Saying no to weed: Public opinion towards cannabis legalisation in Uruguay

Abstract

Aims: This research aimed to explore people’s beliefs associated with opposition to cannabis legalisation in Uruguay. In 2014, Uruguay passed a national law regulating the production, sale and consumption of recreational marijuana. However, two-thirds of the Uruguayan public expressed disagreement towards the government’s new policy.

Methods: This study used logistic regressions on data from a national probabilistic sample of 1512 adults in Uruguay. Opinions and beliefs towards cannabis legalisation were collected in face-to-face interviews, using a battery of questions included in Vanderbilt University’s AmericasBarometer national survey in 2014.

Findings: Results showed that opposition to legalisation in Uruguay is independently associated with the beliefs that the new cannabis law will worsen the public security conditions in the country, that it will serve as a gateway to the use of harder drugs, and that the law will ultimately be ineffective to curb illegal trafficking. They also showed the importance of political ideology.

Conclusions: Public views towards cannabis liberalisation are more intertwined with concerns about public security and apprehension that it will open the gate to heavier drugs than with concerns about individual health and demographic factors. The paper underscores the importance of belief systems and political socialisation over personal behaviour of use.

Keywords:

• Cannabis
• marijuana legalisation
• Uruguay
• public opinion
• drug policy

Introduction

Why do people reject cannabis legalisation? Why do persons who live in a country that has legalised recreational marijuana oppose the policy? The paradigmatic case of cannabis legalisation in Uruguay provides an excellent opportunity to explore the complexities of public opinion towards marijuana legalisation and its implications for public policy development. Uruguay is the first country in the world to fully legalise the production, sale and consumption of recreational cannabis. Moreover, the swift implementation of total regulation led by the Uruguayan national government has turned this South American country into a potential model for legalisation around the region (Hetzer & Walsh, 2014). Yet all known public opinion studies and surveys show that a majority of Uruguayans – between 60% and 65% – have consistently opposed the legalisation policy, even after it was passed (Cruz, Queirolo, & Boidi, 2016; Garat, 2015; Inunza & Pardo, 2014). These attitudes are in sharp contrast with the development of public opinion in other countries where increasing trends of public support for cannabis legalisation have been followed by modest processes of policy liberalisation (Hathaway, Erickson, & Lucas, 2007; van der Sar, Brouwers, van de Goor, & Garretsen, 2011). In the United States, for instance, the general public’s push for liberalisation via referenda has brought drug policy to the forefront of the national policy and political debate (Caulkins et al., 2014). In Uruguay, conversely, marijuana legalisation was a top-down process in which the government passed, with little public support, a very comprehensive regulatory bill on cannabis (Kilmer, Kruithof, Pardal, Caulkins, & Rubin, 2013).

There is the belief among observers and practitioners that initiatives conducive to marijuana legalisation are spurred by demands and mobilisation coming from the society. That is in part the result of studies focussed on developed countries. However, the Uruguayan case shows that legalisation is possible even when the public and legislators do not share the same priorities. In this paper, using a national survey of adults in Uruguay conducted in 2014, we explore the factors associated with the public rejection of marijuana liberalisation reform.¹ We particularly concentrate on the reasons why most citizens opposed legalisation just when the new regulation law was being enacted. More importantly, we want to understand what the beliefs that drive opposition to cannabis legalisation are, especially in a country that has decided to open the market to recreational users.

The importance of this study cannot be overstated. We need to understand the ideological frameworks that inform the debates on legalisation in countries where different rationales and political agendas are at play. In Latin America, for instance, discussions about marijuana legalisation not only involve issues of individual rights and freedoms, but also include considerations about public insecurity, the war on drugs, and institutional capability. In several contexts, the latter issues usually are viewed more critically than the former. With the discussion of cannabis legalisation policies spreading across regions, we must be able to develop a theoretical groundwork for interpreting the upcoming debates in the public opinion arena and how they shape the policy outputs. We thus examine five different beliefs or concerns regarding legalisation. First, that cannabis is harmful to health; second, that marijuana is a gateway to harder drugs; third, that legalisation will be ineffective; fourth, that marijuana users are a threat to society; and finally, that security conditions in the country will worsen with the new law. The particular case of Uruguay is appropriate to explore the relationships and contradictions between public opinion and policy and to contribute to a comprehensive approach to drug policy reform.

