Local Prohibitions on Marijuana: Factors Associated With Bans on Medical and Recreational Businesses

Abstract

Marijuana legalization in U.S. states has introduced new regulatory processes, which allow for some degree of local authority. A common provision in state laws has provided a local option, which enables municipalities to prohibit commercial cannabis facilities within their borders. This paper examined potential determinants of prohibition decisions, with an emphasis on community and local government characteristics. Using a multivariate logistic regression model, determinants of local decisions to prohibit cannabis businesses in Michigan were analyzed. A unique feature of Michigan’s approach to legalization included separate local options to opt into medical marijuana and opt out of recreational marijuana. The results found that both community and local government characteristics explain variation in prohibitions on marijuana businesses. In general, local governments were more tolerant of marijuana businesses in areas with lower densities of senior citizens and higher densities of Black residents. However, the effects of some determinants varied based on whether the decision was to allow medical or recreational marijuana. Higher income areas were more likely to opt out of recreational marijuana, but they were no different from other areas on the likelihood of opting into medical marijuana. Prohibition decisions also were affected by local government characteristics, such as the number of members on city council, female representation, and having a council-manager form of government.

Keywords

cannabis marijuana marijuana legalization local government

Introduction

Public approval of marijuana use has increased substantially within the U.S. over the past 30 years. According to the Pew Research Center (2019), the legalization of recreational marijuana had support from 67% of U.S. adults during 2019 whereas only 16% supported it in 1989. While the cultivation, distribution, and use of marijuana remain illegal activities based on federal law, most states have legalized the use of marijuana in some form among adults. By February 2022, a total of 37 states have legalized the use of marijuana for medicinal purposes while recreational use also has been legalized in 18 of those states (National Conference of State Legislatures, 2022).¹ However, despite the growing tolerance of marijuana-related activities reflected in public opinion polls and state laws, there still remains a non-negligible portion of the population who continue to view legalization in a negative light.

Attitudes toward the legalization of marijuana have been found to vary across different demographic and socioeconomic groups within the population. Opposition to marijuana legalization tends to be relatively stronger among older age groups, women, and those with conservative political views (Denham, 2019; Elder & Greene, 2019; Ellis et al., 2019; Galston & Dionne, 2013; Nielsen, 2010; Schnabel & Sevell, 2017; Schwadel & Ellison, 2017). Patterns of marijuana use also have been found to differ across population subcategories, with higher rates of consumption observed among young adults (Dai & Richter, 2019; Mauro et al., 2019) and males in lower income groups (Carliner et al., 2017). Less favorable attitudes toward marijuana may reflect other underlying personality traits, such as being risk averse or moralistic, which may vary across population groups and local geographic areas.

Given the authority of local governments to restrict commercial cannabis activities, either through ordinances or zoning regulations, factors influencing city governing bodies have the potential to shape the geography of retail markets. The potential for divisiveness in policy discussions on marijuana legalization has been evident in the design of state laws. Several states have granted local governments the legal option to prohibit business activities related to marijuana, such as commercial growing or selling, from occurring within their borders (Dilley et al., 2017; Leon & Weitzer, 2014; Mosher & Akins, 2019). This form of local option can reduce retail access to marijuana for a nontrivial share of communities (Beltz et al., 2020; Camors et al., 2020). For example, within the first year of legal retail sales of recreational marijuana in Colorado, a majority of cities decided to ban them (Aguilar & Murray, 2014). In the experience of Washington State, 2 years after recreational marijuana had been legalized, approximately 30% of the population were living in communities where retail sales had been banned (Dilley et al., 2017).

This study analyzes whether variation in community characteristics can explain local decisions to prohibit cannabis businesses in the state of Michigan. Given the existing patterns in public opinion on marijuana across different subgroups, certain factors may lead to a higher likelihood of banning marijuana businesses if the views of elected officials reflect those of the general population. However, for a variety of reasons, the link between community characteristics and municipal cannabis policies may be weak or nonexistent. Similar to other states, Michigan municipalities only had authority over business location decisions and were not allowed to ban the consumption of marijuana among its residents, who could travel to other cities to purchase it. Negative views on the legalization of consumption may not extend to judgments on the spatial proximity of businesses in a state where marijuana is already legal and regulated. It is also possible that mayors and city councilors have different opinions when compared to the general population. The results of this paper identified factors influencing local decisions to allow commercial cannabis businesses in Michigan municipalities, which had the authority to opt into medical marijuana and opt out of recreational marijuana.

Background on Marijuana Legalization in Michigan

Restrictions on marijuana in Michigan have been loosened multiple times since voters first approved a statewide ballot proposal to legalize medical marijuana use in November 2008. This ballot proposal was approved by 62.7% of voters and had majority approval within all of Michigan’s 83 counties.² Voter approval of medical marijuana resulted in the adoption of the Michigan Medical Marihuana Act (MMMA), which permitted the legal use of marijuana among qualified patients who were diagnosed with one or more debilitating medical conditions. This law also created a state registration system for both qualified patients and primary caregivers who were allowed to assist qualified patients with the medical use of marijuana. Several years later, further changes were made to state laws on medical marijuana that expanded the types of activities that were permitted under Michigan law. In 2016, the MMMA was amended to allow the manufacture and consumption of marijuana-infused products, such as topicals, tinctures, and edible substances, by registered patients. However, commercial entities located in Michigan were prohibited from growing, processing, or selling marijuana for medicinal purposes until the implementation of the Medical Marihuana Facilities Licensing Act (MMFLA) of 2016, which established a regulatory process for distributing operating licenses.

