Online gambling and international regulation: an outside bet

Abstract

This paper will consider the factors which hinder regulation of gambling at an international level, particularly focusing on regulation at EU level. These obstacles include: (a) the wide divergence in approach to regulation of gambling activities and of gambling advertising in different jurisdictions and the lack of consensus as to the risks of online gambling; (b) the status of gambling as an activity of questionable morality; (c) the borderless nature of the Internet and the ‘lowest common denominator' regulation problem; (d) the lack of gambling tax harmonization within the EU; and (e) the development of new forms of gambling such as betting exchanges and spread betting and the new regulatory challenges which these activities give rise to. After assessing these obstacles, the paper will then briefly assess some of the regulatory options available to the EU along with the desirability of such regulation.

Keywords:

• gambling
• regulation
• online gambling
• tax harmonization
• spread betting
• betting exchanges
• country of origin approach
• country of destination approach
• E-commerce Directive

Introduction

Online gambling has, to a large extent, brought gambling in from the cold. Up until recently gambling was perceived as an activity largely confined to certain societal groups. However, online gambling has opened up gambling to a broader customer base and has transformed gambling into the subject of mainstream socio-legal debate. In particular the question of whether gambling should be regulated at international level and of what form such regulation should take has become the focus of attention, especially within the European Union. 1 The reasons why online gambling has led to increased calls for gambling regulation may include the wide variety of cultural attitudes towards gambling which can be found within the EU and the fact that the increase in cross-border activity which online gambling has produced has highlighted this lack of consensus. However, this paper will focus not on the reasons why regulation has been called for but rather on the factors which hinder regulation of gambling at an international level, particularly focusing on regulation at EU level.

The divergence of approach to gambling in EU Member States

The first major obstacle to regulation of gambling activity at EU level is the wide divergence in approach to gambling regulation in different Member States and the lack of consensus among them as to the risks of online gambling. Although it has been pointed out that most Member State legislation shares common purposes such as the protection of minors and the control of addiction and compulsive behaviour, the extent to which each Member State restricts the provision of gambling services to fulfil these shared objectives varies greatly. Thus, restrictions placed on online gambling range from almost no restriction to almost complete restriction. In between are those Member States in which foreign operators are prohibited from operating. In addition, the restrictions placed on the advertising of gambling vary as between Member States and as between different types of gambling. The variation in restrictions imposed by Member States emphasizes the lack of consensus as to the risks of online gambling. The divergence in opinion centres on the issues of criminality and the increased potential for addiction created by online gambling. Since there are as yet few studies on the psychosocial impacts of online gambling or on the potential it creates for criminal activity, the impact of online gambling on society is uncertain and a cause for concern, especially since some preliminary small-scale studies point towards an increased level of problem gambling among online gamblers. 2 Governments may fear that countries that export gambling services may also be exporting social costs that other governments may have to address without the benefit of the tax revenues traditionally generated by land-based gambling enterprises.

The status of gambling as an activity of questionable morality

It is clear that there is a wide divergence in opinion among Member States as to the morality of, and the risks attached to, gambling, in particular online gambling. To a greater or lesser extent, all Member State governments recognize that gambling is an activity of questionable morality as a certain proportion of gambling revenue comes from problem and pathological gamblers. Thus, the second major obstacle to regulation at EU level is that such regulation would essentially involve the EU dictating to the Member States how to deal with an activity of questionable morality. In relation to other ‘moral issues’ such as abortion, prostitution and drug control, the EU has taken a non-interventionist approach 3 and to date the ECJ has generally accorded a wide margin of discretion to Member States in gambling cases and has upheld Member States’ right to place restrictions on the provision of online gambling services where there is objective justification for the restrictions and where the restrictions are not just a disguised means of discriminating against foreign service providers. 4

This fundamental lack of agreement between Member States as to the morality of, and risks attached to, online gambling is probably the greatest obstacle to regulation at EU level. However, several other significant impediments exist. Two such impediments are the lack of gambling tax harmonization within the EU and the development of new forms of gambling. However, perhaps more important than these two issues is the impediment to EU-wide regulation caused by the borderless nature of the Internet. It is on this issue that the remainder of the article will primarily focus.