The legalisation process in Uruguay

Consumption of small quantities of recreational cannabis has been legal in Uruguay since 1974; however, production and sale were fully prohibited until Law 19,172 was enacted in late 2013 (Arraras & Bello-Pardo, 2014). Initiatives to legalise the retail of recreational marijuana had been on the congressional agenda since the early 2000s in attempts to reconciling the contradictions in the law that permitted consumption but banned any form of distribution (Garibotto, 2011). However, by 2011, all legalisation proposals sponsored by different political factions and supported by civil society organisations had been ultimately unsuccessful (Garat, 2015; Kilmer et al., 2013). In 2012, President José Mujica sent a bill to Congress that aimed to regulate the production, distribution and consumption of recreational marijuana that finally passed as law in the Uruguayan Senate in December 2013 (Walsh & Ramsey, 2015). Law 19,172 had been presented by the Executive as part of his Strategy for Life and Coexistence (Presidencia de Uruguay, 2012). The decision to advance the legalisation of recreational cannabis was made by the government in response to growing concerns about the expansion of organised crime in Uruguay and its repercussions on public security (Garat, 2015; Walsh & Ramsey, 2015). However, most Uruguayans did not show enthusiasm for the new regulations. Despite the role that pro-legalisation activists played in pushing for policy change, all public opinion surveys conducted throughout 2013 indicated that at least two of every three Uruguayans did not support the bill regulating the production and sale of recreational cannabis (Garat, 2015; Kilmer et al., 2013). In stark contrast to the United States, where recreational and medical marijuana legalisation initiatives have been driven by popular referenda, the majority of the Uruguayan population has remained clearly opposed to the liberalisation of cannabis, even two years after the passing of the law.²

Reasons to oppose marijuana legalisation

What factors explain the opposition to marijuana legalisation in a country that, such as Uruguay, has gone all the way to allow and regulate the use of recreational marijuana? The emerging scholarship on public opinion of cannabis legalisation provides some hints about the variables associated with the public attitudes towards cannabis liberalisation. They revolve around three types of factors: individual characteristics, relations with the drug, and political socialisation. In the first type of factors, research in the United States, Latin America, and Australia has pointed out to the relevance of age and cohort in the support for drug policy reform. Young people tend to have more positive public attitudes towards cannabis liberalisation policies than other age groups (Carroll, 2005; Cintrón & Johnson, 1996; Fetherston & Lenton, 2005; García Sánchez & Ortiz Riomalo, 2014), although these views seem to change as people age (Nielsen, 2010). By the same token, studies in the United States have shown that race and religious affiliation are strongly associated with views towards drug policies, with non-Whites and evangelical Christians less likely to support marijuana legalisation (Caulkins, Hawken, Kilmer, & Kleiman, 2012; Meares, 1997). In addition, religious involvement seems to prevent cannabis use among different social groups (Nguyen & Newhill, 2016; Ulmer, Desmond, Jang, & Johnson, 2012), which tends to accentuate the opposition views towards legalisation.

In the second group of factors, people would support or oppose marijuana legalisation because it affects them directly (van der Sar et al., 2011; Williams, van Ours, & Grossman, 2011). In several countries, people who have been exposed to marijuana or who have used marijuana and other illicit drugs would be more likely to support cannabis legalisation (García Sánchez & Ortiz Riomalo, 2014; Makkai & McAllister, 1993; Palamar, 2014). Cruz, Queirolo, and Boidi (2016) suggest that support for marijuana legalisation among users and, especially, their relatives might be driven by the expectations of avoiding problems with law-enforcement. Conversely, people with children at home would tend to oppose legalisation out of anxiety over their family’s well-being (Caulkins, Hawken, et al., 2012).

Finally, for years studies have shown the key role that political ideology plays in the attitudes towards legalisation (Caulkins, Hawken, et al., 2012; Lenton & Ovenden, 1996; Palamar, 2014). Timberlake, Lock, and Rasinski (2003) propose that ideology and political norms are part of the broader construct of political socialisation, which shapes the views towards drug policy and cannabis liberalisation. Conservatives, for example, would tend to reject cannabis liberalisation more than any other political group, whereas liberals are more likely to support it (Goode, 1998; Timberlake, Rasinski, & Lock, 2001). In 1973, Baer (1973) found that political views towards cannabis were important, but they turned out to be less significant than the personal’s history of use. The recent literature, however, has underlined the prominence of political views as the use of cannabis has become more socially acceptable (Caulkins, Coulson, Farber, & Vesely, 2012; Galston & Dionne, 2013).

Ideology and political views are thus related to how people and stakeholders construct arguments that advance or resist cannabis liberalisation policies. In the public arena, opposition to legalisation of marijuana has revolved around many different arguments. These opinions have the potential to shape the debate about legalisation, whether underpinning the coalitions in favour of liberalising reforms or hardening the positions of those who oppose them. In this paper, we concentrate on five types of arguments that were echoed in the debate about cannabis regulation in Uruguay (Garat, 2015). The first, and perhaps more widespread contention, has to do with harm. Opponents to marijuana liberalisation argue that although it may be considered a “soft” drug, cannabis is still a harmful substance, with damaging consequences for the personal health of consumers (Stimson, 2010). The range of potential adverse effects of legalising the use of marijuana cited by opponents goes from mental health to dependence syndrome, to physical health, such as respiratory illnesses (Volkow, Baler, Compton, & Weiss, 2014). In Uruguay, some health professionals, including prominent members of the ruling party, publicly criticised the initiative of legalisation, pointing out its potential health risks (Gil & Isgleas, 2013; Sociedad de Psiquiatria del Uruguay, 2012).

