Balancing conflicting interests: stakeholders’ interpretations of ‘moderation’ in Swedish gambling advertising legislation

ABSTRACT

On January 1st, 2019, a license-based system allowing online commercial companies to operate and advertise in the Swedish market was introduced. According to the Gambling Act, advertising should be ‘moderate’. Legislative changes are windows of opportunities where stakeholder interpretation and adaption may result in new agenda-setting based on conflicting interests. This article examines how stakeholder groups initially perceived and interpreted this key concept of moderation. Data from semi-structured interviews, seminars and press conferences, as well as policy documents and statements, were analyzed using a framework method approach. Contrary to what is stated in the legislation, stakeholders defined ‘moderation’ in terms of both content and volume. Although industry and media representatives requested further guidance for interpreting ‘moderation’, they favored self-regulation and allowing market consolidation. There was a notable ambivalence within and between stakeholder groups concerning balancing freedom of expression, market protection, consumer protection and public health. In conclusion, legislation is a blunt steering instrument when key concepts are largely left to stakeholders, with opposing interests, to interpret. Our study shows the incremental and interactive character of policy-making, where new legislation resulted in new policy responses. Case law and market consolidation take time and regulatory interventions need to take this into account.

KEYWORDS:

• Gambling advertising
• moderation
• online gambling
• Sweden
• gambling regulation
• consumer protection

Introduction

Gambling has been associated with both pleasure and tragedy throughout human history. Potential hazards related to gambling, such as fraud, money laundering and other criminal activities (Binde et al., 2022; Spapens et al., 2009) as well as grave social (Wardle et al., 2019) and economic (Ladouceur et al., 1994) consequences for individuals and their families (Kalischuk et al., 2006) have often led to specific regulations. Public health concerns over the social and economic impacts of gambling (Hilbrecht et al., 2020; Marionneau et al., 2023) have increased research into gambling regulation and policy addressing these issues (Egerer et al., 2018; Gainsbury & Wood, 2011; Hellman et al., 2017; Jensen, 2016; Rolando et al., 2020; Rossow & Hansen, 2016; Selin, 2016; Sulkunen et al., 2018). Since decisions to gamble are likely to be influenced by a variety of cognitive biases and illusions which marketing may trigger or reinforce (Binde, 2009; Binde & Romild, 2019; Goodie & Fortune, 2013; Landman & Petty, 2000; Lopez-Gonzalez et al., 2020) gambling regulations typically include marketing rules. The purpose of such regulation is usually to ensure that gambling offers are not pushy, do not exaggerate the chances of winning, and do not in any other way inappropriately entice people to gamble.

A main concern for policymakers has been that massive marketing will contribute to overconsumption of gambling, especially among younger persons (Buil et al., 2015; Labrador et al., 2021; Monaghan et al., 2008; Planzer & Wardle, 2011; Thomas et al., 2023). In recent years, such concerns have led to stricter marketing rules in addition to applicable national government regulations and international laws. Within the EU, there are guidelines for advertising and marketing communications (CEN, 2011; European Commission, 2014, 2021; International Chamber of Commerce [ICC], 2018) as well as an industry code of conduct (European Gaming and Betting Association [EGBA], 2020). However, each country has its own regulatory framework for gambling marketing (Sant, 2016).

Gambling advertising is a relatively well-researched topic, with most studies focusing on how, and to what degree, such advertising may contribute to harmful and excessive gambling (for overviews, see: Binde, 2022; Newall et al., 2019; Parke et al., 2014). Nevertheless, there are very few empirical studies on how marketing regulation is interpreted and implemented. To our knowledge, there are only a handful of previous studies on interpretation of marketing regulation, focusing on Romania (Simion & Dumitru, 2018), Spain (Lopez-Gonzalez & Griffiths, 2016) and online marketing in Europe (Hörnle et al., 2019). There are also a few studies that address moderation in gambling advertising to some extent, for example concerning social media (Gainsbury et al., 2015; Lindeman et al., 2023) or with a gender focus (Goedecke, 2022; Kroon, 2022). Additionally, overviews and analyses of gambling advertising regulation have appeared in reports from authorities and independent research companies (European Commission et al., 2019; Regulus Partners, 2019).

In Sweden, a new Gambling Act was adopted on January 1st, 2019 (SFS, 2018:1138) introducing a new license-based system allowing online commercial companies to operate in the Swedish market. With these changes in the gambling legislation and the re-regulation of the gambling market, Sweden makes a pertinent case for studying stakeholders’ views on gambling marketing regulation and policy change. In this way we aim to highlight the interactive character of policy processes and how stakeholders respond, reposition themselves and exploit a ‘window of opportunity’ following the introduction of new legislation. According to Kingdon (2014), windows of opportunity for policy change open when a defined problem, an available solution and political motivation with adequate sources, coincide. In this article we examine (i.e. by interviews, policy documents and statements, as well as seminars and press conferences) how stakeholders, at the time the legislation entered into force, interpreted the moderation concept in the legislation and thereby becoming de facto policymakers.

