Høringssvar fra European Gaming and Betting Association
EGBA Contribution
The European Gaming and Betting Association (EGBA) is the Brussels-based industry body representing the leading online gaming and betting operators established, licensed and regulated within the EU. On the basis of these licenses EGBA members operate in a large number of Member States
Question 1:
Responsibility measures
Rambøll's contact with potential licence applicants shows that it will be possible to introduce a number of responsibility measures in a licensing model (Model 2B), but that a responsibility regime corresponding to the current basis for Norsk Tipping's operations will not be possible in a licensing model.
What level of responsibility do the consultative bodies think should be aimed for in a future Norwegian gaming model?
First and foremost, EGBA respectfully but strongly disagrees with the statement in the background text of the consultation: "…contacts with potential license applicants shows that it will be possible to introduce a number of responsibility measures in a licensing model (Model 2B), but that a responsibility regime corresponding to the current basis for Norsk Tippings operations will not be possible in a licensing model."
As it will be further developed, it must be born in mind that the level of consumer protection and the types of responsibility measures can be set by the legislator equally or even better for a licensing model than for a monopoly system. Moreover, a licensing system allows the public regulator to supervise and control not only the state-owned operators, but also the private ones, which are currently already accessed by consumers who are resident in Norway, which means that the overall level of consumer protection is higher in the licensing model.
We consider that gambling regulatory frameworks should be aimed at achieving a high level of responsibility and consumer protection. Achieving this level requires an evidence based approach, which recognizes the central role of the consumer, reflecting it in the considerations and policy setting of the legislator. It is relatively effortless for the digital consumer to find access to an unregulated offer on the internet, if the regulated offer is not sufficiently attractive or the public policy and its implementing measures are not adjusted to the realities of the digital environment and behavior of the digital consumer. A high level of consumer protection can only be achieved when the largest number of the targeted audience are channeled to the Norwegian regulated offer and away from the unregulated offer. This goal can only be reached in the licensing model.
.Importantly, there have been a significant number of European countries that have already regulated the online gambling market and which have gained valuable experience with consumer protection measures. EGBA would strongly encourage the Norwegian government to take these experiences and best practices into consideration as they can provide invaluable input. EGBA stands ready to provide its experience and knowledge to help the Norwegian government regulate the online market under a multi-licensing regime and a high level of consumer protection.
Question 2:
What level of protection do the consultative bodies think will be possible to establish under a licensing model?
Data and experience in Member States of the EU shows that a high level of consumer protection can be achieved under a multi-licensing model equally or even better than under a monopolistic licensing system. In fact, due to the fact that competition between operators leads to an improvement of the offer, multi-licensing models are more likely to channel consumers towards the regulated offer. Therefore, in some situations the level of consumer protection may be even higher in multi-license systems than in monopolies. It should also be borne in mind that ultimately, if the model chosen fails to encompass a maximum number of players, irrespective of how stringent the responsible gambling measures will be, they will not reach a sufficient number of players and thus not achieve their purpose.
With multi-licensing models, both the legislator and the regulatory authority are able to design a regulatory framework, which achieves a pre-defined level of protection, allows gambling operators to obtain a license to provide services under clear and well-defined terms, in particular if they comply with the (consumer protection) licensing requirements determined by the legislator. If the licensing applicant or operating licensee does not meet or no longer meets the requirements then the license can either be refused or withdrawn. Similarly, the license can be suspended or withdrawn if issues would appear after the granting of the license. Thus, the level of consumer protection is always established by both the legislator and the regulatory authority, irrespective of the regulatory model.
It should be emphasized that the level of protection established and well-defined by the legislator is the same or higher in a multi-licensing system compared to a monopolistic licensing, with the essential difference in a multi-licensed system that the regulatory authority's power is to license, monitor, supervise and control more than one licensee. As stated above, the CJEU, in particular by its ruling of 24 January 2013 in the Stanleybet, William Hill & Sportingbet case (joined cases C-186/11, C-209/11), showed no preference for one or the other model, but obliged the Member State to consistently establish such a licensing model with adequate control mechanisms. If a Member State opts for a monopolistic licensing model, the CJEU case law obliges the Member State to be consistent with the objectives pursued by the Member State’s own legislation, and to establish a very strict control of the monopolistic operator in order to attain the stated objectives of a particularly high consumer protection.
