State Initiatives to Promote Cannabis Industry Entrepreneurship Among Disproportionately Impacted Communities: A Multi-State Analysis

Abstract

Objectives:

Given the historic cannabis-related injustices in the US, several states with legalized nonmedical cannabis have initiatives to promote industry participation among disproportionately impacted communities and areas (DICAs). This study assessed these initiatives and relevant outcomes across states.

Design:

This was a mixed-methods study using publicly available data.

Setting:

Evaluation was attempted among 17 states with non-medical cannabis retail (as of May 2025) and state-level social equity (SE) entrepreneurship initiatives.

Main Outcome Measures:

Data were collected regarding: (1) characteristics of state SE entrepreneurship initiatives, including: (a) licenses available and/or reserved, (b) selection process, (c) eligibility criteria, and (d) training and financial support; and (2) relevant outcomes, including: (a) number of SE applications and SE licenses issued; and (b) percent of: SE applications licensed, total licenses issued to SE entrepreneurs, and total licenses held by each sex, race, and ethnicity.

Results:

Eleven states reserved a number/percentage of licenses, and certain states reserved specific licenses (n = 2) or specified licenses available (n = 3) for SE entrepreneurs. Each state required majority ownership, and most considered DICA as possible (n = 12) or required criteria (n = 3), required state residency (n = 2) or included other criteria (eg, DICA, farming-related) integrating residency (n = 13), and considered cannabis-related convictions as possible (n = 12) or required criteria (n = 2). Most provided support (training n = 14, financial n = 12, reduced fees n = 10). Percentages ranged for SE applications licensed (Arizona: 2.0% to Massachusetts/Michigan: 100%), licenses issued to SE entrepreneurs (Washington: 2.4% to Maryland: 100%), female-held licenses (Illinois: 22.0% to New Jersey: 44.0%), and minority-held licenses (Vermont: 11.3% to New Jersey: 60.3%).

Conclusions:

This multistate assessment marks a key step in evaluating SE entrepreneurship initiatives. However, outcomes (eg, licenses issued to SE entrepreneurs or DICA subgroups) were not clearly associated with characteristics of these initiatives, suggesting the need to consider different outcomes or allow greater time for SE initiatives to mature and have an impact.

Introduction

Since 2014, 24 states and DC have legalized nonmedical cannabis (as of May 2025¹), with most of these states (all but DC and those with recently passed laws) also establishing cannabis retail.^1,2 Part of the motivation for legalization was to rectify the historic disproportionate impacts of criminal justice policies and high incarceration rates, economic hardship, and social inequities among racial/ethnic minority and low-income populations.^2–6 Thus, these states have also enacted laws providing explicit pathways to either expunge, pardon, seal, or annul the records of those with low-level cannabis convictions, as well as to reinvest cannabis revenues in disproportionately impacted communities and areas (DICAs).^2,5,7–10

Acknowledging that efforts beyond these are necessary to address these disproportionate impacts, many states with legal nonmedical cannabis have implemented SE initiatives to address inequitable participation in cannabis entrepreneurship,^2,11 which has emerged for various reasons (eg, limited business experience or financial resources.^12–14) Such initiatives involve allocating or reserving licenses and/or providing resources (eg, training, financial assistance) for DICAs.^8,15–20

Despite the rapid expansion of cannabis retail and related SE initiatives across states, research linking SE initiative characteristics to key outcomes is just emerging. One 2024 study of 12 cities in 10 states assessed various dimensions of these initiatives, including licensure selection/allocation, eligibility criteria, and financial and training resources for SE entrepreneurs.¹⁹ Results of that study suggested that issuing SE and non-SE licenses at equal ratios and with multiple SE criteria, more discrete DICA and cannabis conviction criteria, and more comprehensive financial and training resources maintained greater fidelity to redistributive justice goals.¹⁹

To date, most analyses of SE initiatives have primarily characterized the initiatives^8–10 or provided outcomes within a single state or a relatively small subset of states.^9,19,21 Thus, additional multistate studies are needed to assess SE initiative outcomes, especially given potential challenges to these initiatives and the limited transparency and sustainability of their outcomes to date.² The current study aimed to advance the literature by providing multistate analyses of SE initiative outcomes. Using drug policy frameworks specific to cannabis^5,6,22 and cannabis SE policies,¹⁹ we identified key aspects of cannabis SE initiatives to promote industry participation among DICAs, indicators of DICA representation in cannabis entrepreneurship, and characteristics of state initiatives that support DICA representation.

Methods

First, we defined our sample of states to assess. Overall, 24 states had legalized nonmedical cannabis retail, as of May 2025; however, we excluded Ohio and Virginia, as Ohio’s nonmedical market was in its initial phases, with nonmedical cannabis license applications only open to current medical cannabis license holders, and Virginia had not yet issued non-medical retail licenses. SE initiatives related to equitable industry entrepreneurship were present in 18 of the remaining 22 states (82%); those without such initiatives included Alaska, Maine, Montana, and Oregon (see Table 1). California was distinct among the states, as its SE initiatives were defined and over-seen by local jurisdictions. As such, evaluation was attempted among the 17 states with state-level SE entrepreneurship initiatives plus California.

TABLE 1.