A key feature of the MMFLA was the shared authority by state and municipal governments on the authorization of commercial marijuana facilities. Under the law, municipal governments were given an opt-in option on whether to allow commercial marijuana facilities within their borders. If a municipal government preferred to opt in, it had to adopt an ordinance that formally specified the allowable types of commercial marijuana facilities that were permitted. Municipal ordinances also could set a maximum limit on the number of each type of facility. However, if a municipal government preferred to prohibit any type of commercial marijuana facility, no action was required. In this case, the municipality simply would defer from opting into the legalization of commercial marijuana facilities.

In addition to a municipal opt-in ordinance, the MMFLA also required that each commercial marijuana facility obtain a state operating license. All license applications were reviewed by the newly created Medical Marihuana Licensing Board, which was an oversight board within the Michigan Department of Licensing and Regulatory Affairs. This board began its review of applications for medical marijuana facilities on December 15, 2017. By July 2018, a total of 106 municipalities adopted opt-in ordinances.³ However, even if a commercial facility obtained a state license and was located within an opt-in municipality, a municipal government still could confine its operations through local zoning regulations.

The legalization of recreational marijuana in Michigan occurred after voters approved a statewide ballot proposal in November 2018. Compared to the referendum on medical marijuana, which occurred 10 years earlier, voters appeared relatively more divided based on geographic variation in the level of support. Overall, the legalization of recreational marijuana had support from 55.9% of voters with majority approval in 50 of Michigan’s 83 counties. The “Yes” vote had the strongest support in Washtenaw County (67.7%) and the weakest support in Missaukee County (39.7%).⁴ As a result of the referendum’s approval, the Michigan Regulation and Taxation of Marihuana Act (MRTMA) was enacted into law on December 6, 2018.

Similar to the earlier law on medical marijuana, the MRTMA provided a mix of state and local authority over the regulation of commercial facilities with operations related to recreational marijuana. As before, commercial facilities still needed to obtain a state operating license. The review of applications for recreational marijuana businesses began on November 1, 2019. Some of the first applications were approved by November 13 with retail establishments opening as earlier as December 1 (Burns, 2019). However, while the MRTMA still granted a local option to municipalities, this decision was now a question of whether to opt out rather than opting in. This opt-out provision was a stark contrast from the earlier law on medical marijuana. If a municipal government preferred to prohibit recreational marijuana businesses within its borders, the city council would have been required to pass an ordinance whereas no action would have been necessary under an opt-in provision. The MRTMA also allowed residents to place an ordinance on a local election ballot via a petition process for a public vote, which could either limit the number of establishments or prohibit them completely. By December 2019, formal opt-out ordinances were passed by approximately 1,400 local governments, which included cities, villages, and townships (WXYZ Detroit, 2019).

Literature Review

Local government decisions on commercial cannabis regulations can be influenced by a variety of motives. According to content analyses of articles in U.S. newspapers, common themes in public discussions on marijuana legalization have included motives related to economic benefits, criminal justice reform, law enforcement, and public health outcomes (H. Kim & Kim, 2018; McGinty et al., 2016). Policy preferences of local residents and their elected representatives also may be influenced by alternative motives related to moral and political beliefs as well as their desire to use legal marijuana products as consumers. In a national survey, McGinty et al. (2017) found that U.S. adults tended to perceive competing motives for legalization as having uneven amounts of persuasiveness, which included higher levels of agreement with pro-legalization arguments related to economic benefits and reduced prison overcrowding. If the relative salience of competing legalization motives varies across different demographic and socioeconomic groups, disparities in community characteristics across Michigan will likely be associated with variation in local approval of commercial cannabis activity.

Prior studies on the diffusion of state-level decisions to legalize medical marijuana have suggested that internal characteristics of state populations have been significant factors in explaining the timing of implementation. For example, a higher likelihood of early adoption has been observed among states with statewide ballot initiatives, citizens favoring liberal political ideology, and a lower percentage of Protestant fundamentalists (Hannah & Mallinson, 2018; Spetz et al., 2019). Even among states that have legalized medical marijuana, the strictness of laws across states has varied with respect to regulations regarding patient registration, product safety, home cultivation, and local authority over dispensary locations (Klieger et al., 2017; Pacula & Smart, 2017). In G. J. Kim et al. (2021) analysis of the strictness of state laws on medical marijuana, regulations on access to medical marijuana were found to be more lenient in states with more liberal state governments, weaker economic conditions, and a higher density of cancer and AIDS patients. Collectively, these studies suggest that tolerance of legal marijuana use has been influenced by political, economic, and public health motives.

Despite prior findings that identified determinants of marijuana legalization, it is not evident that similar factors would influence representatives in Michigan municipalities. The decision to prohibit commercial entities from locating within a municipality does not alter the state law that legalized the possession or consumption of cannabis products. While the ease of legal availability might increase the prevalence of marijuana use following legalization (Everson et al., 2019), the ability to purchase marijuana in surrounding jurisdictions would circumvent a city’s attempts to limit usage. Local prohibition also could undermine efforts to replace illegal suppliers with licensed firms in a regulated market. Given the limitations of local ordinances to restrict marijuana consumption, it is possible that community characteristics have less influence on policy decisions related to the regulation of cannabis businesses.