It is furthermore noteworthy to mention that the growth of the Internet usage has not led to an increase in the incidence of problem gambling, which lies around 0.5% to 1.1% of the overall population in Europe.[1] This is corroborated by the fact that national problem gambling prevalence surveys around the world and peer-reviewed research have evidenced that problem gambling has remained remarkably stable over the past decade despite the significant increase of Internet access and online gambling opportunities[2]. The SINTEF study conducted on behalf of the Norwegian Gaming and Foundation Authority in 2007 confirmed that despite a substantial increase “in the Norwegian gambling market, the prevalence of gambling problems has been stable”.[3] The Dutch WODC of 2011 marked a slight reduction of problem players in the Netherlands in comparison to 2005.[4]
A 2010 research study on National Gambling Regulations and the Prevalence Rates of Pathological Gambling confirmed that there is no real correlation between national regulation and pathological gambling and that the level of gambling addiction remains stable throughout the various national systems.[5].
Figure 1: Prevalence Rates of Pathological Gambling
Country |
Year |
Screen |
Percentage |
Article |
Sweden |
2008 / 2009 |
PGSI |
0.3 |
As cited in British Gambling Prevalence Survey 2010 |
Norway |
2008 |
NODS |
0.8 |
Float, A., & Bakken IJ. (2007). Gambling and gambling problem in Norway 2007 [Gambling and Gambling, Problems in Norway 2007]. Oslo: SINTEF Health. |
Canada |
2003 |
PGSI |
0.5 |
Cox B., Yu N., Afiffi T & Ladouceur RA. (2005). National Survey of Gambling Problems in Canada. Canadian Journal of Psychiatry. 50, 213-217. |
New Zealand |
2006 / 2007 |
PGSI |
0.4 |
Cox B., Yu N., Afiffi T & Ladouceur RA. (2005). National Survey of Gambling Problems in Canada. Canadian Journal of Psychiatry. 50, 213-217. |
UK |
2010 |
PGSI / DSM-IV |
0.7 / 0.9 |
British Gambling Prevalence Survey 2010 |
Germany |
2007 |
SOGS |
0.6 |
Federal Center for Health Education (BZgA) (2008). Gambling behavior and problematic gambling in Germany in 2007 [Gambling behavior and problem gambling in Germany in 2007]. |
Iceland |
2005 |
PGSI |
1.1 |
Olason DT., Finnbogadottir H., Hauksdottir A., & Barudottir SK. (2003). An Icelandic version of the Problem Gambling Severity Index: A psychometric evaluation. Paper presented at the 27th Nordic Psychiatric congress, Reykjavik, Iceland. |
The above table also evidences that similar rates of problem gambling are found in both monopolistic models and multi-licensed models. Studies carried out in both Norway[6] (monopoly system) and the UK[7] (multi-license system) show a similar rate of problem gambling (0.5 to 1%).
Data in other countries show that a multi-license system allowing for an attractive and quality offer allows to channel consumers towards the regulated offer, which complies with the standards set by public authorities under their licensing conditions. In contrast, if the regulation does not allow for a sufficiently attractive offer, the consumer will look (irrespective of the chosen model) for a more competitive offer, which is provided by operators outside of the Norwegian regulated offer and thus outside of the protective umbrella of the Norwegian or EU-EEA consumer protection legislation. In the internet era, the alternative unregulated offer is just “one click away”, and the consumer will look for it – irrespective of attempts to block access – if the regulated offer is not competitive enough. This has been recognized by the European Commission, which stated that:
“The development of an attractive range of legal gambling opportunities is also key to effectively prevent consumers from going on unregulated sites.”[8]
“It is important for authorised operators to be able to offer sufficiently attractive products, because in the absence of credible and sustainable offers consumers will continue to turn to unregulated gambling websites, with the ensuing potentially harmful effects.”[9]
“The right balance needs to be struck and measures to protect consumers should not have the adverse effect of leading players to seek more attractive offers on unregulated sites.”[10]
In short, a well-functioning multi-license gambling regulation is able to set the same or higher consumer protection standards as a monopoly system, while, in turn, being able to channel players towards the regulated offer more effectively. Due to high channeling, a multi-license regime reaches more consumers than a monopolistic regime, which renders the multi-license regime a better fit for ensuring players are safe.