Summary of Cannabis Social Equity (SE) Programs to Promote Equitable Industry Entrepreneurship Among Disproportionately Impacted Communities and Areas (DICAs) Across^a18 US States With Legal Nonmedical Cannabis Retail

	Licenses Available and Selection Process			N Possible SE Eligibility Criteria		Eligibility Criteria											Training and Financial Assistance			Outcomes
State	License reserve	Prior. Rev.	Lottery vs merit selection	Needed	Possible	Ownership	Resid	Income	DICA Resid	Social Disadvantage or DICA Member	Race/Ethnicity	Veteran/Military	Personal Conviction	Family Conviction	Prior SE Career or Impact	Farming Related	Training/TA	Fee Waiver	Grants/Loans	% White-Ownede	% SE Licenses/Total Licenses
Arizona	LR	-	LS	3	4	≥51%	b	P	P	-	-	-	P	P	-	-	Y	-	-	n/a	15.3%
Colorado	-	-	MS	1	3	≥51%	-	P	P	-	-	-	cP	cP	-	-	Y	-	Y	11.6%	12.8%
Connecticut	LR	-	LS	0	0	≥65%	b	R	R	-	-	-	-	-	-	-	Y	Y	Y	n/a	12.8%
Delaware	LR	-	LS	1	3	≥51%	b	-	P	-	-	-	P	P	-	-	Y	Y	-	n/a	37.6%
Illinois	LR	PR	LS	0	0	≥51%	b	-	R	-	-	-	cR	cR	-	-	Y	-	Y	13.3%	54.1%
Maryland	LR	-	LS	1	3	≥65%	b	-	P	-	-	-	-	-	-	-	Y	-	Y	−24.4%	100%
Massachusetts	LR	-	MS	1	3	≥51%	b	P	P	-	P	-	cP	cP	P	-	Y	Y	Y	n/a	47.7%
Michigand	-	-	MS	1	3	>50%	b	-	P	-	-	-	P	-	-	-	Y	Y	Y	−0.3%	16.2%
Minnesota	LR	-	LS	1	5	≥65%	b	-	P	-	-	P	P	P	-	P	-	-	Y	n/a	98.1%
Missouri	LR	-	LS	1	5	>50%	b	P	P	-	-	P	cP	cP	-	-	Y	Y	-	n/a	66.7%
Nevada	LR	-	LS	0	0	≥51%	b	-	R	-	-	-	cR	cR	-	-	-	Y	-	30.1%	2.6%
New Jersey	LR	PR	MS	1	2	>50%		P	P	-	-	-	P	-	-	-	Y	-	Y	−20.8%	n/a
New Mexicod	-	-	MS	2	6	≥50%	b	P	-	-	P	-	P	-	-	P	-	Y	Y	32.6%	n/a
New Yorkd	LR	PR	MS	1	5	≥51%	-	-	-	P	P	P	-	-	-	P	Y	Y	Y	0.8%	44.9%
Rhode Islandd	LR	-	LS	1	5	≥51%	R	P	P	-	-	-	cP	cP	P	-	Y	Y	Y	n/a	n/a
Vermont	-	-	n/a	1	3	≥51%	R	-	P	-	P	-	cP	cP	-	-	Y	Y	Y	−3.2%	39.6%
Washington	LR	-	MS	2	4	≥51%	b	P	P	P	-	-	cP	cP	-	-	Y	-	-	n/a	2.4%
SE initiatives exclusively led by local jurisdictions
California	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	(L)	12.4%	3.3%

Abbreviations: Resid, Residency; TA, technical assistance; LR, licenses reserved; PR, priority review; LS, lottery selection; MS, merit-based selection; R, required criteria; P, possible criteria; L, delegated exclusively to local jurisdictions.

^aIncludes California although SE initiatives exclusively led by local jurisdictions. Excludes: Ohio and Virginia (no active nonmedical cannabis markets as of May 2025); Alaska, Maine, Montana, and Oregon (no initiatives to promote SE entrepreneurs). See Table 2 and Supplemental Digital Content Table S1, available at: http://links.lww.com/JPHMP/B632, for additional details.

^bResidency integrated into other criteria (eg, DICA residency, farming-related). Arizona: 1 possible DICA criterion addresses residency. Connecticut: 1 DICA requirement addresses residency. Delaware: 1 possible DICA criterion addresses residency. Illinois: One possible DICA criterion addresses residency. Maryland: All 3 possible criteria address residency. Massachusetts: 3 possible criteria address residency. Michigan: 1 possible DICA criterion addresses residency. Minnesota: 1 possible DICA criterion addresses residency. Missouri: 2 possible DICA criteria address residency. Nevada: 1 DICA requirement addresses residency. New Jersey: 1 possible criterion addresses DICA and income. New Mexico: 2 possible criteria address race (AI/AN, Black Hispanic, Asian/PI); possible criterion related to farming involves residency. Washington: Two possible DICA criteria address residency.

^cConsidered 1 criterion (either one or the other).

^dAdditional criteria: Michigan: Primary caregiver; New Mexico: Sexual or gender minority; New York: Women/females. Rhode Island: Additional possible criteria: for applicants with a minimum of 10 full-time employees, applicants with ≥51% of employees who meet ≥1 of these criteria: (1) currently reside in DICA, (2) own or familial arrest, conviction, or adjudicated delinquent for any offense eligible for expungement.