The Swedish case

From 1974 to 2019, the Swedish gambling market was divided between the state (lotteries, sports betting and number games, and later EGMs and casinos), the equestrian sports sector (horse betting), and nonprofit organizations and charities (lotteries and bingo). From around 2010 onwards it became increasingly clear that regulations had not kept pace with the digitalized and international gambling market, with foreign-based online companies operating in the Swedish market. These companies eventually gained the largest share of the online market with massive advertising aimed at Swedish consumers, primarily via television channels broadcasting from other countries and thus outside the reach of Swedish legislation.

During this period the gambling market expanded rapidly. For example, gross investments in advertising amounted to about EUR 115 million in 2010. As gambling companies prepared for the new legislation, advertising expenditure reached EUR 655 million by 2018, after almost doubling between 2016 and 2018 (Swedish Gambling Authority, 2023). Most of the advertising was from online gambling companies without a license in Sweden targeting the Swedish market (Swedish Gambling Authority, 2019b). Political debates and several inquiries (SOU, 2006: 64; 2006: 11; 2008, 124; 2015, 34) were accompanied by increasing popular discontent. Surveys showed that public confidence in gambling companies was low, especially among non-gamblers, and there was a widespread certainty that advertising for online gambling was dangerous and harmful (Kroon & Lundmark, 2020).

After the introduction of the new Gambling Act on January 1st, 2019, the market was divided in three different sections; one in which the state kept their exclusive rights; one for the non-profit/public benefit sector (lotteries and bingo); and a new section open for all gambling companies that meet the requirements (Swedish Public Health Agency, 2021). The legislation had the explicit goal of attracting at least 90% of all actors within two years. By early 2019, there were 72 license holders in total (Swedish Gambling Authority, 2019a). At the same time, license holders’ expenditures on advertising decreased substantially. However, the volume of advertising exploded when previously purchased campaigns were now being made public (Swedish Agency for Public Management, 2019, 2020).

Chapter 15 of the Gambling Act concerns marketing of gambling products. As in the previous legislation, marketing may not target children below 18 years of age and this ban was now extended to also include persons enrolled in the national self-exclusion registry. Gambling companies may not direct marketing messages toward persons who have terminated their accounts unless they have actively consented to this. Moreover, all advertising must include the age limit for gambling and contact details of an organization providing information and assistance in relation to problem gambling (§§1,2,3).

These requirements were relatively straightforward, but the main challenge was how to interpret the primary legal requisite that advertising should be ‘moderate’ (§1). This concept was first introduced as an amendment to the Lottery Act (SFS, 1994: 1000, §47a) in 2017 and adopted without any significant modification in the new Gambling Act of 2019. The preliminary work on the new Gambling Act (Prop., 2017, 220; 2016, 8; SOU, 2017, 30) included deliberations about the risk that harsh restrictions on gambling advertising would make commercial gambling companies less inclined to seek licensing. That would, in turn, endanger one of the main aims of the new regulatory system, namely high coverage and thereby increased consumer protection and prevention of problem gambling. Hence, restricting the volume, rather than only the content, of gambling advertising, was ruled out by lawmakers. Nevertheless, laws that are poorly designed or not implemented as intended can harm vulnerable populations (Gostin et al., 2019). In this way, analogous to developments in the alcohol area (Noel et al., 2017), policy-making was largely left to the industry and other stakeholders, who were expected to define ‘moderation’ and monitor adherence to self-defined guidelines. The purpose of this study is, then, to explore stakeholders’ perceptions and interpretations of ‘moderation’ in Swedish gambling advertising legislation, at a time when the gambling market was under reconstruction.

Methods

Data material was collected from different sources during the first years after the new legislation was introduced (2019–2022). Semi-structured interviews were performed in close proximity to the introduction of the new legislation, i.e. during June-August 2019. In addition, policy documents and statements, as well as seminars and press conferences were reviewed in order to follow developments regarding stakeholder interpretations of ‘moderation’.

Stakeholders

Stakeholders are defined as actors that, because of their position, could have an active or passive influence on the process of decision-making and implementation (Varvasovszky & Brugha, 2000). Interviewees were chosen within the stakeholder groups based on their position in the company or organization. That is, spokespersons or other representatives, such as general secretaries or marketing managers. Participants were recruited and informed of the research purpose via e-mail. Based on previous knowledge of the gambling advertising field, we identified four stakeholder groups: Government and authorities, Gambling industry, Media and advertisers, and Mutual support associations for persons subject to negative consequences of gambling. The stakeholder groups included in this study are presented in further detail below.