Lack of gambling tax harmonization within the EU

Turning briefly back to the lack of gambling tax harmonization within the EU, it is clear that this could impede EU-wide regulation. At present different direct taxation regimes and in particular gambling tax regimes exist in the Member States and online gambling service providers therefore engage in ‘tax jurisdiction shopping’. 5 For example, due to its favourable tax regime, other EU jurisdictions are already losing gambling tax revenue to Malta. Since such ‘tax jurisdiction shopping’ is, in the majority of cases, permitted under EU law, some degree of tax harmonization would seem desirable in order for an EU-wide consensus on regulation of gambling to be reached. Without tax harmonization, online gambling service providers would be attracted to tax competitive jurisdictions. This would create a risk of distortion of the internal market and other Member States would be likely to oppose the ensuing loss of gambling tax revenue. 6

However, several obstacles stand in the way of such harmonization including the existing substantial divergence between taxation regimes and the requirement for unanimous decision-making set out in Articles 93 and 94 EC Treaty. The borderless nature of the Internet presents a further significant obstacle to harmonization in that there are many tax competitive jurisdictions outside the EU in which online gambling service providers are encouraged to locate. Thus, tax-friendly jurisdictions within the EU could oppose gambling tax harmonization on the ground that harmonization would probably increase the overall tax burden on gambling operators if they remain within the EU and would thereby encourage them to establish themselves outside the EU. As well as this obstacle, another issue that would have to be broached before gambling tax harmonization could be achieved is the desirability of direct-tax harmonization. This is a contentious issue as while the harmonization of indirect taxation is necessary in order to complete the internal market, the case for direct-tax harmonization is less compelling.

The development of new forms of gambling

Furthermore, the development of new forms of gambling may impede EU-wide regulation since they may raise issues from a regulatory point of view that differ from those raised by traditional forms of betting. The two most prominent of these novel gambling forms are betting exchanges (peer to peer betting) and spread betting. Both are conducted primarily on the Internet, with over 90% of spread betting transactions occurring online. 7 In relation to betting exchanges, concerns have been expressed about the increased potential for insider trading and several high-profile cases in the horseracing world involving race fixing and insider trading lend credence to such claim. While economic analysis of the level of insider trading on sports betting exchanges in comparison to the level at bookmakers indicates that the level of insider trading is in fact lower on the exchanges, 8 the perception that betting exchanges facilitate insider trading may prove more influential than the reality. 9 As regards spread betting, financial spread betting providers must comply with the Market in Financial Instruments Directive (MiFID). 10 However, sports spread betting providers are not subject to such a strict regulatory regime in most EU jurisdictions. 11 This lack of regulation in most Member States may prove problematic given the substantial potential losses which spread bettors can incur. 12

The borderless nature of the Internet: the choice of law problem

We will now move on to consider the impediment to EU-wide regulation caused by the borderless nature of the Internet. 13 If, despite the significant difficulties mentioned above, an EU-wide consensus were reached, whether the EU's efforts would be effective in the absence of a broad international consensus is questionable. It is overly optimistic to believe that an EU-wide consensus would lead to a broad international consensus, considering differing cultural attitudes to gambling worldwide and the differing economic imperatives driving governments. It has been said that approaches towards Internet gaming can be grouped into four categories: maximum protection/compliance, prohibition, laissez-faire and tolerance. While EU Member States' approaches all fall within the first two groups, which are at one end of the spectrum, other jurisdictions have adopted the latter two approaches, which represent the polar opposite view. While the Internet may eventually force convergence of national laws due to the inability of countries (in most instances) to enforce their laws against foreign operators, gambling laws are likely to be among the last laws to submit to this convergence by default. In any event, such convergence, if it occurs, is a long way off. Thus, even if an EU-wide consensus were reached, the question remains as to the effects this would have, if any, in view of the borderless nature of the Internet.