^eRed text indicates higher licensee % vs state population %; blue indicates lower.

Next, we developed a codebook to guide data collection, which was informed by cannabis SE frameworks and prior research.^5,6,19,22,23 Data were collected regarding: (1) characteristics of state initiatives to promote SE entrepreneurship, including: (a) types of licenses available and/or reserved for SE entrepreneurs, (b) selection process, (c) eligibility criteria for SE entrepreneurs, and (d) training and financial support for SE entrepreneurs; and (2) relevant outcomes, including: (a) number of SE applications and SE licenses issued; and (b) percentages of: SE applications resulting in licenses, total licenses issued to SE entrepreneurs, and total licenses held by sex, race, and ethnicity.^5,6,19,22,23

Then, 2 trained MPH- and PhD-level researchers systematically reviewed relevant resources (eg, state laws, state agency websites/reports) to identify sources of SE-related information, as these often varied in terms of their documentation in state laws vs other official texts (eg, cannabis licensing board requirements, processes, etc.).²⁴ Based on this review, we identified core information sources (eg, legislation, cannabis licensing board, department of public health) in each state. To assess relevant state laws and statutes, the researchers examined state legislative documents (per best practices for legal research^25,26) and were supervised by a legal expert (author YTY). They used cannabis licensing board websites and other relevant online sources (eg, SE task force or council reports) to identify data pertaining to SE-related activities and processes for implementing SE-related laws and initiatives.

After defining the core information sources for each state, the 2 researchers systematically reviewed these resources to refine the initial coding framework in collaboration with the study team. Finally, the 2 researchers independently dual-coded variables (alongside their data sources) and compared their data after coding the first 2 states. After addressing any discrepancies, they refined the coding scheme to reduce ambiguity as needed. They then independently dual-coded the remaining states and compared all codes and data sources to ensure reliability. Finally, available outcomes were compared to state population characteristics and contextualized with regard to SE eligibility criteria, resources, and licensure-related characteristics. (Note that quantitative analyses assessing SE initiative characteristics in relation to outcomes were considered but underpowered and inappropriate due to data limitations.)

Results

SE entrepreneur eligibility and resources

Table 1 provides an overview of licenses available and selection process, eligibility criteria, and resources for SE entrepreneurs. Table 2 summarizes data regarding: (1) the types of licenses available in general and to SE entrepreneurs and the selection process; (2) the number of: (a) overall licenses issued, (b) licenses available or allocated for SE entrepreneurs, and (c) SE applications submitted; and (3) the percentages of: (a) SE applications that were licensed, and (b) total licenses that were issued to SE entrepreneurs. Supplemental Digital Content Table S1, available at: http://links.lww.com/JPHMP/B632, provides additional details regarding eligibility criteria and resources for SE entrepreneurs.

TABLE 2.

Characteristics of State Cannabis Social Equity (SE) Initiatives to Promote Industry Entrepreneurship Among Disproportionately Impacted Communities and Areas (DICAs) Among^a18 US States With Legal Nonmedical Cannabis Retail