Question 3:
How do the consultative bodies view the idea that revenue can be secured for the voluntary sector from gaming under a licensing scheme, compared with the current monopoly in which Norsk Tipping's surplus goes directly for the purposes?
We respectfully but strongly disagree with the economic assumptions that underpin this question. We provide extensive evidence and arguments on the economic facts in a separate annex that is attached to this contribution.
It is useful to point out that as online gambling is an economic activity (of a cross-border nature involving the Internet and online commerce), monopolies in this field are, in principle, contrary to the freedom to provide services recognized on Article 56 of the Treaty of Functioning of the European Union (hereinafter “TFEU”) and Article 36 of the EEA Agreement. Whilst in some situations Member States are allowed to issue regulations that restrict the freedom to provide services, these restrictions must be justified by the pursuit of a legitimate public interest. It is, however, established case law that economic objectives such as increasing state revenue or ensuring a certain return to good causes, are not legitimate objectives in justifying a restriction of the gambling offer. As a result, this question of securing income, should be treated as secondary.
Furthermore, the compliance of the Norwegian current legislation with the EEA Agreement, in particular, in respect of some of the business practices of Norsk Tipping, has raised concerns lately. Such business practises are e.g. the large advertising investment and the aggressive tonality in advertising by Norsk Tipping. We will not further expand on this topic in this contribution, but we would like to flag to the Norwegian Authorities their obligations to respect the EEA Agreement. It is established case law that a state monopoly justified invoking consumer protection related reasons shall at all times fulfil the requirements of a so called hypocrisy test of consistency, meaning in practise that a monopoly shall not advertise more than necessary and strictly refrain from enticing consumers to gamble more. In this regard, in Markus Stoß and Dickinger and Ömer the CJEU imposes strict requirements on any admissible advertising for gambling. In particular, the CJEU limits all advertising strictly to “what is necessary for channelling”, i.e. it prohibits advertisements and other actions aimed at expanding the market beyond what is necessary to entice customers to stay with the legal offer.
On the other hand, as demonstrated in Annex I, the liberalization of the online gambling market does not affect the level of revenues stemming from its offline counterpart. There is hardly any cannibalization between online and land based gaming, in addition to which the online gaming operators, which would be interested in becoming licensees, already have a market share in Norway[14]. As a result, the market share of Norsk Tipping is not likely to decrease significantly.
An example of that can be found in Denmark, where in spite of the activity of private and regulated online gambling operators, the GGR for land based casinos increased to 95 M DKK in Q3. This is the strongest quarter since the beginning of 2012.[15]
Question 4:
In the event of a continuation of the monopoly model, what measures do the consultative bodies think should be implemented in order to ensure that the level of gambling problems continues to be held low, and that Norsk Tipping succeeds in channeling players to regulated games?
A great part of Norwegian consumers play on unregulated sites. In this regard, a report prepared by the Menon Consultancy[16] acknowledges that international gaming operators have recently reached a significant market share in Norway, which is likely to increase in the future. Also, the decreasing market share of the monopoly increases the loss of state income. Despite the significant market share of the unlicensed offer, the level of gambling problems in Norway is low as demonstrated above. As a result, maintaining the monopoly in Norway cannot be justified by invoking reasons relating to prevention of problem gambling.
Norway has imposed marketing restrictions and payment blockings to restrict the activities of foreign operators and to enforce the monopoly. However, the authorities have admitted themselves that these measures are not efficient.[17] In this regard, establishing a licensing system would bring most of the companies already operating in Norway to the regulated market where they would comply with the standards set by the Norwegian authorities. It should also be noted that maintaining the monopoly by imposing various blocking measures has turned out to be impossible in practice, as the blocking measures are not only legally questionable but also inefficient.[18]
As a result, we would strongly advise the Norwegian authority to opt for a multi-license system, which would allow them to control and monitor the activity of the private operators, which that are already offering their services in Norway, and thus to oblige them to comply with the Norwegian standards of consumer protection. Implementing a multi-license regime, operators conducting business in Norway would compete on equal terms and the Norwegian authorities would be able to supervise and regulate the operations.