The second anti-liberalisation argument points to the role of marijuana as a potential gateway to the use of harder drugs, such as cocaine and heroin. It usually overlaps with arguments about harm but emphasises the influence of marijuana use in creating the conditions for later addiction behaviours (Office of National Drug Control Policy, 2008; Volkow et al., 2014). In Uruguay, opponents of the regulation in the medical community and political opposition argued that the marijuana law would increase consumption of other drugs and would saturate health services for people with addictions (Diario La Republica, 2013; Gil & Isgleas, 2013).

In several places, especially in Latin America, the debate about the potential harms of drug legalisation is intertwined with arguments regarding public security and policies to tackle crime (Nieto & Morini, 2014). While some proponents of cannabis legalisation – including the Uruguayan government – see it as a way to reduce the high human costs of the war on drugs (Global Commission on Drug Policy, 2014; Presidencia de Uruguay, 2012), opponents of cannabis legalisation are concerned that availability of drugs on the streets would increase crime and empower drug trafficking organisations (Stimson, 2010). In an analysis of media treatment of the marijuana issue, researchers from the University of the Republic in Uruguay found that the local press most frequently associated cannabis with drug trafficking and crime (Filardo, Aguiar, Musto, & Pieri, 2012).

The fourth argument behind opposition to cannabis legalisation refers to more diffuse dangers to society. Public opposition to marijuana liberalisation laws sometimes taps into the idea that cannabis consumers may be a threat to the community and its values in general. This belief is anchored in normative views about society, as some of the proponents of restricting the availability of cannabis would use religious and ideological reasons to argue that drug consumption is not only harmful to individuals but also to the family and the society overall (Bindrim, 2014). Although the growing push for legalisation has led liberalisation opponents to articulate evidence-based arguments, religious and conservative leaders in Uruguay pointed that cannabis liberalisation is morally unacceptable as it contributes to the erosion of family bonds and social values (Conferencia Episcopal del Uruguay, 2015; Larrañaga, 2013).

Finally, a frequent argument about cannabis legalisation states that a state-managed cannabis industry will be ineffective to regulate marijuana consumption, stem its illegal traffic, and protect communities from the dangers of its use. In the United States, some voices point to the fact that laws in Colorado and Washington have been unable to generate the expected results regarding harm reduction (The Gazette, 2015). In Uruguay, opposition leaders voiced their disapproval arguing that the government will not be able to regulate the cannabis black market associated with the surplus of individual home cultivation (Larrañaga, 2013).

Despite the abundance of public opinion polls about drug-related policies in countries undergoing reform, few studies have addressed the reasons why people support or reject drug policies. Notable exceptions are the works of Sznitman and Bretteville-Jensen (2015) and Caulkins, Coulson, Farber, and Vesely (2012). While Sznitman and Bretteville-Jensen found that the belief that marijuana has medical benefits is relevant for the support of legalisation of medical cannabis in Norway and Israel; Caulkins and his colleagues indicated that people might support cannabis legalisation based on judgments of its effectiveness.

We, in contrast, wanted to know whether concerns of harm, normative views or ideological positions were driving the strong opposition towards the government’s initiative in Uruguay in order to identify areas from where cannabis laws might be challenged in the future. There, cannabis legalisation raised a critical query as it broke away from the conventional view that policies and public opinion move in the same direction (Millhorn et al., 2009). The marijuana legalisation initiative in Uruguay thus provides a unique opportunity to study the reasons why people oppose legalisation, even when their government chooses to legalise the production and sales of cannabis. Still, little is known about the factors associated with the public opinion towards these initiatives, despite their importance for helping us understand how drug policy and cultural attitudes towards drug use shape each other and transform the societal behaviours around psychotropic substances. As some observers insist, popular support is not the only driver of legalisation, and liberalisation initiatives require leadership and organisation (Caulkins, Coulson, et al., 2012), and often the right political conditions (Hughes, 2009; Scheerer, 1978). However, when marijuana legalisation passes in the midst of broad public disapproval, it is important to pinpoint the areas and arguments under which cannabis laws may be challenged, especially in countries where the views towards some illicit substances and the war on drugs tend to be different from the developed North.

Methods

This research used a national survey data from the 2014 AmericasBarometer in Uruguay. The AmericasBarometer is an initiative of the Latin American Public Opinion Project at Vanderbilt University. It collects public perceptions on institutions and government performance using national probabilistic samples of the adult population of each country.³ The survey in Uruguay was conducted between March and April 2014. We worked with the LAPOP leadership to design the survey questions and a stratified probabilistic national sample to represent precisely the adult Uruguayan population. A total of 1512 face-to-face interviews were conducted. The survey included questions about opinions and beliefs on marijuana legalisation, cannabis use, political opinions, demographics and potential associated factors.