State	Types of Licenses Available Overall and for SE Entrepreneurs	# of Licenses Issued Overall	# of Licenses Available or Allocated for SE Entrepreneurs	# of SE Applications Submitted	# of SE Applications Selected and License Type	% of SE Licenses/Applications	% SE Licenses/Total Licenses
Arizonab (as of May 2024)	Overall: Retailer, cultivation, processing, manufacturing, packaging, storing, testing lab1 SE: Same as overall2	170 (capped)3	26 available for each license type;2 Lottery selection4	13015	26 retailer licenses2	2.0% (capped)	15.3%
Colorado (as of May 2025)	Overall: Cultivation (516), delivery (18), hospitality (15), operators (11), product manufacturers (232), stores (691), testing facilities (7), transporter licenses (13)6 SE: Same as overall7	14506	No allocation; accelerator licenses reserved for SE to work with accelerator-endorsed licensees8 (retail, cultivation, product manufacture)7	No data found	185 total9	n/a	12.8%
Connecticut (as of April 2025)	Overall: Retailers (42), delivery (10), cultivator (14), food & beverage manufacturer (9), micro-cultivator (7), transporter (4), product manufacturer (3), product packagers (1)10 SE: Same as overall10	9010	50% of each license type reserved11	106812 (as of December 2024)c	27 total SE: 9 retailers, 6 delivery, 5 food & beverage manufacturer, 2 micro-cultivator, 2 transporter, 2 product manufacturer, 1 product packagers; 34 total equity joint venture: 30 retailer, 1 delivery, 2 micro-cultivator, 1 transporter; DIA: 14 cultivator10	n/a	12.8%
Delaware (as of November 2024)	Overall: Retail (15), cultivator (60), product manufacturer (30), testing lab (4)13 SE: Same as overall14	125 (capped)14	47 reserved: 15/30 retail; 20/60 cultivator; 10/30 manufacturer; 2/5 testing labs14	52415	47 total: 15 retailer, 20 cultivator, 10 manufacturer, 2 testing13	9.0% (capped)	37.6%
Illinois (as of July 2024)	Overall: Dispensary (217), cultivation centers (21), craft growers (87), infusers (55), transporters (163)16 SE: Same as overall16	54316	55 dispensing licenses per year reserved17 d	~2700 (for lottery conducted in 2024)16	294 dispensing, 87 craft growers, 54 infusers, 139 transporters16 e	10.9%	54.1%
Maryland (as of June 2024)	Overall: Standard dispensary, micro dispensary, standard grower, micro grower, standard processor, microprocessor18 SE: Same as overall18	20518 f	First round of lottery—exclusively reserved for SE applicants across all license types19	170819	205 total: 75 standard dispensary, 10 micro-dispensary, 20 standard grower, 30 micro grower, 40 standard processor, 30 microprocessor18	12.0%	100%
Massachusetts (as of May 2025)	Overall: Cultivator (353), manufacturer (281), retailer (482), delivery (177), courier (163), microbusiness (44), transporter (27), craft marijuana coop (6), testing lab (20), microbusiness delivery (6), research facility (7) SE: Same as overall20	1566 (approved, pending, reopened)20	No reserved licenses in general, but exclusive access to delivery licenses till 2025 at least20,21 Note: EEA: economic empowerment applicant; DBE: disadvantaged business enterprise; EEA, DBE, and SE not mutually exclusive	74720 g	747 total;20 367 SE, 135 EEA, 186 DBE, 59 to 2+ groupsg	100%	47.7%
Michigan (as of April 2025)	Overall: Grower (Class A, B, C, excess, 1001), processor (285), retailer (853), microbusiness (Class A & other; 17), transporter (22), safety compliance facility (22), consumption area (4), educational research license (1), cannabis event organizer (39)22 SE: Same as overall22	224422 (216 licenses issued in May 2025)	No reservations	~3522 (only in 4/2025)	~35 SE licenses issued (new or renew);22 20 grower, 1 processor, 13 retailer, 2 Class A microbusiness, 2 cannabis event organizer	100%	16.2% (only for May 2025)
Minnesota (as of May 2025)	Overall: 312 for microbusiness (293), testing facility (8), transporter (3), wholesaler (3), delivery service (5); 324 licenses available for those capped: mezzo-business (100), cultivator (50), manufacturer (24), retailer (150). Others (microbusiness, testing facility, transporter, wholesaler, delivery service, med combo) not capped23 SE: Same as overall23	312 (license applications qualified for preliminary approval)	50% of capped license types allocated;23 Lottery for capped licenses, general review for those without cap24	1741 (as of 5/2025, carried over from fall 2024)23	No licenses, including general licenses, issued23 and lottery not yet conducted for capped licenses; 306 of 312 licenses preliminarily approved are for SE applicants: 306 total: 2 testing facility, 293 microbusiness, 3 transporter, 3 wholesaler, 5 delivery service23	17.6% (some capped; column 2)	98.1%
Missouri (as of 2023; 2024–2025 not yet available)	Overall: Cultivation (3), dispensary (5), manufacturing (4), microbusiness delivery (16), microbusiness wholesale (32), seed to sale (6), transportation (6) SE: All microbusiness licenses reserved for SE—2 types: wholesale and dispensary25	7226	48 reserved for each year (16 dispensary, 32 wholesale)27	208327	96 awarded to date (39 in the first round and +57 in second); 32 microbusiness dispensary, 64 microbusiness wholesale25 (48 if timepoint matched with column 3)	4.6%	66.7% (to match the timepoint of data in column 3, % = 48/72)
Nevada (as of May 2025)	Overall: Cultivation (122), production (97), testing labs (8), lounges (2), retailers (106), distributors (46) SE: Eligible only for consumption lounge licenses; 10 reserved for SE, 10 for others28	38129	10 of first 20 independent consumption licenses reserved28	3030	10 (however, 6 of them have been rescinded)30	33.3%	2.6%
New Jersey (as of March 2025)	Overall: Cultivator (517), manufacturer (383), wholesaler (20), distributor (15), retailer (1293), delivery (34), testing lab (9)31 SE: Same as overall31	2271 (total licensees: 691 annual, 1580 conditional)31	≥30% of total licenses go to 3 types: SE, diversely owned, and impact-zone businesses (municipalities with large population, or high unemployment, crime, or marijuana arrest rate); not mutually exclusive32	741 SE, 589 economically disadvantaged, 1283 impact zone, 2076 diversely ownedh	530 SE, 412 economically disadvantaged, 988 impact zone, 1626 diversely owned (approved, including conditional)31	71.5% SE, 69.9% economically disadvantaged, 77.0% impact zone, 78.3% diversely owned	n/a (categories not mutually exclusive)
New Mexico (as of December 2024)	Overall: Producer (400), micro-producer (675), manufacturer (900), retailer (1006), consumption area-(28), courier (56), testing lab (6), research lab (0)33 SE: Same as overall33	~307133	No reservations,34 but goal of 50% representation35	No data found	No data found	n/a	n/a
New York (as of December 2024)	Overall: Cultivator (221), processor (292), distributor (185), retailer (448), microbusiness (272), Conditional adult use retail dispensary (CAURD) (569)36 SE: Same as overall36	~1500 (licenses issued in 2023–2024)36	50% of all adult-use licenses reserved;36 CAURD: for justice involved businesses or qualifying nonprofits37	499536	673 total: 76 cultivator, 145 processor, 60 distributor, 253 retailer, 139 microbusiness36	13.5%	44.9%
Rhode Island	Overall: Retail licenses38 SE: Same as overall38	24 (new licenses to open)38	6/24 reserved38	No data found	No data found	n/a (capped)	n/a
Vermont (as of December 2024)	Overall: Cultivator (394), manufacturer (91), retailers (102), testing lab (2), wholesaler (9);39 also propagation license (no data on #) SE: Same as overall39	60139	No reservations40	No data found	100 total SE: 66 cultivator, 16 manufacturer, 18 retailer; 138 total economic empowerment: 89 cultivator, 20 manufacturer, 26 retailer39	16.6% SE, 23.0% economic empowerment	39.6% (across both SE and economic empowerment)
Washington (as of December 2024)	Overall: Retailers (476), transportation (12), research (2), producers/processors (1198), cooperative licenses (10)41 SE: Retailers, producers, processors available42	169841	Up to 52 retail, 10 producer, 100 processor;41 all forfeited, revoked, or cancelled licenses also reserved43	494 total applicationsi	40 total awarded; annual report indicates 3 SE retailer licensesi	8.1%	2.4%
SE initiatives exclusively led by local jurisdictions
California (as of May 2025)	Overall: Cultivation (3654), distributor (1053), microbusiness (365); nursery (243), processor (119); retailer (1216), retailer-non storefront (305), infusion (102), packaging (27), shared-use (9), manufacturer (435) SE: Same as overall44	7528 (adult use: 4379, adult and medicinal use combined: 3149)45	Varies by city,46 but available for each license type44	~280047 (as of December 2023)	249 total: 233 commercial retailer, 16 commercial microbusiness44	8.9%	3.3%