Question 5:
How do the consultative bodies view such far-reaching liability instruments?
Whilst we fully agree that responsibility tools are essential to inform, educate and protect the consumer, experience and evidence shows that these behavioral tools are only effective when the consumer is empowered by these tools, rather than that they are forced upon the consumer by law.
We are not aware of any other country, using monopolistic or licensing regimes, applying such far reaching measures. It is, therefore, not possible to make an evidence based assessment of the results that could be brought by the application of those measures.
However, its application is able to entail harmful results as regards consumer protection, in particular, in what concerns consumers’ privacy. The potential incompliance of the measures with European and Norwegian data protection legislation is dealt with in question 10, but beyond that, our experience shows that consumers do not feel comfortable when private and, in particular, public gambling operators request and store large amounts of data regarding their gambling habits.
Consequently, and paradoxically, applying such far reaching measures may create the impression on Norwegian consumers that their privacy is undermined in the regulated regime and perhaps better protected by operators outside the Norwegian regulatory framework. In addition, the sharing of information about consumers between competing operators may result in consumers’ exposure resulting from security breaches, or misuse of information.
As we have stated, we are not aware of any other country using such far reaching measures, and, therefore, we cannot properly assess the negative impact that the measures would cause on the safety of Norwegian consumers. Nevertheless, evidence shows that less intrusive measures are effective on fighting problem gambling. We consider that regardless of the model that is chosen, it is advisable to use the tools that are proven to be effective, instead of developing new tools that are likely to push consumers outside the regulated offer.
Question 6:
Is it feasible to introduce such instruments in a licensing scheme?
The same consumer protection instruments that have to be introduced in a monopoly system, can also be introduced in a licensing scheme. Thus, the question is rather whether the said instruments are effective. Additionally, EU-regulated operators are already obliged to comply with the consumer protection standards of all the countries in which they hold a license.
Equally important is the fact that consumer protection and fraud prevention measures enhance the trust of consumers in online gambling operators. Therefore, having a high level of Corporate Social Responsibility (CSR) is a significant competitive element for operators. An explanation of these measures can be found in the following videos:
https://www.youtube.com/watch?v=ayOfTn2-S5k Maris Bonello, Unibet
https://www.youtube.com/watch?v=XMkDLHOBBiU Joachim Hausler, bwin.party
In addition, all EGBA members comply with the CEN (European Committee for Standardisation) Workshop Agreement on Responsible Remote Gambling Measures (CWA 16259: 2014), the first set of pan European consumer protection measures for online gambling in the EU. Their compliance is audited on a regular basis by eCOGRA, an independent auditing company.
In addition to individual measures, many operators also use advanced software to detect gaming problems and contact customers proactively. Further, many operators reserve the right to suspend or close a player’s gambling account, if it seems that gambling is causing the player problems. These solutions combined with well-educated personnel are the most important factors in tackling gambling problems.
Question 7:
Will a licensing scheme with extensive responsibility tools such as the Gaming Authority describes be attractive to possible applicants for a licence?
It is possible to set extensive responsibility tools in a license system that is at the same time attractive to potential applicants for a license. Examples of that can be found for instance in Denmark and the UK.
These tools are independent from the model that is chosen by the national legislator to regulate gambling in a given country (that is, monopoly, single or multiple licensing scheme). There are indeed adequate mechanisms which can be applied regardless of whether or not a single or a multiple license system has been chosen.
The Consultation letter states that "mandatory limit tools in which players must set their own loss limits that apply across all gaming operators" shall be considered. Such a mechanism shall be set up in a realistic manner, while a "loss limits applicable across all gaming operators" will factually most likely not work out. In addition, tools which lead to inefficient but hindering services, like the ones described in the question above, are very likely to discourage players to gamble on regulated sites, private or state-owned.
Question 8:
What specific measures do the consultative bodies think may not be possible or desirable to implement under a licensing model?