Below the state level, disparities in community characteristics have been associated with earlier decisions on the location of marijuana dispensaries. Two years following a state law that legalized recreational marijuana, Dilley et al. (2017) observed that 30% of local governments in Washington State had temporary or permanent bans on marijuana dispensaries in place. Other studies have examined business decisions on where to locate marijuana dispensaries. While earlier studies have found the location of medical marijuana dispensaries to be correlated with certain community characteristics, such as existing commercial activity, some additional findings have been mixed. For example, Morrison et al. (2014) estimated that medical marijuana dispensaries in California had a higher likelihood of locating in areas with lower income levels whereas this relationship was not observed in other studies that focused on Los Angeles, CA (Thomas & Freisthler, 2016) and Denver, CO (Boggess et al., 2014). Considering that convenient access to legal marijuana products may lead to higher rates of marijuana use, city-level decisions to prohibit sales will have potential consequences on local public health. Furthermore, perceptions of a link between legal access and usage may heighten the weight that government decision makers place on motives related to quality-of-life concerns and their personal tolerance of being in close proximity to marijuana users.

One previous study by Camors et al. (2020) examined the influence of female representation in local government on policy outcomes related to cannabis businesses in the state of Oregon. After a statewide referendum approved adult-use recreational marijuana in 2014, municipalities in Oregon were allowed to “opt out” by prohibiting cannabis businesses. However, the procedures for opting out were contingent on voting results from the 2014 referendum. If the statewide referendum was opposed by 55% or more of a city’s voters, the city council had the option to pass a prohibition ordinance. However, municipalities also had the option to prohibit cannabis businesses if less than 55% of its voters opposed the referendum, but it would have required voter approval in a future citywide election. In their analysis of Oregon’s opt-out process, Camors, Chavez, and Romi estimated that opt-out measures were more likely to be placed on local ballots in cities with higher property tax rates and greater female representation in mayoral offices and city councils.

Although Michigan and Oregon both provided local options to prohibit cannabis businesses, distinct circumstances in each state may limit the applicability of findings to other states. Compared to Oregon, Michigan is a more populous state with a larger number of municipalities, which provides a larger sample for more precise estimation. Unlike other states, Michigan did not allow municipalities to levy a local sales tax on marijuana that was separate from state sales taxes. Michigan also legalized marijuana later than early adopting states, such as Oregon, which enabled local policymakers to observe the experiences of other communities. Later adoption might have alleviated uncertainty and concerns with legalization, such as unknown consequences related to crime, drugged driving, or conflict with the federal government. In addition, municipalities in Michigan had the authority to pass a prohibition ordinance regardless of how many voters opposed legalization in the statewide referendum. Local choices initially were provided when medical marijuana was legalized and then later for recreational marijuana after it also was legalized. This design feature provides an opportunity to compare the influence of community characteristics on decisions for both types of marijuana use.

Data and Methods

Dependent Variables and Empirical Model

Local government decisions to adopt cannabis ordinances in Michigan were examined using multiple data sources. The outcomes of interest included two dependent variables, which were binary variables indicating whether a municipality decided to (1) opt into medical marijuana legalization and (2) opt out of recreational marijuana legalization. As mentioned earlier, these decisions only affected whether commercial activities would be allowed within a municipality’s borders and could not prohibit residents from possessing or consuming marijuana. The opt-in decision on medical marijuana was measured with an indicator variable equal to “1” if a municipality passed an ordinance by July 2018. This cut-off date provided a time window for observing local ordinance decisions that included a full year following statewide legalization of medical marijuana plus an additional 6 months after license applications for medical marijuana facilities became available. A list of opt-in municipalities at that point in time was obtained from a report conducted by the City of Grand Rapids (2018).

Similarly, a variable indicating opt-out decisions to prohibit recreational cannabis businesses was coded with a value of “1” if a municipality passed an ordinance by November 2019, which included all opt-out decisions made prior to the opening of the first marijuana retailers in December. This cutoff date allowed for a full year of observing ordinance decisions following the statewide legalization of adult-use recreational marijuana in Michigan. A document listing all opt-out municipalities, which was prepared by the Michigan Regulatory Agency, was obtained from a television news outlet in Detroit (WXYZ Detroit, 2019).

Variation in municipal decisions to pass marijuana ordinances was estimated using a multivariate logistic regression model. The functional form of the logistic model can be seen in Equation 1. The dependent variable, Y_i , is a binary variable indicating whether a local government passed an ordinance related to cannabis businesses. Separate estimates of the model were obtained for the medical opt-in and recreational opt-out decisions. The vector, X_i , includes variables measuring the demographic, socioeconomic, and political characteristics of citizens in a local area. Additional independent variables measuring characteristics of a local government are included in Z_i . The cumulative density function of the standard logistic distribution is denoted by Φ. Coefficients of the model are denoted by β and γ. The error term of the index function is denoted by ε_i and assumed to follow a standard logistic distribution.

Pr (Y_{i} = 1 | X_{i}, Z_{i}) = Φ (X_{i} β + Z_{i} γ + ε_{i})

Independent Variables

The models for opt-in and opt-out decisions controlled for demographic, socioeconomic, and political characteristics of local areas. Data for these independent variables were obtained from various sources. Variables on municipal-level population characteristics were obtained from the 2010 decennial U.S. Census (U.S. Census Bureau, 2010), which included a municipality’s total population and the proportion of residents who were age 65 or older, Black, and Hispanic. Based on prior studies finding a negative relationship between age and attitudes toward marijuana (Nielsen, 2010; Schnabel & Sevell, 2017; Schwadel & Ellison, 2017), it is expected that marijuana prohibition will be preferred in areas with more senior citizens. Stronger support for cannabis businesses was expected among larger communities with more racial and ethnic diversity.

Additional variables on socioeconomic characteristics were available at the county level. Median household income and the Gini coefficient for income inequality were obtained from the 2014–2018 American Community Survey (ACS) 5-year estimates (U.S. Census Bureau, 2019). Low-income areas were expected to have more interest in the potential economic benefits of legal cannabis markets and oppose prohibition. Higher values of a county’s Gini coefficient indicate a greater degree of local income inequality, which has a less certain hypothesized effect on local ordinance decisions.⁵ Social cohesion and public trust might be weaker in areas with more income inequality, which could reduce tolerance of marijuana use. On the other hand, economic development could be a stronger motive in jurisdictions where income levels in some neighborhoods lag behind others.