^aIncludes California although SE initiatives exclusively led by local jurisdictions. Excludes: Ohio and Virginia (no active nonmedical cannabis markets as of May 2025); Alaska, Maine, Montana, and Oregon (no initiatives to promote SE entrepreneurs).

^bSE licensing process done once in 2022.⁴⁸ Only establishment licenses mentioned in website.².

^cTwo types: SE and Equity joint venture (SE partners with licensed producer, cultivator or dispensary);⁴⁹ SE applicants may also apply for provisional cultivator license for facility in DICA without participating in lottery.⁵⁰.

^dAnother 55 licenses reserved under “Qualifying applicant” and 75 under “tied applicant” (not mutually exclusive); SE applicants can participate in all 3 lotteries.¹⁷.

^eTotal SE awardees, including conditionally accepted.

^fOverall cap on licenses (not yearly limit);⁵¹ 205 does not reflect capped licenses.

^gSame n as applications as all approved, pending, or reopened businesses. 367 SE: 116 courier, 24 cultivator, 136 delivery, 29 manufacturer, 50 retailer, 5 transporter, 5 microbusiness delivery, 2 other microbusiness; 135 EEA: 44 courier, 11 cultivator, 23 delivery, 10 manufacturer, 43 retailer, 3 transporter, 1 research facility; 186 DBE: 49 courier, 11 cultivator, 2 delivery, 42 manufacturer, 82 retailer, 5 transporter, 5 microbusiness, 1 craft marijuana coop; 59 to 2+ groups: 3 courier, 13 cultivator, 16 delivery, 5 manufacturer, 3 retailer, 1 transporter, 1 microbusiness delivery, 13 other microbusiness, 3 craft marijuana coop, and 1 testing lab.

^h677 SE: 205 cultivator, 113 manufacturer, 11 wholesaler, 11 distributor, 318 retailer, 18 delivery, 1 testing lab; 534 economically disadvantaged: 138 cultivator, 67 manufacturer, 7 wholesaler, 7 distributor, 297 retailer, 18 delivery; 1190 impact zone: 314 cultivator, 199 manufacturer, 13 wholesaler, 14 distributor, 624 retailer, 24 delivery, 2 testing lab; 1936 diversely owned: 392 cultivator, 274 manufacturer, 19 wholesaler, 19 distributor, 1184 retailer, 45 delivery, and 3 testing lab.³¹

ⁱPer personal communication.

The types of licenses available to SE entrepreneurs were the same as the general types of licenses in most states (n = 14; Table 2), and 11 states reserved a number/percentage of licenses across all or specific license types (Arizona, Connecticut, Delaware, Illinois, Maryland, Minnesota, Nevada, New Jersey, New York, Rhode Island, Washington). There were certain states where specific licenses types were reserved for SE entrepreneurs (ie, regular licensees not eligible, n = 2; Massachusetts: courier, delivery operator, social consumption establishment; Missouri: micro-businesses—either wholesale or dispensary) or where SE entrepreneurs could only pursue certain licenses types (ie, “SE entrepreneurs” could qualify for only certain licenses; n = 3), including Missouri (only microbusiness licenses), Nevada (only independent consumption lounge licenses; 10 reserved for SE, 10 for other applicants), and Washington (only retailer, producer, and processor licenses). States were divided in their use of lottery-based (n = 9) vs merit-based (n = 7) selection processes.