The prevalent view nowadays is that traditional state-based laws are capable of ordering the online world. However, the question of which state's laws should apply is fraught with difficulty. While the parties can make a choice of law as part of their agreement, in reality parties often fail to make such a choice. Furthermore, contractual autonomy is not something which exists outside legal systems but rather is created or at least must be recognized and enforced by states. Thus, the parties' choice of legal system may not be recognized and may be partly or wholly replaced with a substantive body of law which appears more closely connected to the contract and to the parties. ‘Close connection’ is a location-centric concept and so brings into play private/public international law (conflict of law) rules. Thus, while greater recognition of contractual choice may improve the situation somewhat, in many cases the competent state will have to be determined by application of traditional conflict of law rules. These rules are designed to vest competence in the state or legal system most closely connected to the facts of the case. However, online activity is intrinsically extra-jurisdictional as messages pass across the Internet as a result of cooperation among many devices, which may be under the control of many different organizations and individuals, who may be located in many different jurisdictions. Thus, it is clear that the traditional ‘close connection’ test is overly flexible in the online context. Therefore, a dichotomy has developed in the debate on competence which distinguishes between the ‘country of origin’ and the ‘country of destination’ as alternative or concurrent regulators.

The country of destination principle and the limits imposed by enforcement jurisdiction

The application of the country of destination principle would lead to worldwide competence, resulting in online service providers having to comply with the laws of hundreds of states. This could lead to legal harmonization at the highest level of regulation. However, perhaps the most acute difficulty with the application of this approach is the enforcement problem. However much states endeavour to regulate online activity originating outside their jurisdiction, they lack the means to enforce their laws. In private law there are some agreements for the reciprocal enforcement of foreign judgments which may ameliorate this situation but in public law such agreements have not been concluded. Therefore, enforcement jurisdiction imposes limits on state regulation of online activity as it means that infringements of the law cannot be sanctioned. Of course, cooperation between states at the enforcement stage would allow states to uphold their own laws even when they would otherwise lack power over foreign wrongdoers. To date there has been a lack of cooperation among states however, especially in the public law sphere. In the absence of such cooperation, states can take certain measures such as prosecuting symbolically without enforcement, prohibiting local actors such as credit card companies or Internet service providers from providing supportive services or blocking foreign illegal content. However, although these measures may cumulatively ensure greater legal compliance with states' regulatory regimes, they all have attendant difficulties. For example, prohibiting local actors from providing supportive services does not target the primary wrongdoers and can be considered unfair. Thus, these unilateral measures are unsuitable to be widely employed and cooperation between states is preferable. The lack of such cooperation means that enforcement jurisdiction remains territorial and so in practice it is difficult for states to regulate online activity originating outside their territories.

Despite the significant difficulties with the country of destination principle, it has been applied by the courts to some websites, though judges have not attempted to answer the question of whether the accessibility of any website in a particular country is enough for that country to assert jurisdiction. Distinctions have been drawn by the courts between sites that are intentionally targeted at the state or could be presumed to be so targeted and sites that are not so targeted, distinctions between passive and interactive sites, between commercial and non-commercial sites, between parties who had other offline contacts with the state and those who had not, and between websites that happen to have an effect in the state and sites which are directed at, and known to cause harm in, the state. Thus, in particular factual situations, justice may have been achieved by applying the country of destination approach, but it is clear that such an approach is not suitable for general application.

The country of origin principle and the regulatory ‘race to the bottom’

While the country of destination approach is flawed, the country of origin principle, whereby once a service provider is operating legally in one jurisdiction it can market its services in others without having to comply with further rules, has not found resonance with states. While such an approach would mean that online publishers could easily foresee their legal exposure and that the regulatory burden on them would be realistic, states have generally not been prepared to forego regulatory competence. This is so because states may wish to protect their citizens from harmful foreign content, the definition of which varies from state to state. A further objection to the country of origin principle is that, while it may ease the regulatory burden on the service provider, often that burden is simply shifted to the online consumer.

However, the most significant difficulty with an exclusive country of origin approach is that it would be likely to lead to a regulatory ‘race to the bottom’, as it encourages forum shopping. Forum shopping is problematic in that it is likely to cause de facto legal harmonization at the lowest level. Since the country of origin approach is likely to lead to ‘lowest common denominator’ regulation, it is only acceptable in so far as the lowest level of regulation is acceptable.