1. Government and authorities encompassed the Department of Finance, in which the responsible minister for gambling regulation was placed. This category also included the Swedish Gambling Authority, the main supervisory authority for all gambling issues. Lastly, we included the Swedish Consumer Agency, which holds main responsibility for advertising supervision. Hence, these authorities have a shared responsibility for supervision of gambling advertising.

2. Industry representatives included the Swedish Gambling Association (SPER) and the Swedish Trade Association for Online Gambling (BOS). The former represented 75% of the gambling market in Sweden and the latter represented online gambling companies and game developers in Sweden. In addition, four gambling companies were interviewed – two newcomers and two state-owned or state-controlled companies with a long history of operating in Sweden.

3. Media and advertisers were represented by one large commercial TV company, the Swedish Media Publishers’ Association, and one large tabloid newspaper controlling content, distribution and placement of advertisements. Moreover, this category included the Association of Swedish Advertisers, which counts 18 of Sweden’s 20 largest advertisers as members.

4. Mutual support associations consisted of four different national associations that offer support for persons subject to negative consequences of gambling. These associations offer counseling, group sessions and financial advice – often provided by former gamblers.

Data collection and analysis

After reviewing all publicly announced and/or accessible seminars and press conferences during the studied period (2019–2022), we included in our analysis a selection of six seminars and three press conferences that were representative of the respective stakeholder groups. Based on the same data collection strategy, we also collected all relevant policy documents and statements, expressing stakeholders’ groups official positions. We conducted 12 semi-structured interviews within all stakeholder groups, (one from government, six from industry, one from media, and four from mutual support groups), either face-to-face (n = 9), via Skype (n = 2) or via telephone (n = 1). Because interpretation and implementation of new legislation was primarily left to industry stakeholders, we chose to perform relatively more interviews within this stakeholder group. Moreover, the re-regulation brought new and more diverse actors into the Swedish gambling market and we suspected there might be a greater heterogeneity in their views. The interviews, ranging between 60 and 90 minutes, were recorded and transcribed. During the semi-structured interviews, we used a topic guide which included questions related to the gambling re-regulation, consumer protection measures, ongoing debates regarding gambling advertising and, specifically, how stakeholder groups interpret moderation. The same topics were explored for the other data sources.

A Framework method was used due to its systematic and flexible approach to analyzing collected qualitative data (Gale et al., 2013). The process started by reading and writing initial impressions, before coding and labeling. Transcripts of interviews, seminars and press conferences, as well as policy documents were coded and grouped according to subject similarity. Coding was reviewed by two of the authors. Codes were grouped first into categories and then themes forming a ‘coding index’ that was used to organize the data. An analytical framework was developed by the authors from the coding index. The final framework had 12 categories gathered into three themes that summarized the data and included references to relevant quotations (Ritchie et al., 2003). A selection of these quotes is included in the results.

Ethical approval

The content of this article does not involve the collection or handling of sensitive personal data, such as information related to individual gambling problems. In accordance with Swedish legislation, ethical approval was therefore not required from the Swedish Ethical Review Authority.

Results

The results are organized in three main themes: First, the issue of how ‘moderation’ should be interpreted, as a question of content or volume. Second, stakeholders’ descriptions of forms and delimitations of gambling (advertising) regulation. Third, stakeholders address the subject of protection, especially between consumer protection and freedom of commercial speech in a legalized market.

Theme 1: moderation – content or volume?

Following the re-regulation in 2019, the number of actors in the Swedish gambling market increased substantially and, consequently, so did the amount of gambling advertising. This notable increase was heavily criticized by the general public as well as central politicians. Stakeholders in our study generally acknowledged that the law only regulates content, but many expressed doubts about the practical interpretation of moderation in terms of both volume and content (IN 2, Mutual support association; IN 8, Media; IN 3, Industry; IN 4 Industry; IN 9, Industry).¹Stakeholders in all groups conveyed a general annoyance and frustration over constant advertising exposure, agreeing that gambling advertising in traditional media channels had been too aggressive in amount and tonality (SP 1, Government and authorities, IN 3, Industry; IN 4, Industry; IN 7, Industry; IN 8, Media; SP 4, Media, IN 5, Mutual support association; IN 1, Mutual support association). Advertising fatigue began due to intensive marketing, and media stakeholders pointed out how it could create problems for other non-gambling industries (DO 5, Media; SP 4, Media).