Each state required majority ownership (typically ≥50%−51% or ≥65%; Table 1; Supplemental Digital Content Table S1, available at: http://links.lww.com/JPHMP/B632). Almost all considered DICA as either a required criterion (n = 3) or a possible criterion (n = 12), and most also required state residency (n = 2) or included possible or required DICA or other criteria (eg, farming-related) that integrated state residency (n = 13). Another common criterion was personal or family member cannabis-related convictions, most often considered possible criteria (n =12 for personal; n = 9 for family) or less often as required criteria (n =2 for personal or family). Less common criteria included racial/ethnic minority status (n = 4), veteran/military (n = 3), farming-related (n = 3), member of a socially disadvantaged group or DICA (without residency; n = 2), prior SE career or impact (n = 2), etc. (eg, primary caregiver, sexual/gender minority, female, n = 1 each).

Most states with SE initiatives provided training and/or technical assistance (TA; n = 14), reduced application and licensing fees to SE applicants/licensees (n = 10), and/or other financial aid (eg, grants, loans; n = 12; Table 1; Supplemental Digital Content Table S1, available at: http://links.lww.com/JPHMP/B632; Supplemental Digital Content Appendix S1, available at: http://links.lww.com/JPHMP/B631).

Outcomes

Shown in Table 2, based on the most recent available data, the percentage of SE applications licensed ranged from 2.0% (Arizona) to 100% (Massachusetts, Michigan). The percentage of licenses issued to SE entrepreneurs (across all licenses) ranged from 2.4% (Washington) to 100% (Maryland).

Table 3 provides the available data on sex, race, and ethnicity of licensees in the 10 states that provided any data publicly (which exclude Arizona, Connecticut, Delaware, Minnesota, Missouri, Rhode Island, and Washington), alongside the racial and ethnic composition of each state’s population for context and comparison. In each state, more males than females held licenses (range of % male: 59.0% in New Jersey to 88.0% in Illinois), despite roughly equal proportions of males and females across these 10 states. In all states, the greatest proportion of licenses was held by White individuals (range: 32.1% in New Jersey to 88.7% in Vermont), except Maryland (23.9% White licensees vs 53.6% Black licensees). Four of the 10 states (Maryland, Missouri, New Jersey, and Vermont) had lower proportions of White licensees compared to their state populations. The 6 other states (California, Colorado, Illinois, Nevada, New Mexico, and New York) had higher proportions of White licensees relative to their state populations. Illinois only provided the proportion of licensees who were White and aggregated the remainder who represented minority race or ethnicity (27.2%).

TABLE 3.

Racial/Ethnic Profiles of State Populations and State Cannabis Licensees, as Well as Differences, Among ^a10 States With Nonmedical Cannabis Retail and Metrics Related to Social Equity (SE) Initiatives to Promote Equitable Industry Entrepreneurship Among Disproportionately Impacted Communities and Areas (DICAs) (as of May 2025)

		Whiteb			Blackc			Asianc			AI/ANc			Mixed/Other Racec			Hispanicc
State	Female	State	Licensee	Diff	State	Licensee	Diff	State	Licensee	Diff	State	Licensee	Diff	Stated	Licensee	Diff	State	Licensee	Diff
Colorado	20.5	66.5	78.1	11.6	4.1	4.3	0.2	3.6	2.4	−1.2	0.6	0.5	−0.1	2.5	1.7	−0.8	22.5	9.8	−12.7
Illinois	12.0	59.5	72.8	13.3	14.1	n/a	n/a	6.1	n/a	n/a	0.2	n/a	n/a	1.8	n/a	n/a	18.3	n/a	n/a
Maryland	36.6	48.3	23.9	−24.4	30.4	53.6	23.2	6.9	13.2	6.3	0.2	0.5	0.3	2.7	4.9	2.2	11.5	3.9	−7.6
Michigan	26.6	74.0	73.7	−0.3	13.8	5.8	−8.0	3.5	4.1	0.6	0.6	0.5	−0.1	2.5	n/a (other: 7.7)	5.2	5.7	2.4	−3.3
Nevada	25.3	45.7	75.8	30.1	9.7	2.2	−7.5	8.9	9.9	1.0	0.8	1.1	0.3	3.9	n/a (other: 2.2)	−1.7	30.3	10.9	−19.4
New Jersey	41.0	52.9	32.1	−20.8	13.0	17.0	4.0	10.3	12.0	1.7	0.2	0.3	0.1	1.7	2.2 (+other 11.7)	12.2	21.9	11.0	−10.9
New Mexico	29.7	35.7	68.3	32.6	2.0	4.1	2.1	1.7	2.2	0.5	8.8	2.7	−6.1	1.7	5.9 (+other: 9.2)	13.4	50.2	44.2	−6.0
New York	27.9	54.2	55.0	0.8	14.4	14.6	0.2	9.4	6.7	−2.7	0.3	0.8	0.5	1.9	n/a (mixed + other: 8.1)	6.2	19.7	15.7	−4.0
Vermont	35.7	91.9	88.7	−3.2	1.4	4.5	3.1	2.0	3.3	1.3	0.3	0.4	0.1	2.0	n/a (other: 1.5)	−0.5	2.3	n/a	n/a
SE initiatives exclusively led by local jurisdictions
California	34.5	34.7	47.1	12.4	5.6	20.2	14.6	15.6	7.5	−8.1	0.4	2.1	1.7	3.0	n/a (other: 2.2)	−0.8	40.3	12.5	−27.8

^a10 states that provided any data publicly (excludes Arizona, Connecticut, Delaware, Minnesota, Missouri, Rhode Island, and Washington). State population characteristics—source: https://usafacts.org/data/topics/people-society/population-and-demographics/our-changing-population/state/+statename.

