmaCann, p. 11)
I suspect the judge’s decision regarding this second type of evidence (studies of government contracting) is more vulnerable than his decisions regarding other matters in the case. For one thing, as the judge acknowledges, courts disagree about whether they should consider evidence of discrimination the legislature itself may not have considered. An appeals court might hold that the judge should have considered this evidence (even if it might not have changed the result in the case). What is more, while the Croson Court suggested that the evidence the state uses must demonstrate discrimination in the relevant market, I don’t think that necessarily forecloses consideration of evidence showing discrimination in closely related markets (as the judge in PharmaCann seems to suggest in the last line quoted above). After all, a clever plaintiff could always describe a particular market as being “new” or somehow distinct from one in which discrimination was previously demonstrated (e.g., Uber is not the same as Lyft, which is not the same as a taxi, which is not the same as a livery service, and so on). And as a practical matter, it might prove impossible to demonstrate discrimination in small markets (like Ohio’s medical marijuana cultivation industry) without making comparisons to other markets.
To be sure, figuring out which comparisons are relevant can be tricky, as I discuss in a previous post on a study commissioned by Maryland to defend its use of racial preferences in marijuana licensing. See New Study Defends Constitutionality of Racial / Gender Preferences in the Award of State Marijuana Licenses. But suggesting that a state can only redress past discrimination that has been demonstrated in the targeted market, and cannot attempt to nip such discrimination in the bud (so to speak) – namely, before it appears – seems to me too strict a reading of Equal Protection doctrine.
3) Discrimination in other states’ marijuana licensing programs and their efforts to combat such discrimination
Lastly, the state offered news accounts “showing disparities perpetuated in other states’ marijuana programs” and citing measures adopted by different states to boost minority participation in their marijuana industries. (PharmaCann, p. 12)
The judge, again, found this evidence irrelevant, both because it was all generated “post-enactment” (PharmaCann, p. 11) and thus (like the contracting studies above) could not have been considered by the legislature, but also because “[n]one of the programs [adopted by other states includes] a mandatory set aside, like like R.C. §3796.09(C)”:
“Illinois’s program allots additional points if 51% of an applicant is owned by a minority, female, veteran, or disabled person. 68 Ill. Adm. Code §1290.70(d)(7). Pennsylvania includes general requirements for businesses to include diversity plans and requirements of the government to foster the submission of diverse applications. 28 Pa. Code §1141.32.
Similarly, Maryland’s law includes requirements that the government shall encourage diverse applicants and ‘[t]o the extent permitted by federal and State law, actively seek to achieve racial, ethnic, gender, and geographic diversity when licensing medical cannabis growers[.]’ Md.Code Ann., Health-Gen. 13-3306. Florida does not include a broad minority set aside, like Ohio, but a provision focusing solely on Black farmers who are members of the Black Farmers and Agriculturalists Association-Florida Chapter . . . Fla.Stat. §381.986.” (PharmaCann, pp. 11-12)
Second, the judge also found that the law was not narrowly tailored to serve a compelling interest. On this point, the court cited several problems with the adoption of the racial quotas contained in RC section 3796.09(C):
For one thing,
“The Defendants did not show evidence of any alternative remedies considered by the legislature before enacting R.C. §3796.09(C).
The Court believes alternative remedies could have been available to the legislature to alleviate the discrimination the legislature stated it sought to correct. If the legislature sought to rectify the elevated arrest rates for African Americans and Latinos/Hispanics possessing marijuana, the correction should have been giving preference to those companies owned by former arrestees and convicts, not a range of economically disadvantaged individuals, including preferences for unrelated races like Native Americans and Asians.” (PharmaCann, p. 14)
Furthermore, the court was troubled that the 15% target was arbitrarily chosen:
“Defendants admit that the 15% stated within R.C. §3796.09(C) was lifted from [the minority set aside program used for government contracting] without any additional research or review by the legislature regarding the relevant labor market described in R.C. §3796.09(C), the medical marijuana industry. . . .
Defendants argue that the numbers asassociated with the contracting market are directly applicable to the newly created medical marijuana industry because of a disparity study conducted by Maryland [Professor’s note: the one I mentioned above.] . . . The Maryland study was not reviewed by the legislature before enacting R.C. §3796.09(C), and is a review of markets and disparity in Maryland, not Ohio. Accordingly, the Court finds this one study the Defendants use to try to connect two very different industries (government contracting market and a newly created medical marijuana industry) have little weight, if any. . . .