The proportion of religious adherents in a county and a county’s marijuana arrest rate also were included as independent variables to control for local interest in legal marijuana. Data on religious adherence was obtained from a 2010 study conducted by the Association of Statisticians of American Religious Bodies (Grammich et al., 2018). County-level arrest rates for marijuana were measured as the total number of marijuana-related arrests per 10,000 people during 2016, which included arrests for possession, sale, or manufacturing. Arrest data were obtained from the FBI’s Uniform Crime Reporting Program (U.S. Department of Justice, 2016). Communities with stronger commitments to religious beliefs were expected to prefer prohibition ordinances whereas higher arrest rates were expected to be associated with greater opposition to prohibition. In addition, citizen support for marijuana legalization was measured using the proportion of county voters who supported the legalization of medical and recreational marijuana in the statewide ballot initiatives from 2008 and 2018, respectively (Michigan Department of State, 2016, 2018).

The logistic models on cannabis decisions also controlled for other variables that indicated a municipal government’s structure and gender diversity. The data for these variables were obtained from the Michigan Municipal League (MML), which publishes an annual directory of elected and appointed municipal officials for all Michigan cities and villages. Since only a small number of urban townships were included in this directory each year, the analysis was limited to cities and villages. Using the 2017 and 2019 editions of the MML’s Directory of Michigan Municipal Officials, it was possible to observe first and last names of all elected officials in city and village government, which included council members, city mayors, and village presidents. Overall, the directory included the names of over 3,500 elected officials from 533 municipalities. Using these directories, a binary variable indicating a council-manager form of government was obtained. Since the decision-making authority of an executive position is relatively weaker under a council-manager government, the influence of council members is expected to have more influence under this form. According to the MML directory in 2019, Michigan had 267 cities and villages with the council-manager form of government, which involves the hiring of a manager to serve as the jurisdiction’s chief administrative officer. The other 266 municipalities had a mayor-council form in which an elected mayor or council president served as chief administrative officer. The number of city council seats, excluding the mayor or council president, also was included as an independent variable to control for a council’s size and relative influence.

Since the MML directory did not indicate the gender of each elected official, gender was identified using Gender API (2020), which is a name-to-gender computing platform. These probability estimates are based on a database of over six million names from 189 countries. When provided with a first name and country of origin, Gender API produces a probability score that estimates the likelihood that a first name is either male or female. The platform reports the probability score for whichever gender is more likely. Hence, all probability scores exceeded 50% and were accompanied by an indicator of the more likely gender. Based on the names in the MML directories, gender probability scores ranged from 53% to 100% and averaged 96.9% and 96.5% in the 2017 and 2019 editions, respectively. Elected officials were classified as being “female” if the Gender API platform indicated that it was the more likely gender. Based on this categorization method, two main independent variables of interest were created to control for female representation in local government. The first gender variable was a binary variable indicating whether a municipality’s executive position, either city mayor or village president, was held by a female. The second variable was a continuous variable that measured the percentage of council members who were female. In the 2 years of MML directory data, the estimated portion of municipalities with female executives ranged from 18.5% to 21.8% while the average percentage of council seats held by women ranged from 34% to 35.5%.^6,7

Descriptive statistics from a sample of 533 Michigan municipalities are shown in Table 1. Despite statewide legalization of marijuana for medical and recreational purposes, local governments in Michigan appeared reluctant to allow cannabis businesses within their communities. Only 9.8% of cities and villages passed an opt-in ordinance to permit medical marijuana facilities while a majority (67.9%) decided to prohibit recreational marijuana facilities. Average values of the variables on female representation suggest that women were underrepresented in local elected offices, but there was wide variation across municipalities. For example, in most municipalities (82.6%), women held 50% or fewer council seats. Some communities had no women on council while others were represented mostly by women.

Table 1.

Descriptive Statistics for Sample of Michigan Municipalities.

Variables	Mean	Standard Deviation	Minimum Value	Maximum Value
Medical marijuana opt-in (0,1)	9.75%	29.70	0%	100%
Recreational marijuana opt-out (0,1)	67.92%	46.72	0%	100%
Population in 1,000s	9.55	36.16	0.11	713.78
% Age 65 and older	15.81%	5.86	5.70%	50.20%
% Black	4.41%	11.72	0%	93.50%
% Hispanic	3.87%	4.68	0%	45.60%
Median household income	53.00	9.87	34.63	80.90
Gini coefficient	44.32	2.76	39.16	49.74
Marijuana arrest rate	26.70	18.35	5.95	165.89
% Religious adherents	39.86%	9.49	16.04%	68.48%
% Yes vote on medical marijuana in 2008	59.46%	5.21	50.49%	71.48%
% Yes vote on recreational marijuana in 2018	52.61%	5.73	39.73%	67.73%
Council-manager government (0,1)	50.1%	50.00	0%	100%
City council size in 2017	5.587	1.215	0	10
City council size in 2019	5.625	1.236	0	10
Female executive in 2017 (0,1)	18.57%	38.92	0%	100%
Female executive in 2019 (0,1)	21.76%	41.30	0%	100%
% Females on council in 2017	35.46%	22.53	0%	100%
% Females on council in 2019	33.98%	20.90	0%	100%
Sample size	533

Note. Sample means, standard deviations, maximum values, and minimum values are provided in the table above.