The choice of law problem as it relates to online gambling

The choice of law problem is particularly acute in the case of online gambling due to the total lack of consensus as to substantive law. If the country of origin principle is applied, gambling businesses would probably seek to minimize compliance costs by establishing themselves in a state with lenient or no regulatory requirements. Indeed, in the absence of legal certainty as to what law applies to online activity, many online gambling operators have already established themselves in jurisdictions such as Antigua, Costa Rica and the Netherlands Antilles, in the hope of reducing regulatory compliance costs. 14 However, the country of destination approach as applied to online gambling is also flawed due to the inability to enforce the laws of the destination state unless the service provider has some presence in that state. Thus, in order to enforce its laws, some states have resorted to arresting executives of online gambling service providers if they happen to enter that state. This is clearly an arbitrary and ineffective means of enforcement.

One might claim that the choice of law/enforcement problem could be circumvented and a border created between the EU and the rest of the world by blocking non-EU gambling sites. Such blocking of ‘undesirable’ content is possible from a technical point of view and has already been orchestrated in China, Cuba and Singapore. However, the imposition of such restraints is unlikely to be a feasible solution as such blocking would essentially involve large-scale censorship which would be unacceptable in Western democratic states. Thus, it does not seem possible to create a border between the EU and the rest of the world, by either legal or technological means. Therefore, the question of what is the most viable regulatory option will now be briefly discussed.

An analysis of the regulatory options available to the EU

Maintaining the status quo is the first regulatory option available to the EU. Some of the difficulties which would attend any attempt to reach an EU-wide consensus can also be perceived as indicators that this is the most desirable option. For example, it could be argued that regulation of gambling at EU level would force Member States to culturally converge. Another possible objection to EU regulation is that it would lead to the removal or loosening of Member State restrictions and this could lead to an increase in social problems such as problem gambling, bankruptcies and crime. However, it is questionable whether Member State restrictions, even if found to be justified, protect Member State citizens against the negative effects of gambling, particularly if the state monopoly or licensee is involved in aggressive and sophisticated promotion of its gambling services. 15 Prohibition and restriction of opportunities to gamble clearly do not eliminate the incidence of problem gambling and it is not known whether or to what extent, if any, they reduce it. In this regard, US studies as to whether the introduction or expansion of casinos leads to an increase in problem gambling and/or bankruptcies have yielded mixed results. 16 Some preliminary small-scale studies seem to point to an increased incidence of problem gambling among online gamblers. 17 However, this could be because existing problem gamblers are attracted to this more convenient form of gambling. More importantly, these studies do not consider whether restrictions on the provision of online gambling services have any positive effect. Thus, although further research into the socio-economic impacts of removing restrictions on gambling appears necessary, particularly within the EU and in relation to online gambling, there is as yet no conclusive evidence to support the proposition that the removal of some Member State restrictions would increase the negative effects of gambling.

Moreover, arguments against all forms of EU regulation must be balanced against the benefit of completing the internal market. Along with the impediment to market growth which it results in, maintaining the status quo is undesirable for a number of reasons. First, case law is reactive as the court can only respond to those cases brought before it. Second, case law is destructive, i.e. the ECJ cannot make positive recommendations as to the law. Third, the ECJ can only apply the law. It is not empowered to take into account the relevant economic context and must only apply the rights to free movement set out in the Treaty. The inability of the court to consider the surrounding economic and social context may, in some cases, limit its ability to implement the free movement rights in the Treaty. Fourth, negative integration gives rise to the problem of uncertainty, given the absence of a doctrine of precedent in the ECJ. Gambling service providers and EU citizens are not clear on what restrictions are lawful or not. Therefore, it is clear that maintaining the status quo in itself cannot lead to a single market for gambling services. However, not all forms of regulation are feasible or desirable so the feasibility of other regulatory options will now be assessed.

It is clear that harmonization of Member State legislation is unfeasible due to the current degree of divergence in Member States' approach to gambling regulation and the forced cultural convergence which would arise from harmonization. A second, and slightly more realistic, form of positive integration is the application of the (exclusive) country of origin principle. However, this approach has not found favour with states worldwide, either generally or in relation to gambling. Unless the majority of states worldwide adopt the country of origin principle, the states which do adopt it will gain little. Their own service providers will probably be subject to foreign law but foreign service providers will be able to operate in their territory without being subject to local law. In light of these difficulties, the application of the exclusive country of origin approach to gambling seems unfeasible.