Regarding the statistics the legislature did review prior to enacting R.C. §3796.09(C), the cited statistics pertaining to the arrest rates of minorities are not directly related to the values listed within the statute. . . . Within the 2013 ACLU study, . . . [it] ‘found that black Ohioans were arrested 41 times more often for marijuana possession than white Ohioans in 2010.’ . . . This number is the evidence most directly related to the effects of discrimination based on marijuana arrests in Ohio statewide. Yet, this statistic, or any of the other statistics cited in the materials, is not reflected in the amount chosen to remediate the discrimination . . ., fifteen percent. This percentage is not based on the evidence demonstrating racial discrimination in marijuana related arrests in Ohio. Therefore, the Court can only conclude the numerical value was selected at random by the legislature, and not based on the evidence provided.” (PharmaCann, pp. 15-16)
Notwithstanding my concerns with some of the judge’s reasoning (noted above), I think he reaches the correct conclusion on the (un)constitutionality of RC section 3796.09 under current Equal Protection doctrine.
So what are the ramifications of the decision?
1. It is unclear what the decision means for the licenses awarded to Parma Wellness and Harvest Grows (the two applicants owned and operated by economically disadvantaged groups).
The judge found the racial preferences to be severable from the rest of Ohio’s medical marijuana law, so the decision in PharmaCann doesn’t jeopardize Ohio’s medical marijuana program.
I suspect the court will order the DOC to award Greenleaf (the plaintiff) a license. But that doesn’t necessarily mean DOC will strip Parma Wellness and Harvest Grows of their licenses. The DOC could just create another license for Greenleaf, much as it did to remedy the scoring error it made regarding PharmaCann (see above). Indeed, neither Parma Wellness nor Harvest Grows was even a party to this litigation, which was brought as a suit against the DOC. I’m not sure who would even have standing (in Ohio courts) to demand the revocation of those two licenses. So even if RC 3796.09(C) is unconstitutional, it may yet still have the desired effect on the racial composition of Ohio’s licensed marijuana industry.
The case also holds two lessons for states considering ways to boost minority participation in the marijuana industry:
2. States should consider race-neutral options to achieve more racial diversity in licensing.
The decision highlights just how demanding the requirements of strict scrutiny are –and why it’s often described as “strict in theory, fatal in fact.” Any race-conscious program to boost minority participation in the marijuana industry will likely be subject to these demanding requirements. (There may be some exceptions, like outreach programs that merely encourage minorities to apply for licenses.)
But the ruling in PharmaCann doesn’t apply to race-neutral measures designed to boost minority participation. For example, a state could consider lowering the fees it charges all applicants for marijuana licenses. In Ohio, these fees are quite steep and may have made it comparatively difficult for economically disadvantaged groups (including many minorities) to seek a license. The state charges $20,000 just to apply for a Level I (large scale) marijuana cultivation license, and it charges another $200,000 annually to keep such a license. Lowering those fees could help minority applicants, who might have less access to loans and other sources of capital as compared to white applicants. But because this move would be race neutral (i.e., it would benefit all applicants, regardless of their race) it would not be subject to strict scrutiny. What is more, pursuing race-neutral options like this first could help bolster the case for the use of racial preferences later (i.e., for showing they are narrowly tailored), if a state was not satisfied with the efficacy of race neutral measures.
3. States need to assemble more evidence to sustain race conscious programs.
One of the key problems in this case was that the state legislature did not do enough homework before adopting the racial preferences for marijuana licenses. Even if another court were to view the evidence more charitably (e.g., to consider some of the evidence the judge discounted or ignored), it still might not be enough to pass strict scrutiny.
This is why I applaud efforts like Maryland’s to commission of studies of racial disparities in the marijuana industry and related industries. Maryland’s study may have been flawed (for reasons I discuss in my earlier post) and it may have come too late to save Maryland’s racial preferences (Maryland’s licensing authority balked at using those preferences before the study was commissioned, as noted in the book on pages 520-525). But if a state is committed to using racial preferences in the award of marijuana licenses, it probably needs to undertake that sort of effort.
That’s it for now. Here are two additional links that might be of interest to the reader:
- Suit Challenging Maryland’s Refusal to Use Racial Preferences in Licensing Has Been Settled
- Ohio Medical Marijuana Cultivation Program (program website with information on the licenses that were at issue in PharmaCann)