Each individual gambling advertisement might be moderate, but after an evening in front of a commercial TV channel, the overall impression of all gambling commercials will be that it is not moderate (IN 4, Industry)

The gambling companies’ intensive marketing also has a negative effect on the conditions for many other advertisers […]. It reinforces advertising fatigue that is already high in Sweden (SP 4, Media)

Stakeholders from the media and mutual support associations were also in agreement about how confusing it was to interpret the concept of moderation. Mutual support associations were concerned about advertising volume and how to decrease problem gambling at the individual level (IN 1, Mutual support association; IN 2, Mutual support association; IN 5, Mutual support association; IN 6, Mutual support association) and both stakeholder groups colorfully illustrated the challenge of defining the limits of moderation.

I usually use the example of if you and I were to go to a buffet and we agree in advance that we are going to eat moderately. The portions could differ greatly. Moderation from your perspective and from mine. It’s a very loose concept and I don’t understand why it was chosen from the beginning (IN 2, Mutual support association)

[Moderation] – what the hell is that? […] The three of us, for example, can define moderation differently. […] I live 70 km from the city. For me, moderation is driving maximum 150 km per day. There are many people who drive far less and who would find it highly problematic if they had to drive that far (IN 8, Media)

Stakeholders, particularly from the industry group, were critical toward the lack of guidance from legislators and authorities regarding interpretation in practice (IN 4, Industry; IN 3 Industry; IN 10, Industry). Instead, industry stakeholders turned to the Marketing Act (SFS, 2008: 486), guidelines provided by BOS and SPER, and their own internal guidelines (IN 7, Industry; IN 9, Industry; IN 12, Industry) for support. Media and industry stakeholders were above all awaiting the development of case law, as well as decisions made by responsible authorities. In this way they acknowledged that the regulation process was yet incomplete and would need more time before determining a proper interpretation (IN 3, industry; IN 4, Industry; IN 8, Media).

The concept of moderation is governed by law, and it only regulates content. […] The legislature says that moderation, whatever it is, will crystallize through case law and thus precedent is created. What we mean is moderate content, that’s exactly what BOS and SPER have jointly developed in these marketing guidelines. There’s our answer to what moderate content is (IN 3, Industry)

It takes a while before [new legislation] becomes apparent. Precedent needs to be established [referring to case law] (IN 8, Media)

In general, government and authority stakeholders emphasized that the legal definition of moderation is clearly about the content and not about the volume. Still, they were also anticipating future court rulings for further guidance (IN 11, Government and authorities; DO 2, Government and authorities).

It was also clear [at information meeting] that moderation does not refer to volume and one might expect a larger volume of advertising. So, it was pretty clear, or at least we’ve been clear, that moderation, it’s not about the amount (IN 11, Government and authorities)

A court ruling regarding the interpretation of moderation was made by the Swedish Patent and Market Court, prohibiting a gambling company’s advertising. The court based their decision on examples of the guidelines from the European Commission (2014/478/EU) and guidelines developed by BOS and SPER. The court decided that the particular type of advertisement assessed was perceived as intrusive and therefore contrary to the requirement for moderation.

A decisive factor in assessing a marketing measure is how the presentation is perceived by an average recipient in the target group to which the marketing is directed. […] the moderation requirement in the Gambling Act is, unlike the more general provisions in the Marketing Act, a specific protective law whose stated purpose is to counteract gambling problems (DO 12, Government and authority)

Mutual support associations deviated from the other stakeholders in our study by not calling for further guidance and specifications. In their view, gambling advertising is by default immoderate and industry will always try to challenge the legal limits and spur each other on.

It only takes one company advertising in a certain way that challenges the limits [.]. And then the next company will follow, and the whole thing escalates. And that’s exactly what happened (IN 2, Mutual support association)

Only a few months after the law had been introduced, the government initiated a new investigation that was to suggest further regulations of gambling advertising aimed at limiting both volume and content (Dir, 2019: 18).

Theme 2: regulation – who, how and when?

The second theme in our analysis includes stakeholders’ views on different aspects of regulation in connection with gambling advertising. That is, issues related to how advertising should be regulated, by whom and in what way. All stakeholders acknowledged the need to regulate gambling advertising in some manner. However, opinions differed regarding the scope and form of such control. Overall, three sub-topics emerged within this theme: 1) a need for new and more restrictive legislation, 2) variants of self-regulation and responsibilization, and 3) expectations of a consolidation of the market over time.

The first topic within this theme concerned legislative amendments. In terms of content, a public inquiry (Dir, 2019: 18) was initiated shortly after the new law was introduced with the aim of proposing legislative amendments to enable further limitations on gambling advertising. The inquiry concluded that it was not the right time to introduce a requirement for so-called particular moderation, which would entail an even stricter regulation of content similar to corresponding regulation in the alcohol area. Instead, the proposal was to limit the amount of advertising for high-risk games via television broadcasts (SOU, 2020: 77). Nevertheless, in a subsequent ministerial memorandum, it was suggested that the Gambling Act’s call for moderation in fact should be changed to a requirement for ‘particular moderation’ (Fi, 2021/02357). This amendment proposal was, incidentally, rejected by the present government and then approved by the Swedish parliament.