Another possible solution would be the application of the country of origin principle but with certain carve-outs or derogations which would allow the country of destination to apply its law on certain crucial matters where justified (partial application of the country of origin principle). This approach represents a compromise between the interests of gambling service providers and the interests of Member States. The problem, however, is controlling the circumstances under which Member States can derogate from the country of origin principle. Guidance as to how partial application of the country of origin principle could be implemented in practice can perhaps be drawn from the E-commerce Directive. Under this Directive certain derogations from the country of origin principle are allowed for reasons such as public policy, public health, public security or the protection of consumers, provided they are proportionate and there is no less restrictive means available. The Member State seeking to derogate must first notify the Commission and the originating Member State, except in cases of urgency. The originating Member State must be given the opportunity of rectifying the situation. The notification to the Commission does not suspend the measures taken. The Member State of destination may implement the measures without waiting for the Commission's decision on whether the derogation is allowed or not. However, the Commission reserves the right to examine the compatibility of the measures with the country of origin rule and may ask Member States to refrain from taking the measures or to discontinue them. If a Member State does not comply with the notification obligation, the Commission may commence infringement proceedings against that state and the service provider against whom the measures were taken may take action before the courts in the destination Member State. 18

Whether such a scheme could work in the case of gambling is questionable. Uncertainties have arisen about the scope of ‘general good’ derogations even in less controversial sectors such as banking and insurance. 19 Thus, the scope of a gambling ‘general good’ derogation would surely be subject to uncertainties and the Commission would have to play an active role in ensuring that only derogations genuinely necessary to protect certain interests are permitted. Otherwise this approach would represent no improvement on the current approach. 20 Thus, although the derogation scheme set out in the E-commerce Directive may not be ideal, it could at least be a starting point.

It has also been suggested that different forms of gambling (such as betting services, casino services, bingo services and lottery services) be treated separately. This does not constitute a regulatory approach but rather is a means of dividing up the task. If it is considered desirable that all forms of gambling are (eventually) to be regulated in the same way, it nonetheless may be a sensible suggestion to regulate certain sectors first as Member States are less likely to oppose a second phase of regulation once they have experienced the effects of the first phase. The opening up of other previously closed-up markets such as the telecommunications, banking and insurance markets demonstrates the value of such incremental regulation. On the other hand, a case can be made for different forms of EU regulation in relation to different forms of gambling. In this regard, a distinction may be drawn between ‘hard gambling’, characterized by high event frequency and rapid staking, which present greater potential for exploitation and therefore a greater justification for strict regulation and other forms of gambling, which may warrant a lighter touch due to delayed game resolution, coupled with limited opportunities for repeat staking. Furthermore, different gaming sectors are characterized by different market structures, price sensitivities and tax regimes and also give rise to different types and levels of social costs. Thus, different forms of gambling may merit separate regulatory treatment at EU level, even though the ECJ case law thus far has largely failed to distinguish between different gambling forms. The partial application of the country of origin approach described above could possibly be flexible enough to cover all gambling forms.

Conclusion

On consideration of the main regulatory options, partial application of the country of origin principle seems the most feasible option. If partial application of the country of origin principle were to be the desired regulatory approach, it could be desirable to allow Member States to opt for host country control in the first regulatory stage, with the (partial) country of origin principle applying otherwise. In the insurance sector such an option was given to Member States in the earlier stages of the regulatory process. While the only two countries who opted against host country control were the UK and the Netherlands, it allowed the EU to implement the principles of the single licence, minimum harmonization and country of origin control at a later stage. 21 In relation to gambling, while most Member States would probably want to retain control as host countries, the UK would opt out of host country control as this would be consistent with the provisions of the Gambling Act 2005. One or more Member States opting out could pave the way for the eventual introduction of a general country of origin principle. Indeed, this phased introduction of the (partial) country of origin principle was the approach favoured by the Commission under the Services Directive, 22 though it was not accepted by Parliament at that time.

Another issue which would have to be considered if the suggested approach were to be applied is the enforcement of restrictions. If the partial country of origin approach were to be adopted, Member States would be largely reliant on each other for enforcement purposes. Thus, in order for Member States to permit (partial) application of the country of origin approach, they would have to have confidence in each other's ability (and willingness) to enforce their gambling laws. To this end, the EU regulatory framework should encompass guidelines as to how Member States can more effectively enforce gambling laws, in particular as against online operators established in their jurisdiction.