As discussed in previous questions, the same responsible gambling measures that can be implemented in a monopolistic regime can also be implemented in a multi-license system.
Furthermore, by establishing a multi-licensing regime, public authorities are able to use the reputation of private operators in order to force them to comply with national public standards. In this regard, reputation is one of the most important assets for regulated companies as consumers in general prefer to gamble in the sites they trust on. We suggest the Norwegian authorities to develop a Norwegian trust mark, which can be used by licensed companies to differentiate themselves from the operators on the black market.
Such a method would not pose a serious burden on operators, and would ensure their compliance with Norwegian law because losing the right to use the trust mark would involve reputation loss and a quick and dramatic loss of their Norwegian market share.
The Gaming Authority points out that introduction of extensive responsibility tools in a licensing scheme will have political, technical and legal challenges, and will demand resources. However, the Gaming Authority does not go into detail on which challenges there are in such a scheme.
Question 9:
What challenges do the consultative bodies think will arise?
We cannot assess the challenges that “extensive responsibility tools” will present without knowing the concrete tools that are to be implemented. However, whatever these tools are, they must be compliant with the obligations that stem from the EEA Agreement, and be applied equally across all licensed operators, including the incumbent ex-monopoly. In particular, the freedom to provide services put forward by Article 56 TFEU and Article 36 of the EEA agreement forbids Member States to put forward measures that discriminate against foreign companies, and establishes that any restriction to provide cross-border services must be justified on public policy grounds, such as the prevention of crime or the protection of consumers.
At the same time, responsibility tools should not impose an excessive burden on operators and, especially, players. Otherwise, the latter will end up playing on the black market. Achieving this balance is not an easy task as too lenient measures might fail to duly protect consumers, while too restrictive measures may impede the channeling of consumers towards the regulated offer. However, other Member States have achieved this balance (i.e. UK), and EGBA is willing to share its expertise built on years of experience on the online gambling sector should Norway request it. Whatever the chosen measures will be, the most important factor for a well-functioning gambling regime is that it’s able to channel the customer efficiently to consume the services of the regulated offer. If high channeling is not achieved, the whole regime fails to protect the consumers despite individual sophisticated responsible gambling measures.
This applies regardless of the regulatory model that is used in the country.
The Gaming Authority is of the opinion that introduction of extensive responsibility tools entails problems concerning the right to privacy under the Norwegian Personal Data Act, in that such a scheme will mean collection and storing of sensitive personal data.
Question 10:
How do the consultative bodies view this issue?
We consider that to set limits that apply across all gambling operators and its implementation measures raise serious Data Protection concerns. Applying such measures would oblige operators to perform data transfers about the gambling behavior of identifiable individuals not only to the Norwegian gambling authority, but also to any other company having a license to operate in Norway. As there are less intrusive means to fight problem gambling, we consider that such an amount of data processing would contravene the principle of proportionality that shall govern data processing according to Article 6(1)(C) of the Data Protection Directive, which was incorporated into the EEA Agreement in 1999.
Protection of regulated operators, regardless of regulatory model
The Gaming Authority outlines various measures which they think may be of interest in barring unregulated operators from the Norwegian gaming market, regardless of regulatory model. The measures outlined are blocking of unregulated operators' web pages, criminalising players, strengthening of the prohibition on facilitating payment, excluding players via third persons and halting marketing from illegal operators.
Question 11:
How do the consultative bodies view the practicability and desirability of introducing such measures in order to bar unregulated operators from the Norwegian gaming market?
Member States have used a wide range of methods to try to prevent the offering of gambling services by unregulated companies. However, they have been so far unsuccessful, in particular, considering that such methods are difficult to enforce against operators based outside Europe, e.g. in Asia. In this regard in the context of court proceedings on the Danish model, the European Commission stated that:
“it is questionable whether these blocking systems could produce the expected results, as either the online gamblers could escape the internet blocking by changing ‘ports’ used, or the prohibition of certain payments could block licit commercial transactions other than relating to stakes and prizes.”[19]
Furthermore, circumventing payment bans is relatively easy for non-EU operators for instance by circumventing coding rules. Additionally, players may use foreign bank accounts to circumvent the prohibition.