Measuring opposition to legalisation

To gauge opposition to cannabis legalisation, we asked the following question: “For weeks now, the marijuana market is decriminalised and regulated by the government. Do you strongly agree, agree, disagree or strongly disagree with this measure?” Response categories were presented on a five-point Likert scale indicating a range from strongly disagree to strongly agree. For analytical purposes, we recoded and categorised these responses into two groups: those who disagreed or strongly disagreed with cannabis regulation and the rest.

We used five different items in the questionnaire as proxies for gauging the reasons against legalisation.⁴ The following beliefs were measured on a five-point Likert scale ranging from strongly disagree to strongly agree: (1) marijuana is harmful to health; (2) marijuana is a gateway to other drugs; (3) regulation will be ineffective to curb illegal sale of cannabis; and (4) people who use marijuana are a threat to society. In addition, to explore the idea that cannabis legalisation will aggravate public security, using an item with three answer choices, we asked respondents whether public security in the country will be better, the same, or worse as a consequence of cannabis legalisation.

To determine the reasons behind opposition to legalisation in Uruguay, we first tested bivariate relations between public opposition to recreational cannabis and the different proxy items tapping the beliefs for opposition. Then, we ran a logistic regression with all the covariates taking into account the complex survey sampling design in the estimation of standard errors using the “svyset” command in Stata 14.1 (Stata, 2015). Following previous research on factors associated to support for cannabis legalisation, we conducted the multivariate analyses in three steps to identify the relevance of demographic factors, personal context factors, and the beliefs why people reject marijuana legalisation. Demographic variables included in the analyses are gender, age, level of education and religious affiliation. Factors related to the individual context of the respondent refer to the history of previous use, relatives who have used marijuana and having children. Finally, in the last step we included the variables that tap beliefs on marijuana, and that has been described above. In this group, we included political ideology to control the possible effect of political socialisation. Political ideology is understood here in the classic left-right dimension and was measured using two dummies created from a single question with a political ideology scale (see Table A1 in the Appendix). The Variance Inflation Factor (VIF) and Tolerance statistics were calculated to detect potential problems of multicollinearity in the explanatory variables. Results indicated no significant problems of multicollinearity among the variables used in the model (VIFs < 2.0; Tolerance > 0.50).

Results

Results show that 60.7% of respondents were against cannabis legalisation in 2014. As mentioned above, this figure matches other surveys that indicated majoritarian rejection towards the legalisation law, even after several months the bill was passed (Inunza & Pardo, 2014). The items designed to tap the beliefs about the consequences of marijuana legalisation also present a disapproving panorama towards cannabis regulation: 68% of survey respondents agreed that consuming cannabis was harmful, whereas nearly 71% shared the belief that marijuana was a gateway to other drugs. When asked if they believed whether people would keep buying illegal marijuana after legalisation, 69% of survey respondents answered agreeing with that idea. Also, approximately 43% of interviewees subscribed to the argument that people who consume marijuana are a threat to society. Finally, a close percentage, 44% responded that public security in the country would worsen as an effect of cannabis legalisation (see Table A2 in the Appendix).

All these opinions seem to be related. Spearman correlations between public opposition to recreational marijuana, reasons to oppose marijuana, and report of marijuana consumption are presented in Table 1. Opposition to marijuana legalisation was positively and significantly correlated with all the beliefs tested in the survey, but it was particularly strong in regards to the idea that the country’s public security situation is going to worsen as a result of the marijuana law (r_s = 0.62, p < 0.001) and that cannabis is a gateway to harder drugs (r_s = 0.56, p < 0.001). The weakest correlation, although still statistically significant, was found in the notion that legalisation will be ineffective to curb illegal cannabis retail (r_s = 0.35, p < 0.001). The results also show that opposition to marijuana legalisation and marijuana consumption are negatively correlated, although this correlation is not as strong as the other correlations (r_s = −0.35, p < 0.001). Previous history of cannabis consumption is also significantly and negatively correlated with all the beliefs that reject different aspects of marijuana, although the relations are not very strong, particularly in connection with the idea that cannabis legalisation will be ineffective to curb illegal sale of the drug (r_s = −0.11, p < 0.01).

Table 1. Spearman correlations for opposition to marijuana legalisation, cannabis consumption and beliefs about marijuana legalisation in Uruguay.