^bRed text indicates higher licensee % vs state population %; blue indicates lower.

^cRed text indicates lower licensee % vs state population %; blue indicates higher.

^dMixed race, per state population statistics.

Shown in Table 3, the ranges of licensees by other races varied (range of Black licensees: 2.2% in Nevada to 53.6% in Maryland; range of Asian: 2.2% in New Mexico to 13.2% in Maryland; range of AI/AN: 0.3% in New Jersey to 2.7% in New Mexico), the percentage of minority licensees was greater than the state population in several states (n =7 for Black: California, Colorado, Maryland, New Jersey, New Mexico, New York, and Vermont; n = 6 for AI/AN: California, Maryland, Nevada, New Jersey, New York, and Vermont; n = 6 states for Asian: Maryland, Michigan, Nevada, New Jersey, New Mexico, and Vermont). Data regarding those reporting mixed or other races should be interpreted with particular caution (given that different labels were provided for these data); 5 states (Maryland, Michigan, New Jersey, New Mexico, and New York) had proportions of mixed/other licensees greater than their state populations. The proportion of Hispanic licensees was lower than the state population in all 10 states and ranged from 2.4% (Michigan) to 44.2% (New Mexico).

Table 1 provides key outcomes—percentage White-owned licenses and percentage of total licenses issued to SE entrepreneurs—alongside characteristics of the states’ SE initiatives. No discernable patterns between specific characteristics of state initiatives to promote SE entrepreneurship and the percentage of White-owned licenses and the percentage of total licenses issued to SE entrepreneurs were found.

Discussion

This study takes an initial step in evaluating state initiatives to promote equitable industry entrepreneurship among DICAs in states with legal nonmedical cannabis retail. This study documented a range of approaches (eg, reserved licenses, training, financial aid) and potential indicators of better licensing outcomes for SE entrepreneurs or underrepresented groups (eg, percentage of racial/ethnic minority licensees, percentage of total licenses issued to SE entrepreneurs). However, there were no clear links between characteristics of state initiatives to promote SE entrepreneurship and these outcomes.

Nonetheless, potential successes and shortcomings of these initiatives should be considered. In 4 of the 10 states, there were lower proportions of White licensees compared to their state populations, and the percentage of minority licensees was greater than the state population in several states, specifically for Black entrepreneurs in 7 states, AI/AN entrepreneurs in 6 states, and Asian entrepreneurs in 6 states. On the other hand, in all 10 states, the proportion of Hispanic licensees was lower than the state population, and more males than females held licenses. Notably, the broad ranges in these figures across states underscore the need for transparency (to the extent possible) in documenting the contexts and potential mechanisms for both shortcomings and successes, which existing literature emphasizes as crucial for interpreting and deepening our understanding of SE initiatives.¹⁰ However, these findings are notable given other literature suggesting that census tracts with more racial/ethnic minorities and socioeconomic disadvantage in the US were more likely to have cannabis retailers,²⁷ although there might be differences in these associations based on race/ethnicity and socioeconomic factors.²¹ Together, these results highlight multiple influences (eg, more exposure to retail but less industry participation and benefit) that might perpetuate disparities faced by certain populations.

These states provided opportunities for SE entrepreneurs to pursue a broad range of license types. A couple of states had unique licensure initiatives, with 2 states reserving certain licenses exclusively for SE entrepreneurs, and 3 allowing SE entrepreneurs to only pursue certain types of licenses. However, none of these approaches seemed to reflect differences in outcomes (eg, percentage of SE applications awarded licenses; number of licenses issued to SE entrepreneurs or to females or racial/ethnic minority groups). Similarly, there were no clear indicators of successful outcomes based on eligibility criteria—neither the types nor the number of criteria, while a prior study suggested better outcomes when states issued SE and non-SE licenses at equal ratios and had multiple SE criteria and more discrete DICA and cannabis conviction criteria.¹⁹ However, contextual factors, like lawsuits and inappropriate manipulation of policies, particularly related to preferential licensing, may undermine SE initiatives.^28–33 For example, Washington’s 6-month state residency requirement has been the focus of one court case disputing the rationale for such longevity of state residency,²⁹ and some companies have canvassed lower-income areas to identify someone to apply for a license backed by the larger company, then ultimately undermined the applicant’s business ownership.^30–33

Importantly, the majority of states provided training and financial assistance, which are essential to support SE entrepreneurs, particularly those with limited business experience or financial resources.² Indeed, the financial considerations for business owners in general and for SE entrepreneurs attempting to enter the cannabis industry are broad-ranging (eg, capital, credit, real estate constraints, and property requirements). Unfortunately, these concerns are multiplied by the fact that prior research documented that non-SE (vs SE) retailers cluster in more affluent neighborhoods with more access to commercial and tourist activity,¹⁹ which may hinder the success of SE entrepreneurs. However, the current study failed to find any clear connection between the availability of training and financial resources on outcomes, which may be partially attributable to the inability to assess the level of financial funding in relation to outcomes. States’ annual budgets often dictate the amount and duration of support for SE entrepreneurs¹⁹; thus, funding levels are not clear nor consistent over time, which also poses barriers to implementation and evaluation.²