Results

Logistic model estimates for a Michigan municipality’s decision to allow commercial activities related to medical marijuana are shown in Table 2. The table provides average marginal effects and standard errors. Considering that the dependent variable indicates the decision to opt into legal sales of medical marijuana, positive signs on average marginal effects reflect greater tolerance of medical marijuana. As expected, characteristics of community residents and municipal governments had statistically significant effects on opt-in decisions. Municipalities with larger population sizes and a higher proportion of Black residents were more likely to pass opt-in ordinances. A lower likelihood of opting into medical marijuana was observed among municipalities with older populations, which was measured as the proportion of residents at ages 65 and above.

Table 2.

Logistic Model Estimates for Municipal Opt-In Decisions on Medical Marijuana Facilities.

Independent Variables	Marginal Effect	SE
Population in 1,000s	0.001**	(0.000)
% Age 65 and older	−0.011**	(0.004)
% Black	0.002*	(0.001)
% Hispanic	0.001	(0.002)
Median household income	−0.002	(0.002)
Gini coefficient	0.005	(0.007)
Marijuana arrest rate	0.001	(0.001)
% Religious adherents	−0.003	(0.002)
% Yes vote on medical marijuana in 2008	−0.001	(0.004)
Council-manager government (0,1)	0.068**	(0.026)
City council size in 2017	−0.000	(0.011)
Female executive in 2017 (0,1)	−0.060*	(0.037)
% Females on council in 2017	−0.001	(0.001)
Sample size	533
% Correctly predicted	0.910
McFadden’s Pseudo R ²	0.180

Note. Results are based on logistic model estimates for opt-in decisions on medical marijuana. Average marginal effects are reported with standard errors approximated using the delta method shown in parentheses.

Statistical significance is denoted by * and ** for significance at the 5% and 1% levels, respectively.

Among local government characteristics, statistically significant effects were estimated for variables measuring the type of government structure and female representation. In comparison to municipalities with a mayor-council form, those with a council-manager were more likely to pass an opt-in ordinance. Female representation in local government was negatively associated with opt-in decisions, but the average marginal effect was only significant for having a female official in an executive leadership role. The proportion of women on city council was a statistically insignificant determinant.

Results for the analysis of opt-out decisions on recreational marijuana are shown in Table 3. Each column of the table provides average marginal effects and standard errors based on one of two model specifications, which differ based on whether a municipality’s earlier decision on whether to opt into medical marijuana was included as an independent variable. As with medical marijuana, the variation in adopting opt-out ordinances in Michigan could be explained by observable community and government characteristics. Since the dependent variable measured whether Michigan municipalities opted out of legal sales of recreational marijuana products, positive signs on the average marginal effects reflect greater opposition to recreational marijuana.

Table 3.

Logit Model Estimates for Municipal Opt-Out Decisions on Recreational Marijuana Facilities.

	(1)		(2)
Independent Variables	Marginal Effect	SE	Marginal Effect	SE
Medical marijuana opt-in municipality (0,1)			−0.341**	(0.064)
Population in 1,000s	0.001	(0.001)	0.001	(0.001)
% Age 65 and older	0.013**	(0.004)	0.011**	(0.004)
% Black	−0.004*	(0.002)	−0.003	(0.002)
% Hispanic	−0.004	(0.004)	−0.003	(0.004)
Median household income	0.008**	(0.003)	0.007**	(0.003)
Gini coefficient	0.010	(0.010)	0.010	(0.009)
Marijuana arrest rate	−0.001	(0.001)	−0.001	(0.001)
% Religious adherents	−0.000	(0.002)	−0.001	(0.002)
% Yes vote on recreational marijuana in 2018	−0.003	(0.005)	−0.002	(0.005)
Council-manager government (0,1)	0.167**	(0.037)	0.194**	(0.037)
City council size in 2019	0.040*	(0.018)	0.041*	(0.016)
Female executive in 2017 (0,1)	−0.005	(0.047)	−0.017	(0.044)
% Females on council in 2017	−0.001	(0.001)	−0.001	(0.001)
Sample size	533		533
% Correctly predicted	0.749		0.747
McFadden’s Pseudo R ²	0.098		0.143

Note. Results are based on logistic model estimates for opt-out decisions on recreational marijuana. Average marginal effects are reported with standard errors approximated using the delta method shown in parentheses.

Statistical significance is denoted by * and ** for significance at the 5% and 1% levels, respectively.

As shown in column 1 of Table 3, some factors that influenced medical marijuana decisions had similar effects on recreational marijuana decisions. Local policies toward recreational marijuana businesses were relatively favorable in areas that had a larger percentage of Black residents and unfavorable where there were a higher percentage of senior citizens. However, unlike the results for medical marijuana ordinances, some factors had different effects on opt-out decisions for recreational marijuana. County-level median income levels had a positive association with the likelihood of opting out, which suggests that the economic benefits of commercial cannabis might be an influential factor among jurisdictions in low-income areas.

With respect to local government characteristics, estimates suggested that municipalities with a council-manager form of government were more likely to opt out. While this finding may contradict the positive relationship between council-manager government and opt-in decisions on medical marijuana that was reported earlier, one plausible explanation could be that municipalities with a professional manager are more proactive on implementing ordinances in response to the provision of a local option by the state government. The positive effect of city council size also might reflect the possibility of quicker action among local governments with more administrative resources. In another departure from the results on medical marijuana, female representation did not appear to influence the likelihood that a municipality passed an ordinance to prohibit businesses related to recreational marijuana.