	Has consumed cannabis	Cannabis is harmful to health	Cannabis is gateway to other drugs	Legalisation will be ineffective	Cannabis consumers are a threat	Public security will worsen with the law
Opposition to legalisation	−0.346**	0.426**	0.556**	0.349**	0.498**	0.616**
Has consumed cannabis	1.00	−0.315**	−0.343**	−0.108*	−0.403**	−0.215**
Cannabis is harmful to health		1.00	0.556**	0.215**	0.478**	0.315**
Cannabis is gateway to other drugs			1.00	0.364**	0.526**	0.420**
Legalisation will be ineffective				1.00	0.258**	0.306**
Cannabis consumers are a threat					1.00	0.395**

*p < 0.05;

** p < 0.01.

To examine the importance of some beliefs in the opposition of Uruguayans to cannabis legalisation, we conducted the multivariate analyses in three steps because we wanted to understand the relationships between demographic and personal history factors before plugging the belief variables. First, we entered only the demographic variables to determine the role of gender, age, education, and religious denomination in the stance against cannabis legalisation (Model 1 in Table 2). Then, we plugged the second model with factors related to the personal context of the respondent: whether she has consumed marijuana or not; if she has relatives and/or friends who use cannabis, and whether she has children or not. Finally, in the third model, we included the belief variables to explore their relation to rejection of cannabis legalisation. We also included her political ideology in this model to control the effects of political socialisation that have been reported in previous research. All results are shown in Table 2, including t-statistics to compare across different covariates. We also included the F-adjusted test for each model, which is a goodness of fit statistic for logistic regressions using survey data (Archer & Lemeshow, 2006). The three models show a good fit, with Model 3 providing the best fit. We aim to identify the most relevant arguments associated with opposition to marijuana legalisation in Uruguay, but we do not make causal claims between them.

Table 2. Factors associated with opposition to cannabis legalisation in Uruguay.

	Model 1		Model 2		Model 3
Variables	Linearised coefficients (SE)	t-statistic	Linearised coefficients (SE)	t-statistic	Linearised coefficients (SE)	t-statistic
Demographics
Female	0.166 (0.12)	1.41	0.051 (0.13)	0.37	0.202 (0.17)	1.16
Age 26–40	0.342 (0.19)	0.18	−0.103 (0.19)	−0.54	0.213 (0.24)	0.87
Age 41–55	0.190 (0.19)	0.99	−0.222 (0.19)	−1.17	0.248 (0.23)	1.05
Age 56 and older	0.360 (0.19)	1.84	−0.345 (0.19)	−1.79	0.043 (0.28)	0.15
Secondary education	−0.853** (0.15)	−5.75	−0.725** (0.15)	−4.78	−0.689* (0.24)	−2.87
Higher education	−1.380** (0.15)	−9.25	−1.211** (0.15)	−7.76	−0.786* (0.29)	−2.66
Catholic	0.523** (0.13)	3.90	0.310* (0.12)	2.52	0.146 (0.19)	0.74
Evangelical	0.639* (0.22)	2.84	0.301 (0.23)	1.32	−0.216 (0.36)	−0.61
Atheist	−0.353* (0.14)	−2.50	−0.404* (0.15)	−2.61	−0.559* (0.21)	−2.55
Personal context
Has children			0.013 (0.15)	0.09	−0.112 (0.23)	−0.49
Has used marijuana			−1.376** (0.16)	−8.48	−0.770* (0.25)	−3.10
Relatives/friends consume marijuana			−0.510** (0.13)	−3.78	−0.206 (0.19)	−1.10
Beliefs
Left-of-centre ideology					−0.005* (0.002)	−2.51
Right-of-centre ideology					0.006* (0.002)	2.59
Cannabis harmful to health					0.306 (0.20)	1.56
Cannabis as gateway to other drugs					1.104** (0.18)	6.20
Law will be ineffective to curb illegal sale					0.686** (0.17)	4.08
Cannabis consumers are threat to society					0.402* (0.16)	2.44
Security will worsen with legalisation					2.456** (0.20)	11.99
Constant	0.749 (0.22)	3.39	1.758** (0.25)	7.10	−1.154* (0.48)	−2.41
N	1.489		1.465		1.213
F-adjusted test (Prob > F)	0.972 (0.474)		0.897 (0.560)		0.627 (0.769)

*p < 0.05;

** p < 0.001.

In the first model, results did not show substantial differences in the way opposition to marijuana legalisation is articulated regarding gender and age, not even when we included age groups in the analysis. Only respondents older than 56 years showed slightly more opposition to cannabis law in relation to the reference group (respondents from 18 to 25 years), but such difference is not statistically significant. When we entered the context factors (Model 2) and the belief variables (Model 3) in the analyses, no age group showed statistically significant differences in terms of opposition to legalisation. However, the variable of education turned out to be statistically significant when associated with opposition to regulation in all three models. Grouped in three different educational groups, results indicate that people with secondary and higher education are less likely to oppose legalisation than the rest of Uruguayan population. In fact, having some level of education, especially higher education reduces the likelihood of opposing the marijuana law. Religious denomination turned out to be a significant factor associated with views opposing legalisation when context and belief variables are absent. Self-defined Catholics and Evangelicals are more likely to reject legalisation than the rest of the population; only people who self-proclaimed atheists are less likely to oppose cannabis regulation.⁵ However, when we included other covariates in Models 2 and 3, linearised coefficients (ant the t-statistic) for the Catholic and Evangelical groups were considerably reduced, and the Evangelical group variable became statistically non-significant when beliefs variables were added. Holding constant all other covariates, atheists remained significantly and negatively associated with rejection of cannabis legalisation.