One possible reason that current results did not suggest clear associations between characteristics of state SE entrepreneurship initiatives and outcomes may be the need to better consider evaluation metrics (ie, how to operationalize initiatives and what outcomes to include). For example, only 4 of the 17 states explicitly considered racial/ethnic minority status and only one considered sex—so the percentage of the total number of licenses issued to SE entrepreneurs from racial/ethnic minority groups or females may not map on to the criteria each state is using to evaluate the success of their SE entrepreneurship initiatives. Additionally, the percentage of applications resulting in licensure may not be an appropriate outcome in some states where the number of licenses is capped, as the numerator cannot increase and thus depends on the denominator (which ideally would increase with support to SE entrepreneurs). Moreover, the percentage of licenses issued to SE entrepreneurs may be inflated by specific approaches that are not intended to be maintained over time. For example, the outcomes in Maryland are highly impacted by the fact that only one lottery has been conducted so far, and only SE entrepreneurs were eligible for that lottery³⁴ (thus leading to 100% of licenses being issued to SE entrepreneurs).

Another potential reason this study did not clearly identify state SE entrepreneurship characteristics that fostered better SE licensure outcomes may be that some of these initiatives have not yet had time to mature or take effect. For example, states with training and TA involving mentorship by regular licensees are compelling but lack evidence, as Colorado had yet to issue an accelerator license (as of February 2024³⁵) and Connecticut’s programs are just being implemented. Along these same lines, the time since nonmedical cannabis retail went into effect and the year these SE initiatives were launched may have had some impact. While most states that have SE initiatives coincided their legislation for establishing cannabis retail and SE initiatives, Colorado and Washington had significant gaps of time between nonmedical cannabis legalization and SE program establishment.^2,10 These gaps may have led to inequitable business opportunities in the initial years of legalization and thus, provide particular challenges in resolving inequities.^19,36

Relatedly, another factor that may have hindered the ability to identify SE initiatives related to SE licensure success is the limited available data. Many states are still grappling with systematic and timely reporting of the number of licenses being issued to SE entrepreneurs. Some states, like Delaware, Massachusetts, and Minnesota, have clear information on total SE license applications and issuances, while other states like Michigan and Washington had less clear public information on the total number of licenses issued to SE applicants. This is likely due to the relatively new nature of these initiatives, including in states that were early to establish nonmedical cannabis markets and/or SE entrepreneurship initiatives. Ultimately, all states should establish a systematic data reporting process, with shared metrics across states, to provide crucial information and facilitate robust evaluation of state initiatives to promote SE entrepreneurship. States should consider openly communicating important SE program characteristics, including the application process (eg, preliminary application, pre-approval), details regarding financial support (eg, financial resource availability), and any local initiatives, as well as key evaluation outcomes, such as SE licensing rates (including by SE-related criteria), SE-owned business survival over time, uptake of training and financial resources, and local variation in outcomes.

Strengths and Limitations

The current study has some key strengths. Importantly, this study assessed all states with active nonmedical cannabis retail and SE entrepreneurship initiatives, describing the characteristics of these initiatives and available data regarding indicators of success. However, this study is limited given its reliance on publicly available documents, which were not uniformly available and did not consistently present information across states. Thus, these data might not reflect the most up-to-date outcomes or practices and may not have been interpreted within the context of each state’s own metrics of success. Additionally, this study focused on state-level initiatives—and did not capture local initiatives—for 2 reasons: (1) SE initiatives are largely at the state-level, and (2) the literature and the available data across states (much less across localities) related to SE initiative evaluation are nascent. Relatedly, quantitative analyses were not appropriate due lack of power because of limitations to available outcome data (eg, diversity of applicants/licensees, SE licensees/applications). Additional research is needed to link program characteristics (eg, merit- vs lottery-based selection, financial resources) to these outcomes as more data become available—including at the state and local levels. Finally, these outcomes may have been impacted by confounders not assessed in this study (eg, other societal or policy factors, financial context within the state). Nonetheless, this study can provide important context for other studies that may more succinctly quantify the nature of these SE initiatives and metrics of success in the future.

Supplementary Material

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (http://www.JPHMP.com).

Implications for Policy & Practice.

• This study revealed significant variations in states’ approaches to enhance equitable cannabis entrepreneurship opportunities and highlights important implications for policy, practice, and research.

• Unfortunately, this study did not identify any specific characteristics of SE entrepreneurship initiatives that were linked to indicators of success. However, limitations to the available data for evaluation and the inability to detect effects point to some crucial steps.

• First, greater transparency and broader sharing of data regarding SE initiative outcomes are needed.

• Relatedly, metrics concerning key outcomes must be carefully considered to reflect various dimensions (eg, key SE licensee equity metrics, resource utilization, intended and unintended outcomes) and consistently collected across states.

• Furthermore, states should ensure the timely implementation of SE-focused initiatives, as they may require time to mature and impact cannabis industry participation among DICAs.

• Finally, research must assess the supports in place for SE entrepreneurs over time, given that resources vary due to state budgetary considerations35 but remain essential for SE entrepreneurs.^19,37

• Also crucial to research evaluating SE initiatives are sustainable efforts and resources to facilitate community-engaged, ethically driven research built on collaboration and trust.¹⁰