As regards the criminalization of players, it should be reminded that it is a general standard to apply administrative sanctions in case of failure to comply with gambling law requirements. Moreover, according to the settled case law of the CJEU[20], measures that a Member State takes must be based on objective, non-discriminatory criteria and cannot be disproportionate. In addition, criminal enforcement against players is a measure that is also questionable from the perspective of civil rights and freedoms enshrined in the Charter of Fundamental Rights of the European Union; in particular, as regards the principle of legal certainty as players are not always aware that they are playing on unregulated sites.
For this reason, we consider that the better way to fight unregulated offer is to channel the demand towards the regulated market. To achieve that goal, regulated operators must be able to maintain an attractive offer to Norwegian consumers, including a straightforward and simple registration process, an attractive and competitive offer ensured by a well-functioning market, among others.
Question 12:
The Gaming Authority outlines three different forms of blocking web pages: blocking IP addresses, blocking DNS addresses and DPI-based blocking. How do the consultative bodies view these different forms of blocking as measures for protecting the Norwegian regulated market?
The blocking of websites is highly unadvisable. This is due to the fact that it is very difficult to enforce against non-European operators, which is the actual offer that poses the greatest threat to Norwegian consumers. In particular offers from Asia often lack adequate consumer protection rules and offer no safety in terms of credit-card and other types of fraud.
Furthermore, giving an administrative authority the discretion to block websites without a court decision contradicts not only the basic right to freedom of expression, but also the right to a fair trial, both rights that are included in the European Charter of Human Rights. Besides that, the compliance of such measures with Article 36 of the EEA Agreement is uncertain.
We consider that a greater level of consumer protection could be achieved on ensuring that the regulated offer is attractive to Norwegian consumers, and that the latter are informed about the risks of gambling on unregulated sites.
[1] Please note that variation may be caused by methodology used.
[2] For instance, the UK Prevalence Study of February 2011 used a sample of 7,756 people and concluded that the rate of problem gaming in 2010 (using two different measures) had slightly increased from 0.6 - 0.7% in 2007 to 0.7% - 0.9% before concluding that “it was impossible to say whether these apparent increases were an upward trend or a temporary fluctuation”, British Gambling Prevalence Survey, 2010.
[3] SINTEF Study, Gambling and Gambling Problems in Norway 2007, p.2
[4] WODC report: “Majority of the players uses self-control techniques”, p.50
[5] “Regulating Gambling in Europe – National Approaches to Gambling Regulation and Prevalence Rates of Pathological Gambling 1997 – 2010″, Simon Planzer (ed.), Zurich 2011 available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2045073
[6] Norwegian SINTEF Study, 2007, available at: http://www.sintef.no/upload/Helse/SINTEF-Rapport%20Pengespill%202007.pdf
[7] UK Prevalence Study (2011), available at: http://www.gamblingcommission.gov.uk/research__consultations/research/bgps/bgps_2010.asp
[8] Communication from the Commission to the European Parliament, the Council, the Economic and Social Committee and the Committee of the Regions. Towards a comprehensive European framework for online gambling * COM/2012/0596 final * page 5
[9] Idem, page 10
[10] Idem
[11] Judgment of the Court (Fourth Chamber) of 16 February 2012. Marcello Costa (C-72/10) and Ugo Cifone (C-77/10).Published in the electronic Reports of Cases, para. 56.
[12] Cf. Shock, ibid n. 5, paras. 101-103.
[13] Cf. Dickinger and Ömer, ibid n. 7, para 69
[14] http://menon.no/upload/2015/05/07/samfunnsokonomisk-analyse-av-pengespill-endelig.pdf
[15] https://spillemyndigheden.dk/en/news/data-gambling-market-q3-2015
[16] http://menon.no/upload/2015/05/07/samfunnsokonomisk-analyse-av-pengespill-endelig.pdf
[17] https://lottstift.no/lotteritilsynet/files/2012/01/Rapport-betalingsformidlingsforbodet-jan-20121.pdf
[18] EC (2011) 6499 on the measure C 35/2010, para 126
[19] EC (2011) 6499 on the measure C 35/2010, para 126. Emphasis added
[20] CJEU case 203/08 Sporting Exchange, para. 50