Immediate context variables returned interesting results regarding opposition to marijuana legalisation in Uruguay. Being a parent does not seem to be statistically related to opposition to marijuana legalisation (t-statistic = −0.013). However, previous personal experiences with marijuana yielded significant and substantive support for marijuana legalisation (t-statistic = −8.48). The likelihood of opposing legalisation also decreased when the respondent alluded to relatives and close friends who consume marijuana, but the relationship is a bit weaker (t-statistic = −3.78) in comparison with reports of personal use. In other words, Uruguayans who have used marijuana and who are closely related to other users are considerably less interested in keeping its prohibition than the rest of the population. However, when belief covariates are introduced in Model 3, linearised coefficients of history of marijuana consumption and having relatives who consume cannabis shrank to the point where the coefficient of having close people consuming cannabis lost its significance. Although the personal history of consuming marijuana remains significant in the complete model, these results suggest that the presence of relatives and friends who consume marijuana is not as important as the personal history of use when different beliefs towards marijuana are accounted for.

As the literature has indicated, other than the variable about prior experience with marijuana, some of the most important factors in explaining views towards marijuana legalisation refer to political socialisation variables. Our results show that Uruguayans were sharply divided along ideological lines over the cannabis legalisation initiative. Not surprisingly, people with right-of-centre political views rejected legalisation more frequently than the rest of the population, whereas people who identified with the political left tended to show less opposition to legalisation than citizens in the centre of the political spectrum. Political ideology remained relevant in relationship with opposition to cannabis legalisation even as we added the belief variables, but the results in Model 3 indicate that most tested beliefs are more strongly associated with rejection of the cannabis law.

Regarding the arguments and beliefs around the use and legalisation of marijuana, two ideas seem to be especially powerful when it comes to opposing marijuana law. First, that public security conditions in the country will worsen as a result of cannabis legalisation (t-statistic = 11.99); and second, that marijuana is a gateway to other drugs (t-statistic = 6.20). Less substantive but still significant are the beliefs that legalisation will be ineffective to curb illegal sales (t-statistic = 4.08) and that cannabis consumers are a threat to society (t-statistic = 2.44). The belief that cannabis use will be harmful to people’s health returned no substantive or statistically significant relationship (t-statistic = 1.56). The views that public security will deteriorate as a result of cannabis legalisation and that marijuana consumption opens opportunities to harder drugs are more strongly associated with the public rejection of legalisation than any other factor tested in the model, including education, previous use of cannabis, and political ideology. In other words, opposition to the regulation of recreational cannabis in Uruguay seems to be filled with people’s fears that marijuana legalisation may jeopardise people’s security and that its liberalisation may lead to serious drugs.

Recent studies examining the Uruguayan legalisation process have underscored the role of President Mujica in the passing of Law 19,172 (Arraras & Bello-Pardo, 2014; Valdomir, 2015). Using the AmericasBarometer 2014, Cruz et al. (2016) found that the President’s popularity was the most important factor connected to positive views towards the law. After looking at the relevance of the belief variables and ideology, we tested whether the public opinion towards the President trumped the beliefs associated with opposition to legalisation. We reran the complete model including the views about the performance of President as a proxy of his popularity.⁶ We found that although the President’s job approval came up as an important and significant factor negatively associated with opposition to legalisation (t-statistic = −5.41), people’s concerns about public security (t-statistic = 10.42) and the belief that marijuana will lead to harder drugs (t-statistic = 5.64) remained strongly and significantly associated with opposition to marijuana law.

Discussion

The recent push for medical and recreational marijuana legalisation in North America, Europe, and Australia suggests that anti-prohibitionist movements have succeeded in persuading part of the public about the benefits of liberalising the distribution and consumption of marijuana (Caulkins et al., 2014; Galston & Dionne, 2013). Uruguay provides an interesting counterintuitive case nonetheless. It shows that despite the government’s decision to allow the production and access to different forms of cannabis, most people remain steadfastly opposed to marijuana legalisation (Inunza & Pardo, 2014; Walsh & Ramsey, 2015). Why? What reasons lie behind disapproval of legalisation?