When a municipality’s earlier opt-in decision on medical marijuana was included as an independent variable, there was little change in the estimates, which can be observed in column 2 of Table 3. As expected, there was a strong correlation between decisions to allow medical and recreational marijuana. Compared to municipalities that did not opt into medical marijuana, opt-in municipalities were associated with a 34.1% pt. reduction in the likelihood of opting out of recreational marijuana. Considering that 67.9% of municipalities opted out of recreational marijuana during the time period of covered by the data, this estimate suggests that allowing medical marijuana was associated with a reduction in the likelihood prohibiting recreational marijuana by a magnitude of approximately 50.2%.

Conclusion and Discussion

The legalization of adult-use marijuana has been met with reluctance by some policymakers in early adopting states. Most of these states have given municipal governments the option to prohibit cannabis businesses in their communities. This paper examined patterns in the adoption of prohibition ordinances in Michigan with an emphasis on local resident and government characteristics in city and village governments. Although most municipalities chose to prohibit cannabis businesses, either by refraining from opting into medical marijuana or electing to opt out of recreational marijuana, it appeared the local characteristics were associated with significant variation in these policy outcomes. This study was the first to examine local option decisions on medical marijuana businesses following statewide legalization of only medical-related consumption. Another key contribution of this study was the ability to compare determinants of decisions to prohibit medical and recreational marijuana businesses by analyzing local government policies within the same state. Given the heterogeneity in the strictness of marijuana regulations across states (G. J. Kim et al., 2021; Klieger et al., 2017; Pacula & Smart, 2017), an analysis of Michigan’s local option provisions provided a clean comparison of prohibitions on commercial activity related to medical versus recreational uses.

Consistent with prior studies on public opinion, local elected officials were more tolerant of commercial marijuana activity when they represented citizens from population groups that have tended to favor marijuana legalization. The finding of a higher likelihood of prohibiting marijuana businesses in communities more senior citizens would align with patterns from earlier studies on public opinion (Ellis et al., 2019; Nielsen, 2010; Schnabel & Sevell, 2017). Communities with a higher proportion of Black residents also were more likely to allow marijuana businesses. Another noteworthy result was the positive relationship between an area’s level of income and the likelihood of prohibiting recreational marijuana, but not medical marijuana. Greater leniency toward recreational marijuana activity in communities with lower income levels may reflect the perception of greater economic gains from recreational versus medical marijuana sales, which could have a stronger influence in poorer communities. This result adds to a body of literature with mixed findings on the relationship between income levels and the location of marijuana businesses. One earlier study on the location of medical marijuana dispensaries in California found a higher density of dispensaries in low income areas (Morrison et al., 2014) while studies in two large cities found no relationship (Boggess et al., 2014; Thomas & Freisthler, 2016).

In addition to the characteristics of a community’s residents, local government characteristics also appeared to affect policy outcomes. The finding of a negative relationship between female representation in local government and medical marijuana opt-in decisions reflects past studies that observed a gender gap in public opinion with women holding less favorable views on marijuana (Denham, 2019; Elder & Greene, 2019; Galston & Dionne, 2013; Nielsen, 2010; Schnabel & Sevell, 2017; Schwadel & Ellison, 2017).⁸ However, this relationship only existed for municipalities that had a female in a position of leadership. In addition, the analysis of Michigan’s opt-out provision indicated that female representation did not have a significant effect on the likelihood of prohibiting recreational marijuana businesses. This result differs from a finding in an earlier study by Camor et al. (2020), who examined local actions to prohibit recreational marijuana businesses in Oregon. In their study, greater female representation increased the odds that a city would present an opt-out measure on an election ballot and allow residents to vote on a prohibition decision. Potential causes for these different outcomes could stem from state-specific features related to statewide regulatory systems, tax rates on cannabis products, the timing of statewide legalization, or media coverage. Cross-state comparisons may be needed to provide a better understanding of contextual variables that could interact with female representation to influence local marijuana policies.

Further examination also should be given to state disparities in the process for arriving at local policy decisions. More specifically, variation in the leadership values of local elected officials is worthy of more attention. A city council’s hesitation to opt into commercial cannabis, at least in the early stages of statewide legalization, may reflect its desire to receive public comments before taking action rather than a preference for long-term prohibition. Likewise, an immediate decision to opt out might be a plan to temporarily delay the opening of the first cannabis businesses in order to provide citizens with an opportunity to provide feedback. Disparities in the design of local option provisions may generate different decisions from city officials based on their leadership styles. For example, an opt-out decision in one state could allow residents to vote on a ballot measure prohibiting commercial cannabis whereas other states may allow city councils to take actions that supersede citizen attempts to influence policies.

Conclusions derived from the analysis of local prohibitions in Michigan will provide valuable insight for policymakers and researchers as other U.S. states decide whether to legalize adult-use marijuana in the future. However, despite the contributions of this paper, there are some limitations in the methodological design that should be considered when assessing their generalizability. Due to the cross-sectional nature of the data used in the analysis, local prohibition decisions were analyzed based on whether they were adopted by a specific date during the rollout of Michigan’s regulations on marijuana. Due to data limitations, it was not possible to observe the sequential order in which nearby jurisdictions adopted prohibition ordinances. Further work utilizing longitudinal data is needed to assess how an early prohibition ordinance by one municipality influences later decisions of other municipalities.

Along with community characteristics, other factors that could affect local policies on marijuana deserve further attention. In this study, the variation in ordinance adoption was partially explained by disparities across municipalities with respect to age composition, racial background, income level, and government characteristics. If local decisions to allow commercial marijuana facilities lead to significant changes in the ease at which marijuana can be obtained legally, subsequent changes in marijuana use might follow similar patterns across municipalities. Not only do municipalities in Michigan have the authority to limit the number of marijuana businesses within their borders, but they also can set limits on the types of businesses, which can include growers, processors, and dispensaries. Considering that local authority to restrict or prohibit commercial facilities has been a common feature in state laws since Washington and Colorado became the first states to legalize adult-use recreational marijuana in 2012, the continued inclusion of this kind of provision may lead to important regional disparities within states. Future studies should consider the implications on the accessibility, price, and consumption of legal marijuana products.