While stricter legislation – even a possible ban on gambling advertising – was being debated by stakeholders and in mass media, mutual support associations expressed little or no faith in gambling companies’ ability to adhere to legislation at all. In their view, gambling companies had been allowed to function without oversight from Swedish authorities for many years and their expectations regarding the effectiveness of stricter legislation were low.

One cannot expect anything when they have run amok as they have done for so many years, completely unregulated, to think they would just fall into line as agreed is incredibly naive (IN 2, Mutual support association)

The second topic within this theme concerns the forms and scope of self-regulation and issues of responsibility. The responsible minister urged the market to develop forms of self-regulation aimed at eliminating what he termed ‘aggressive’ advertising. Stricter legislation was, then, only to be activated if this strategy did not produce satisfactory results.

If they came back with self-regulation that led to a noticeable change that reduced the aggressiveness and extent of advertising, then of course we would choose that path before a change in the law. The industry came back with self-regulation, but our assessment is that this regulation is not enough, that self-regulation will not lead to a noticeable change, which means that we will not have in place a sufficiently good level of protection for consumers (SP 2, Government and authorities)

The question of who to hold responsible for setting up clear guidelines regarding moderate gambling was addressed by several stakeholder groups – should it be the authorities, the gambling companies, or possibly the media? (IN 3, Industry; IN 4, Industry; IN 8, Media).

I think strong legislation is fundamental, but self-regulation is so much more. It’s not just the guidelines that you have to follow, it’s about this dialogue. Gather together, interpret the law and regulations, hold trainings. […] That’s the work we’re really doing, learning from each other (IN 4, Industry)

Advertisers must take their own responsibility with self-measures and self-regulation, especially when it comes to advertising for products and services that pose risks to public health (SP 4, Media)

Both industry and media stakeholders recognized their mutual responsibility for ensuring moderation in advertising (IN 4, Industry; IN 7, Industry; IN 10, Industry, IN 8, Media; DO 5, Media) and supported self-regulation measures through mutual cooperation (SP 4, Media; SP 9, Industry). Having said that, a media stakeholder was not convinced that gambling companies were acting in a sufficiently responsible way. As a consequence, the media needed to protect its own brand and reputation, not least in relation to other advertisers.

We will not have a good environment for our advertisers if we do not take responsibility for these things, because it will drain the value of our advertising breaks if there are gambling companies that you do not like. So, we need to take responsibility in order to maintain our value, quality and credibility (IN 8, Media)

One of the most divisive issues within this sub-topic was, however, whether some form of risk classification of gambling products should be introduced. Such classifications could refer either a) to the product itself, or b) to the individual gambler. In this case views varied within, as well as between, stakeholder groups. According to annual reports, the state-controlled and state-owned companies, generally had comparatively less income from high-risk games. These companies favored risk classification and a possible ban on online casino advertising. Consequently, they decided to cease with such online casino advertisements in April 2019.

Now we chose to remove [advertising for high-risk products]. Not because it was not moderate in content, because we thought it was. But mainly because we want to be more restrictive when it comes to red [risky] products (IN 7, Industry)

In contrast, commercial online gambling companies, who are highly dependent on revenue from so-called high-risk games such as online casino and slots, argued that gamblers should be classified according to individual risk. In their view gambling is never totally risk-free, but individuals are more or less susceptible to problematic gambling behavior.

Addictive and unhealthy behavior by individuals is easy to track by the gambling industry thanks to the large amount of data which is continuously collected from all gamblers. Any future risk classification system must be based on the conditions and actions of the individual (DO 16, Industry)

Conflicting views on risk-classification and advertisement of high-risk products were also notable within the government and authority stakeholder group. The public inquiry on the future regulation of the gambling market (SOU, 2020: 77) proposed that the Swedish Gambling Authority should develop a model for risk classification of different forms of gambling. However, the Gambling Authority was skeptical toward this idea, while the Public Health Agency, on the other hand, was more positive about the proposal.

The third and last topic within this theme differed somewhat from the two former topics by its more ‘wait-and-see’ approach. That is, a focus on market consolidation and the importance of time. Industry and media stakeholders agreed that an increase in marketing volume is to be expected when a market is being re-regulated. As the market consolidates, the number of actors will gradually decrease and thus so will gambling advertisements (IN 4 Industry; IN 3, Industry, IN 8 Media; IN 12 Industry, SP 4, Media).