Assuming that people are increasingly tilting towards legalisation, the emerging literature has focussed on the individual factors furthering the case for legalisation – such as personal history of use – and has neglected to keep an eye on the other side of the coin. Our research has explored the variables and beliefs linked to opposition to recreational cannabis legalisation in a context where groundbreaking liberalisation is already taking place. It shows that opposition to cannabis legalisation is closely connected with concerns about security despite the efforts of President Mujica’s government to present cannabis legalisation as a strategy to tackle insecurity and reduce harm (Presidencia de Uruguay, 2012). Moreover, the findings indicate that cannabis legalisation is viewed by most Uruguayan respondents as a potential contributor to insecurity rather than a solution to it. They also suggest that public rejection of marijuana liberalisation is closely related to concerns that cannabis consumption would lead to the use of harder drugs.

These concerns do not seem to be driven by apprehension towards the effects of consumption on personal health, which in the multivariate analysis remained statistically non-significant. Rather, they are related to the perceived ineffectiveness of the new regime not only to reduce insecurity but also to curb illegal sales of the drug. This has relevant implications for the design of drug policies in Latin America, where the discussion about legalisation is driven by concerns of public insecurity generated by criminal organisations (Dammert, 2009; Nieto & Morini, 2014). They suggest that proponents of legalisation in Uruguay – and perhaps in other places in Latin America where support of cannabis legalisation is scarce (Cruz et al., 2016) – would be more successful swaying the public by addressing the anxieties on insecurity generated by liberalisation than by tackling the issue of health. In the case of Uruguay, although the government initially presented the law initiative as a strategy to confront rising crime, not much discussion about the issue of crime followed after the law was approved.

The findings also reveal a strand in the belief system that indicates the presence of moral concerns about legalisation. Studies in the United States have shown that these anxieties may be linked to religious views about the world (Stylianou, 2004). In the Uruguayan case, our findings show that religious affiliation plays a limited role in the public disapproval towards legalisation, especially when beliefs and context variables are considered. Moral concerns thus have an independent relation with attitudes towards legalisation, although they are not as prominent as the apprehensiveness towards security or political ideology seem to be. In fact, in opposing legalisation, the effects of Catholic and Evangelical affiliations are significantly reduced when the history of consumption is introduced, not when beliefs of moral threat or harm are entered in the model.

Our exercise also confirms in Uruguay what the literature has pointed out elsewhere: political ideology is a key factor regarding attitudes towards cannabis-related policies (Palamar, 2014; Timberlake et al., 2001). In Uruguay, political identities played a major role as cannabis liberalisation was part of the ambitious liberal political agenda of President Mujica, which also included abortion law and equal marriage (Bidegain Ponte, 2013). Hence, opposition towards legalisation was strongly manifested by people on the political right, whereas support came from people aligned to the left. In Australia, Matthew-Simmons, Sunderland, and Ritter (2013) found that attitudes towards drug policies are part of ideological systems in which prohibitionist attitudes were closely related to support for the use of the criminal justice system to tackle drug issues. In our findings, the weight of security and moral concerns in relation to legalisation may indicate the role of those ideological systems in the responses towards cannabis liberalisation, but further research is necessary to understand how ideology systems work in full-fledged legalised regimes.

Finally, our Uruguayan data confirm the findings of several studies regarding the importance of previous history of cannabis use in the support for legalisation (Lancaster, Sutherland, & Ritter, 2014; Trevino & Richard, 2002; Williams et al., 2011). However, our figures suggest that beliefs that cannabis is a gateway to harder drugs and that security will worsen with legalisation trump the effect of knowing that relatives consume marijuana. The presence of those beliefs in the model also reduce the statistical significance of the effect of previous cannabis use. In other words, in general, belief systems are probably more critically related to the attitudes towards regulation than the variable of having experienced with it.

In any case, from this study emerges a picture in which many Uruguayans are rejecting cannabis legalisation primarily not because they are worried about health-related risks but because they are concerned about the situation of security in the country as a result of liberalisation. The findings also show that disapproval to the new regulation of cannabis is closely related to lower education and absence of previous personal experimentation with marijuana. However, some of the most important factors for predicting negative attitudes towards marijuana regulation refer to ideological belief systems, which ultimately refer to political socialisation. As shown throughout this paper, rejection of cannabis regulation was strongest among conservatives and people who view liberalisation as a threat to the overall political community. This may be an expected reaction to a policy that was actively pursued by a liberal administration, but it highlights the weight of notions alien to the discussion of drug liberalisation. Although the Uruguayan government succeeded in establishing a full regulatory regime by appealing to problems of public security and personal well-being, it seems to have failed in generating public acquiescence around the appropriateness of such policy. In the end, this may turn out to be a liability for future government administrations seeking to enforce the measures designed to reduce harm to public health and insecurity.