Supplemental Material

Supplemental Material, sj-docx-1-cdx-10.1177_00914509221105285 - Local Prohibitions on Marijuana: Factors Associated With Bans on Medical and Recreational Businesses

Supplemental Material, sj-docx-1-cdx-10.1177_00914509221105285 for Local Prohibitions on Marijuana: Factors Associated With Bans on Medical and Recreational Businesses by David M. Yaskewich in Contemporary Drug Problems

Supplemental Material

Supplemental Material, sj-xlsx-1-cdx-10.1177_00914509221105285 - Local Prohibitions on Marijuana: Factors Associated With Bans on Medical and Recreational Businesses

Supplemental Material, sj-xlsx-1-cdx-10.1177_00914509221105285 for Local Prohibitions on Marijuana: Factors Associated With Bans on Medical and Recreational Businesses by David M. Yaskewich in Contemporary Drug Problems

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

David M. Yaskewich

Notes

Supplemental Material

Supplemental material for this article is available online.

References

Aguilar

Murray

(2014). Colorado cities and towns take diverging paths on recreational pot. The Denver Post. https://www.denverpost.com/2014/12/19/colorado-cities-and-towns-take-diverging-paths-on-recreational-pot-2/

Beltz

Mosher

Schwartz

(2020). County-level differences in support for recreational cannabis on the ballot. Contemporary Drug Problems, 47(2), 149–164. https://doi.org/10.1177/0091450920925581

Boggess

L. N.

Perez

D. M.

Cope

Root

Stretesky

P. B.

(2014). Do medical marijuana centers behave like locally undesirable land uses? Implications for the geography of health and environmental justice. Urban Geography, 35(3), 315–336. https://doi.org/10.1080/02723638.2014.881018

Burns

(2019). Michigan recreational marijuana retail sales to begin Dec. 1. MLive.com. https://www.mlive.com/public-interest/2019/11/surprise-decision-means-recreational-marijuana-available-in-michigan-dec-1.html

Camors

C. A.

Chavez

S. L.

Romi

A. M.

(2020). The cannabis industry within the USA: The influence of gender on cannabis policy and sales. Sustainability in Accounting, Management and Policy Journal. Advance online publication. https://doi.org/10.1108/SAMPJ-12-2018-0330

Carliner

Mauro

P. M.

Brown

Q. L.

Shmulewitz

Rahim-Juwel

Sarvet

A. L.

Wall

M. M.

Martins

S. S.

Carliner

Hasin

D. S.

(2017). The widening gender gap in marijuana use prevalence in the U.S. during a period of economic change, 2002-2014. Drug and Alcohol Dependence, 170(1), 51–58. https://doi.org/10.1016/j.drugalcdep.2016.10.042

City of Grand Rapids. (2018). Data compendium for informed policy: Zoning for cannabis-related uses. https://www.grandrapidsmi.gov/files/assets/public/departments/planning-department/cannabis-data-compendium-7-18-18.pdf

Dai

Richter

K. P.

(2019). A national survey of marijuana use among US adults with medical conditions, 2016-2017. JAMA Network Open, 2(9), e1911936. https://doi.org/10.1001/jamanetworkopen.2019.11936

Daniller

(2019). Two-thirds of Americans support marijuana legalization. Pew Research Center. https://www.pewresearch.org/fact-tank/2019/11/14/americans-support-marijuana-legalization/

10.

Denham

B. E.

(2019). Attitudes toward legalization of marijuana in the United States, 1986-2016: Changes in determinants of public opinion. International Journal of Drug Policy, 71(1), 78–90. https://doi.org/10.1016/j.drugpo.2019.06.007

11.

Dilley

J. A.

Hitchcock

McGroder

Greto

L. A.

Richardson

S. M.

(2017). Community-level policy responses to state marijuana legalization in Washington State. International Journal of Drug Policy, 42(1), 102–108. https://doi.org/10.1016/j.drugpo.2017.02.010

12.

Elder

Greene

(2019). Gender and the politics of marijuana. Social Science Quarterly, 100(1), 109–122. https://doi.org/10.1111/ssqu.12558

13.

Ellis

J. D.

Resko

S. M.

Szechy

Smith

Early

T. J.

(2019). Characteristics associated with attitudes toward marijuana legalization in Michigan. Journal of Psychoactive Drugs, 51(4), 335–342. https://doi.org/10.1080/02791072.2019.1610199

14.

Everson

E. M.

Dilley

J. A.

Maher

J. E.

Mack

C. E.

(2019). Post-legalization opening of retail cannabis stores and adult cannabis use in Washington State, 2009-2016. American Journal of Public Health, 109(9), 1294–1301. https://doi.org/10.2105/ajph.2019.305191

15.

Galston

W. A.

Dionne

E. J.

Jr . (2013). The new politics of marijuana legalization: Why opinion is changing. Governance Studies at Brookings, 1–17. http://www.brookings.edu/wp-content/uploads/2016/06/Dionne-Galston_NewPoliticsofMJLeg_Final.pdf

16.

Gender

A. P. I.

(2020). Gender API. http://www.gender-api.com

17.

Grammich

Hadaway

Houseal

Jones

D. E.

Krindatch

Stanley

Taylor

R. H.

(2018). US religion census congregations and membership study, 2010 (County File). Association of Statisticians of American Religious Bodies. https://www.thearda.com/Archive/Files/Descriptions/RCMSCY10.asp

18.