We are also bothered by advertising, that there is so much. But it’s a natural part of a re-regulation (IN 4, Industry)

All industries that end up in re-regulation, there will be a peak before it stabilizes and you get your position. […] it’s like three phases. We have the time before re-regulation, a little bit like the wild west and quite messy. Then we will enter into a huge hype […] when they have received their licenses. It may last for 18 months. Maybe 12. Maybe 24. […] And then, at some point we will move slowly but surely into a new normal market (IN 8, Media)

As a result, several stakeholders within the industry group stated that a modification of the gambling regulation, for example stricter marketing rules, was premature. In fact, such measures could severely distort the new regulation (IN 4, Industry; IN 8, Media; IN 12, Industry). License holders have a competitive advantage by being allowed to market their services in Sweden. Industry stakeholders argued that if that incentive vis-à-vis unlicensed companies were to be restricted, the motivation of being a license-holder, and thereby being bound by Swedish regulation and paying tax to the state, would no longer be enticing (IN 3, Industry; IN 4 Industry; IN 12, Industry). Gambling companies could choose to give up their licenses while still targeting the Swedish market, and other companies would become less interested in acquiring a license. Different stakeholders expressed the need for more time to see the full effects of the new law (IN 3, Industry; DO 5, Media; IN 12, Industry).

Theme 3: protection – consumers or market?

The third theme in our analysis consists of topics that are more overarching in character, relating to different understandings of what gambling advertising is all ‘about’. One of the most fundamental issues in this respect is whether a legal product such as gambling should be subject to special restrictions or not – something which potentially impinges on freedom of trade. An important element in freedom of trade is the right to advertise one’s products. All stakeholders in the industry and media groups agreed that licensed gambling companies must be allowed to make their presence, and at least some of their services, known to the public. Some stakeholders took this right one step further and called upon a constitutionally guaranteed freedom of expression. These stakeholders maintained that limiting the amount of gambling advertising is an infringement on freedom of expression and incompatible with highly valued democratic principles.

It is legitimate for commercial companies, whatever they market as long as the activity is legal, to be allowed to market their products and services. There are some goods and services that need a special regulation, including gambling, because they have some special effects among consumers (IN 3, Industry)

It [Gambling advertising] was very intrusive […] But it is also a commercial freedom of expression that is very important for democracy. Freedom of expression is still the cornerstone of democracy (IN 8, Media)

Government and authority stakeholders showed similar concerns, mainly referring to the fundamental principles of the European Union.

There are legal challenges in limiting the amount of marketing. These are partly considerations of freedom of the press and freedom of expression, and partly of EU legal considerations regarding free movement [of services], which is one of the EU’s fundamental freedoms (SP 7, Government and authority)

Despite recognizing the fundamental principles that legal products and services may be promoted in a free market, stakeholders within all groups acknowledged the need for particular regulation of some products due to their potential negative effect on consumers. Government and authority stakeholders were concerned that increased volume and ‘aggressive’ content of advertising could lead to increased consumption, and thereby more problem gambling (SP 2, Government and Authority). Moreover, consumer protection and public health were closely related to each other:

We will do what is necessary to ensure that Swedish consumers are protected in the gambling market. This is not because it is annoying to have lots of commercials when you’re watching football. This is about public health; it is about protecting the thousands of vulnerable consumers in this market, and it is the state’s duty to act (SP 2, Government and authorities)

However, industry stakeholders argued that gambling advertising was a minor precursor to gambling problems (IN 3, Industry; IN 4, Industry). Although some mutual support associations questioned whether gambling advertising should be allowed at all – in the same way as tobacco advertising (IN 2, Mutual support association) – they agreed with industry stakeholders about the limited impact of gambling advertising in general. In their view, advertising is not primarily directed toward those who already gamble:

Advertising is aimed at those who do not gamble […] And either you get mad, like my mother who cannot watch TV because she just starts thinking about me and my [gambling] history. But then there are those who think ‘yeah, yeah, but maybe I should try it out (IN 2, Mutual support association)

For mutual support associations, the detrimental effects of bonuses as a marketing strategy were of much greater importance than advertising as such (IN 1, Mutual support association; IN 2, Mutual support association; IN 5, Mutual support Association; IN 6, Mutual support association). Bonuses for them were considered a trigger, especially when associated with their preferred gambling type and previous wins.

Bonuses were also a topic of interest among industry stakeholders, but their take on the issue was quite different. They argued that limitations on marketing, especially regarding bonuses, constitute a notable competitive disadvantage against unregulated companies. Moreover, they maintained that these limitations are potentially detrimental for consumer protection because gamblers are likely to keep gambling in many different companies when chasing sign-up bonuses (IN 3, Industry; IN 9, Industry; IN 12, Industry).