The results of this paper also hint at the importance of conducting research on cannabis liberalisation and drug policy in Latin America and the Caribbean. In countries such as Colombia, Jamaica, Guatemala, and Mexico, where there is an enduring discussion about marijuana legalisation, the debate about drug policy change frequently gravitates towards issues of criminal justice and the war on drugs. Research on marijuana legalisation cannot ignore people’s concerns about security and its relationship with the political context. Many citizens may support drug policy change and marijuana liberalisation as a matter of individual rights, but many others may oppose out of moral panic towards drugs, as it was the case in the United States during the 1980s (Reinarman & Levine, 2004). In sum, we need to understand how people in other countries construct their views on drug policy and how these perspectives shape the policy debate and the actual legislation in the rest of the world.

Declaration of interest

This paper is part of a research funded by the Open Society Foundations (Grant id: OR2013-10585). There are no conflicts of interest posed by the authors and the submission of this manuscript.

Notes

Notes

1. This article focuses on one of the studies conducted by the Latin American Marijuana Research Initiative (LAMRI), a project with the objective of monitoring regulation of the cannabis market in Uruguay.

2. A survey commissioned by the Latin American Marijuana Research Initiative (LAMRI), conducted in October 2015, found that only 37.4% of respondents agreed with the new cannabis legislation.

3. For more information about the methodology of the AmericasBarometer and the questionnaire used in each country, visit: http://www.americasbarometer.org/.

4. See the Appendix for the specific wording of items in the questionnaire and their treatment in the statistical analyses.

5. We created dummies for each denomination and used the group of “No religion” as reference group.

6. Not shown here but available upon request.

Appendix. Variables used in the analyses

Table A1. Variables and items in the AmericasBarometer 2014 Survey.

Personal variables	Items in AmericasBarometer 2014	Values
Opposition to cannabis legalisation	For weeks now, the marijuana market is decriminalised and regulated by the government. Do you strongly agree, agree, disagree or strongly disagree with this measure?	Recoded to:Agree or strongly agree = 0Disagree or strongly disagree = 1
Religious denominations (different dummies)	What is your religion, if any? [If the respondent says that he/she has no religion, probe to see if he/she should be located in option no religion or option atheist]	No religion = 0Catholic = 1Evangelical = 1Atheist = 1
Respondent has children	Do you have children? How many?	No = 0Yes = 1
Have tried marijuana?	Some people use or have at least experimented with marijuana. Have you ever tried marijuana? (1) No, never; (2) Yes, one time; (3) Yes, two or more times	No = 0Yes = 1
Family or friends consume marijuana	Do you have friends or relatives who consume marijuana?	No = 0Yes = 1
Political ideology: Dummy for Left (1–4) Dummy for Right (7–10)	On this card there is a 1–10 scale that goes from left to right. One means left and 10 means right. Nowadays, when we speak of political leanings, we talk of those on the left and those on the right. In other words, some people sympathise more with the left and others with the right. According to the meaning that the terms “left” and “right” have for you, and thinking of your own political leanings, where would you place yourself on this scale?	Left = 1No left = 0Right = 1No right = 0
Cannabis is harmful to health	Now I 'm going to read some statements that reflect different views on marijuana, and I will ask you to tell me to what extent you agree or disagree with these statements. We will use again the 1–7 scale, where 1 means “strongly disagree” and 7 “strongly agree”. You may also indicate any intermediate score. Marijuana is harmful to health. To what extent do you agree or disagree with this statement?	Agree = 1Else = 0
Marijuana as gateway	Marijuana is the gateway to other drugs such as cocaine, LSD or pasta base	Agree = 1Else = 0
Law will be ineffective to curb illegal sale	Although regulated, people will continue to buy illegal marijuana	Agree = 1Else = 0
Cannabis consumers are threat to society	People who use marijuana are a threat to society	Agree = 1Else = 0
Security will worsen with legalisation	In terms of public security, as a result of this law, will the country be better, the same or worse than now?	Country will be worse = 1Else = 0

Table A2. Descriptive summary of the variables.

Variables	% (Counts)
Female	52.5 (795)
Age groups
16–25	15.5 (234)
26–40	25.9 (392)
56–65	16.1 (244)
66+	17.2 (250)
Education
Primary	29.5 (439)
Secondary	49.1 (742)
Higher	21.4 (323)
Religious affiliation
No religious affiliation	27.9 (422)
Catholic	35.3 (534)
Evangelical	9.4 (142)
Atheist	14.9 (225)
Respondent has children	75.7 (1.144)
Respondent has consumed cannabis	22.1 (334)
Relatives/friends have consumed cannabis	48.6 (724)
Left-of-centre ideology	41.4 (577)
Right-of-centre ideology	25.6 (357)
Agreed Cannabis harmful to health	68.3 (994)
Agreed Cannabis is gateway to other drugs	70.7 (1.030)
Agreed Law will be ineffective to curb illegal sale	69.0 (1.000)
Agreed Cannabis consumers are threat to society	43.4 (635)
Believe Security will worsen with legalisation	44.0 (640)