Hannah

A. L.

Mallinson

D. J.

(2018). Defiant innovation: The adoption of medical marijuana laws in the American states. Policy Studies Journal, 46(2), 402–423. https://doi.org/10.1111/psj.12211

19.

Holman

M. R.

(2017). Women in local government: What we know and where we go from here. State and Local Government Review, 49(4), 285–296. https://doi.org/10.1177/0160323X17732608

20.

Kim

G. J.

Hwang

S. J.

Berry

F. S.

(2021). Explaining the strictness of medical marijuana regulations in states. The Social Science Journal, 58(2), 186–205. https://doi.org/10.1016/j.soscij.2019.04.003

21.

Kim

S. H.

(2018). Framing marijuana: How U.S. newspapers frame marijuana legalization stories (1995-2014). Preventive Medicine Reports, 11, 196–201. https://doi.org/10.1016/j.pmedr.2018.07.003

22.

Klieger

S. B.

Gutman

Allen

Pacula

R. L.

Ibrahim

J. K.

Burris

(2017). Mapping medical marijuana: State laws regulating patients, product safety, supply chains and dispensaries, 2017. Addiction, 112(12), 2206–2216. https://doi.org/10.1111/add.13910

23.

Leon

Weitzer

(2014). Legalizing recreational marijuana: Comparing ballot outcomes in four states. Journal of Qualitative Criminal Justice and Criminology, 2(2), 193–218. http://www.ssrn.com/abstract=2511034

24.

Mauro

C. M.

Newswanger

Santaella-Tenorio

Mauro

P. M.

Carliner

Martins

S. S.

(2019). Impact of medical marijuana laws on state-level marijuana use by age and gender, 2004-2013. Prevention Science, 20(2), 205–214. https://doi.org/10.1007/s11121-017-0848-3

25.

McGinty

E. E.

Niederdeppe

Heley

Barry

C. L.

(2017). Public perceptions of arguments supporting and opposing recreational marijuana legalization. Preventive Medicine, 99, 80–86. https://doi.org/10.1016/j.ypmed.2017.01.024

26.

McGinty

E. E.

Samples

Bandara

S. N.

Saloner

Bachhuber

M. A.

Barry

C. L.

(2016). The emerging public discourse on state legalization of marijuana for recreational use in the US: Analysis of news media coverage, 2010-2014. Preventive Medicine, 90, 114–120. https://doi.org/10.1016/j.ypmed.06.040

27.

Michigan Department of State. (2016). 2008 Michigan election results. https://mielections.us/election/results/08GEN/

28.

Michigan Department of State. (2018). 2018 Michigan election results. https://mielections.us/election/results/2018GEN_CENR.html

29.

Morrison

Gruenewald

P. J.

Freisthler

Ponicki

W. R.

Remer

L. G

. (2014). The economic geography of medical cannabis dispensaries in California. International Journal of Drug Policy, 25(3), 508–515. https://doi.org/10.1016/j.drugpo.2013.12.009

30.

Mosher

C. J.

Akins

(2019). In the weeds: Demonization, legalization, and the evolution of U.S. marijuana policy. Temple University Press.

31.

National Conference of State Legislatures. (2022). State medical marijuana laws. http://www.ncsl.org/research/health/state-medical-marijuana-laws.aspx

32.

Nielsen

A. L.

(2010). Americans’ attitudes toward drug-related issues from 1975-2006: The roles of period and cohort effects. Journal of Drug Issues, 40(2), 461–493. https://doi.org/10.1177/002204261004000209

33.

Pacula

R. L.

Smart

(2017). Medical marijuana and marijuana legalization. The Annual Review of Clinical Psychology, 13(1), 397–419. https://doi.org/10.1146/annurev-clinpsy-032816-045128

34.

Pew Research Center. (2019). 2019 Pew Research Center’s American trends panel. http://www.pewresearch.org/wp-content/uploads/2019/11/FT_19.11.08_Marijuana_Topline-For-Release.pdf

35.

Schnabel

Sevell

(2017). Should Mary and Jane be legal? Americans’ attitudes toward marijuana and same-sex marriage legalization, 1988-2014. Public Opinion Quarterly, 81(1), 157–172. https://doi.org/10.1093/poq/nfw050

36.

Schwadel

Ellison

C. G.

(2017). Period and cohort changes in Americans’ support for marijuana legalization: Convergence and divergence across social groups. The Sociological Quarterly, 58(3), 405–428. https://doi.org/10.1080/00380253.2017.1331715

37.

Spetz

Chapman

S. A.

Bates

Jura

Schmidt

L. A.

(2019). Social and political factors associated with state-level legalization of cannabis in the United States. Contemporary Drug Problems, 46(2), 165–179. https://doi.org/10.1177/00914550919827605

38.

Thomas

Freisthler

(2016). Examining the locations of medical marijuana dispensaries in Los Angeles. Drug and Alcohol Review, 35(3), 334–337. https://doi.org/10.1111/dar.12325

39.

U.S. Census Bureau. (2010). 2010 Decennial census. http://data.census.gov/

40.

U.S. Census Bureau. (2019). 2014-2018 American community survey 5-year estimates. http://data.census.gov/

41.

U.S. Department of Justice. (2016). FBI uniform crime reporting program data: County-level detailed arrest and offense data, 2016 [Data file and codebook]. Inter-university Consortium for Political and Social Research [distributor]. https://doi.org/10.3886/icpsr37059.v3

42.

WXYZ Detroit. (2019). Here’s a list of 1,400+ communities that have opted out of recreational marijuana. https://www.wxyz.com/news/heres-a-list-of-the-1-300-communities-that-have-opted-out-of-recreational-marijuana

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

3.80 MB

0.03 MB