The effect of the bonus rule is that at-risk gamblers are chasing bonuses. Before, they called customer service and said ‘I want a bonus.” And so, they kept them as a customer and were able to talk to them. Now, they go to other gambling companies and register as new customers and get a bonus. […] And then you can jump around among all those over 80 licensed companies (IN 12, Industry)

Bonuses in gambling advertising have been interpreted and implemented differently by all stakeholders, providing an example of the diverse viewpoints and vested interests that are to be considered while legislating.

Discussion

Concern over harmful overconsumption of gambling has led to stricter marketing rules worldwide, for example, prohibiting advertising of online casinos or even a general ban on the promotion of gambling services. This is also the case in Sweden, where the new Gambling Act introduced in 2019 included several regulations aimed at preventing harmful gambling, including specific rules about gambling advertising – most notably that such advertising should be ‘moderate’. The purpose of this study was to analyze stakeholder interpretations of ‘moderation’ in advertising following the re-regulation of the Swedish gambling market. Our study discusses three main themes: the definition of moderation in terms of content or volume, how gambling advertising should be regulated and by whom, and who the legislation is meant to protect. Our results explore the complexity of how policy processes continue and how stakeholders respond and reposition themselves even after a new legislation has been introduced.

All the stakeholders in our study were aware that the law only regulates content, not volume. Nonetheless, ‘moderation’ was recurringly referred to in terms of both content and volume – suggesting great uncertainty about the meaning of this central term at the time. This could be interpreted as a genuine need for further guidance, as expressed by several stakeholders. Furthermore, the conceptual confusion may reflect gambling companies’ inclination to position themselves exactly within the limits of the law or to challenge these.

Despite the fact that lawmakers had ruled out regulation of volume, the window of opportunity (Kingdon, 2014) seemed to be open even after the law had been introduced. Political responses seemed to move in circles and most stakeholders in all groups were preoccupied with volume as well as content. The responsible minister was one of the most adamant spokespersons for stricter regulation of both content and volume of gambling advertising, even when the new legislation was fresh off the press. For other stakeholders, the content issue materialized as a question of risk classification, either of the gambling product or of the individual gambler. Industry representatives’ positions on this matter were clearly related to their economic positions in the market and their portfolios of gambling products. In the results, we also identified different views on the feasibility and desirability of a risk classification system between government and authority stakeholders. The way that a problem is defined plays a crucial role in agenda-setting processes as, depending on the description of the problem, different solutions will be developed. Generally, legislation needs to be accurate and distinct but also flexible to be applicable to different situations. Obscure definitions in legal concepts might increase the chances of policy adoption (Baier et al., 1986; Matland, 1995) but may also, from the legislator’s point of view, hamper steering and successful implementation. From the receiving end of the table, lack of clarity could entail a window of opportunity for continued negotiations about the interpretation of policies. The ‘true’ impact of new regulations is then likely to be delayed and all parties prepared to put resources into continued deliberations about central concepts and premises. The volume of gambling advertising was initially perceived as highly intrusive and bothersome, especially by the general public – generating immediate policy responses and ongoing deliberations about what the problem actually was. However, because definition and understanding of the problem was not clear to all stakeholder groups, the agenda-setting process continued (Kingdon, 2014).

Differences in interpretation also extended to responsibility. Who is responsible for maintaining ‘moderation’ and protecting market agents/consumers – and in what way? Stakeholders in this study were varyingly eager to suggest new (and stricter) legislation, self-regulation, or allowing the market to consolidate over time. All stakeholder groups seemed to agree that the government and authority stakeholders have overall responsibility for protecting the general public from harm, including overseeing compliance with rules and regulations. However, opinions varied within stakeholder groups, including government and authorities, about the specific details within the regulatory system. Industry and media stakeholders have a responsibility for gambling advertising as indicated in the Marketing Act (2008: 486, 23 §) and have recognized both the need for mutual cooperation and the existence of common interests, highlighting the risk of advertising fatigue and consequent economic losses. In this way, media stakeholders are forced to consider both content and volume in their advertising strategies.