As the recent European Parliament Resolution demonstrates, regulation at EU level is not likely to occur in the near future (if at all) owing to the difficulties of reaching EU-wide consensus and the ambiguity as to its desirability. 23 However, some form of regulation would seem to be necessary as the minefield of regulatory risk that is the current EU gambling market damages the internal market more generally. While one might think gambling regulation to be a comparatively minor issue, in fact it has implications beyond the gambling sector, as it epitomizes the ongoing battle between Member State protectionism and EU internal market principles.

Notes

1. See Swiss Institute of Comparative Law (2006).

2. Williams and Wood (2007).

3. See Kurzer (2001).

4. See Case C-6/01 Anomar [2003] ECR I-8621; Case C-260/04 Commission vs. Italy [2007] ECR I-07083; Case C-243/01 Gambelli [2003] ECR I-13031; Case C-124/97 Läära [1999] ECR I-06067; Case C-42/02 Lindman [2003] ECR I-13519; Joined Cases C-338/04, C359/04 and C-360/04 Placanica [2007] ECR I-1891; Case C-275/92 Schindler [1994] ECR I-01039; Case C-67/98 Zenatti [1999] ECR I-07289.

5. A 2006 survey of Member State regulation of gambling described the differences between the Member States in their fiscal treatment of gambling services as ‘enormous’: see Swiss Institute of Comparative Law (2006, p. xxv).

6. See Report of the Casino Committee (2008, para. 5.7.2).

7. See Brady and Ramyar (n.d.).

8. Patton et al. (2006, pp. 682–684).

9. Betfair has indicated that it has experienced adverse reactions from authorities in most Member States when attempting to enter their markets (questionnaire conducted by the author answered by Scott Ferguson, Field Education Manager with Betfair).

10. Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments, OJ 2004 L145/1.

11. In a 2006 survey for the European Commission of Member State regulation of gambling, it is significant that spread betting was not mentioned other than in the section on the UK. In the UK sports spread betting firms are governed by the Financial Services and Markets Act 2000 and need to be authorized by the Financial Services Authority (FSA).

12. Whereas in other forms of gambling one's losses are determined by a combination of two factors nominally within one's control: the stake and the frequency of staking, spread betting losses are determined additionally by the degree to which the player gets the bet wrong. For example, in financial markets a minimum bet could typically be £5 a ‘tick’. This means that for every plus or minus 1 cent move in price of the underlying, the bettor gains or loses £5. While some sites have stop-loss/stop-win policies whereby the maximum loss or win which a bettor can make on a particular bet is predetermined, on other sites one must choose to set a stop-loss.

13. On the general issue of regulatory competence over online activity see Kohl (2007).

14. Research suggests there are around 2300 gambling websites worldwide. Antigua has the largest number with around 537 sites followed by Costa Rica (474), Kahnawake (Canadian Reservation) (401) and the Netherlands Antilles (343) (Department for Culture, Media and Sport, 2007).

15. Huls (2007, p. 73).

16. Swiss Institute of Comparative Law (2006, pp. 1439–1441).

17. One study found that 75% of those who bet on the Internet were either problem or pathological gamblers – compared with 20% of offline gamblers. See Cullen (2008).

18. Directive 2000/31/EC on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market, OJ 2000 L178/1, Article 3.

19. Tison (2002).

20. One may also argue that, since gambling was specifically excluded from the scope of the E-commerce Directive in 2000, the approach taken in that Directive is not suitable to apply to gambling. However, since 2000, the internal market in services is nearer completion and, given the enormous growth of online gambling, it seems to be an area requiring EU regulation.

21. Seatzu (2003).

22. Directive 2006/123/EC of the European Parliament and of the Council of 12 December 2006 on services in the internal market, OJ 2006 L 376, 27.12.2006, pp. 36–68.

23. See European Parliament resolution of 10 March on the integrity of online gambling (2008/2215(INI)). Retrieved from http://www.europarl.europa.eu/sides/getDoc.do? pubRef=-//EP//TEXT+TA+P6-TA-2009-0097+0+DOC+XML+V0//EN. This Resolution states that online gambling regulation should remain the purview of the Member States. However, it does see room for universal standards on issues like age limits, protection of children and problem gamblers and also sees the need to regulate gambling advertising.