Similar to what Selin (2016) found in his study of gambling policy reform in Finland, there is a ‘paradoxical situation’ between stakeholder’s interests; the market aspect of increased revenue and the continuous concern over protecting consumers. Our study shows that gambling advertising is a balance between protecting businesses in a legal market and consumer protection. As far as the former is concerned, some stakeholders saw advertising restrictions as a potential infringement of freedom of trade, while also exploiting this window of opportunity to define this as an issue of freedom of expression (Kingdon, 2014). When it comes to the latter, an underlying question was, of course, the potential effects of gambling advertising at large. This question was mainly addressed by mutual support associations, who generally opposed all gambling advertising but did not consider it to be aimed at people experiencing gambling problems primarily. Mutual support groups were largely more focused on helping individuals dealing with the consequences of problem gambling than on dissecting the precise definition of the concept. Generally, these stakeholders discussed vulnerable populations referring to children under 18 years old or individuals who have chosen to self-exclude. Nevertheless, they were generally vague and contradictory about who is actually to be protected; the population at large, all gamblers, or at-risk or people experiencing gambling problems and their families. Media stakeholders were primarily concerned with the population at large, or rather their viewers/readers and other (potential) advertisers. Mutual support associations tended to highlight the needs of people experiencing gambling problems and their families, while government and authority stakeholders were more imprecise in defining target groups for various protective measures. The responsible minister was, for example, varyingly concerned about consumers, vulnerable consumers, and public health. These dual aims are also reflected in the present legislation, where protection of ‘the market’ and ‘the consumer’ are to be balanced vis-à-vis each other.

Our study focuses on the first period after the new legislation was introduced and how stakeholders were positioning themselves in this re-regulated market. The most important limitation lies in the varying availability of data material, partly reflecting the complexity and structure of the Swedish gambling market. Stakeholders are more or less active in producing policy documents, statements and press conferences, etc., - and are varyingly accessible for interviews and research participation. In this way we cannot be certain that we have captured the sentiments of different stakeholder groups comprehensively. At the time of writing, when the law has been in operation for four years, statistics show that the volume of gambling advertising has stabilized at the same level as prior to the introduction of the new legislation (Swedish Public Health Agency, 2022). Our results suggest that preconditions for successful regulation are hampered when central concepts are not well defined, when there is insufficient consensus about what problem is to be solved, by whom and in what way. In addition, timing is vital, as markets evolve continuously, alongside shifting public opinion and political leadership (Liu et al., 2010). From a policy perspective, this suggests that rapid changes to newly introduced regulations may be premature because the full effects are not recognizable until the first regulations have had time to bed in.

Investigations of how legislation concerning gambling advertising and public health varies between countries, and how policy processes and the choice of policy instruments are constructed and implemented, are obvious venues for future research.

Conclusion

With the re-regulation of the Swedish gambling market, a license-based system was introduced allowing commercial online gambling companies to offer and advertise their services in Sweden. The new Gambling Act stipulates that gambling advertising should be ‘moderate’. Our study explores stakeholders’ perceptions and interpretations of this concept at a time when the gambling market was under reconstruction.

All stakeholder groups agreed that moderation referred to content only, but the agenda-setting process continued as advertising volume was recurringly addressed. Nonetheless, there was some disagreement between stakeholder groups about what problem the regulation was meant to address, the distribution of responsibility, and the choice of relevant regulatory instruments. Most stakeholder groups favored self-regulation within the market and allowing the market to consolidate, while recognizing the need for government regulation and surveillance. There was a notable ambivalence within and between stakeholders concerning how to balance freedom of expression, market protection, consumer protection and public health.

Our study shows the incremental and interactive character of policy-making, where new legislation resulted in new policy responses. When key concepts are not well defined, legislation may be a blunt steering instrument that leaves interpretation to stakeholders with opposing interests. Case law and market consolidation take time and regulatory interventions need to take this into account. This suggests that Kingdon’s windows of opportunity do not open only once but on multiple occasions. In other words, policy processes are probably better viewed as interactive and circular, rather than linear and stationary.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data that support this research are available on request from the corresponding author, unless it compromises the anonymity of individuals where this has been requested.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14459795.2024.2323044

Additional information

Funding

The work was supported by the Forskningsrådet om Hälsa, Arbetsliv och Välfärd [2016-07091].

Notes on contributors

Katya González Díaz

Katya González Díaz is a PhD student in the Department of Public Health Sciences at Stockholm University. She is working with gambling related issues involving the re-regulation of the Swedish gambling market and its relation to public health. Her main research interests include public health, policy and behavioral addictions.

Jenny Cisneros Örnberg

Jenny Cisneros Örnberg is associate professor in political science and senior lecturer at the Department of Public Health Sciences, Stockholm University. Her main research interests include the intersection between domestic and international policy-making in the field of public health, with a focus on alcohol- and gambling policy and regulatory change on national and international level.Cisneros Örnberg is PI for the research program REGAPS (Responding to and Reducing Gambling Problems Studies).

Therese Reitan

Therese Reitan is professor of public administration at Södertörn University and affiliated researcher at Stockholm University. Her main research interests include policies, organization and professional practices aimed at alcohol and drug misuse, public administration and leadership, as well as road safety policies.

Notes

1. Note: Codes are provided to identify the data type (IN = Interview, SP = Seminar or Press conference, DO = Policy document and statements) followed by an assigned number, and type of stakeholder group.