(a) Schedule 1 and misleading practices

Schedule 1 lists 31 commercial practices which are always unfair, regardless of the impact on the consumer. Some of these are misleading practices and might be relevant to loot boxes. For example, paragraph 7 prohibits a trader from:

There might be occasions where this is used as a tactic to persuade consumers to purchase a loot box quickly. However, this will only be relevant where the statement that the product is available only for a limited time is ‘false’; in the context of offering loot boxes it will more commonly be true.Footnote ²⁷ Similarly, paragraph 27 of Schedule 1 prohibits ‘including in an advertisement a direct exhortation to children to buy advertised products…’ While such communications are more likely to be regarded as marketing communications than advertising, CAP Guidance specifically states that ‘[m]arketing communications directed at children should not include a direct exhortation to buy an advertised product’.Footnote ²⁸

If the practice is not specifically prohibited, the second step is to consider whether a practice is unfair under one of three ‘small general clauses’:Footnote ²⁹ as a misleading action; misleading omission; or aggressive practice. Regulation 5(2) provides that a commercial practice will be a misleading action:

In addition, it must be demonstrated that the misleading action ‘(b)…causes or is likely to cause the average consumer to take a transactional decision he would not have taken otherwise’.

The practice does not have to contain false information and could also be a misleading practice if it is technically accurate but deceptive.Footnote ³⁰ It must be untruthful or deceptive as to the matters in paragraph (4), which include both the nature and the main characteristics of the product. This provision will be relevant in some cases (for example where the game provides inaccurate odds about particular outcomes). Recently published Guidance points out that ‘[i]n-game purchasing and advertising happens within a unique context of gameplay, time pressure and (in some cases) chance, and therefore brings with it specific risks of misleading consumers’.Footnote ³¹ However, it will not be applicable to many of the concerns considered in this paper, where other provisions are more appropriately utilised.

An alternative is to proceed on the basis that there has been a misleading omission. Regulation 6(1) states that a commercial practice is misleading if: ‘in its factual context, taking account of the matters in paragraph (2)’ it omits or hides material information, provides such information ‘in a manner which is unclear, unintelligible, ambiguous or untimely’ or (except where it is already apparent) fails to identify its commercial intent. Again, the practice must cause or be likely to cause the average consumer to ‘take a transactional decision he would not have taken otherwise’. The matters in paragraph 2 are:

1. (a) all the features and circumstances of the commercial practice;

2. (b) the limitations of the medium used to communicate the commercial practice (including limitations of space or time); and

3. (c) where the medium used to communicate the commercial practice imposes limitations of space or time, any measures taken by the trader to make the information available to consumers by other means.

‘Material information’ in this context means both the information ‘which the average consumer needs, according to the context, to take an informed transactional decision’ and ‘any information requirement which applies in relation to a commercial communication as a result of a Community obligation’. Whether information is material will be determined on a case-by-case basis.Footnote ³²

Spence-Jones and Xiao view it as ‘clearly arguable that omitting to publish odds relating to items contained in video game loot boxes can be a misleading omission’ and regard this as the simplest route for enforcers to follow.Footnote ³³ This is a possible avenue, although the conclusion of the Court of Appeal in PLT Anti-Marketing that there is ‘no obligation on a supplier to identify its particular mark-up’ on products it supplies might imply that odds need not be disclosed.Footnote ³⁴ King and Delfabbro argue that ‘there seems to be a growing need for consumer protection measures for in-home purchasing systems’.Footnote ³⁵ However, they again focus heavily on information remedies, such as the display of age restrictions on games and the providing of regular statements on spending activity. This reflects the approach taken in China, where game developers are required to disclose the odds of receiving items from loot boxes on the assumption that doing so will facilitate informed decision-making.Footnote ³⁶ It has been suggested that the disclosure regime in China has not been wholly successful.Footnote ³⁷ In particular, while odds have been disclosed in the majority of cases, they have frequently not been provided in a prominent and transparent manner.Footnote ³⁸

To fall foul of the provisions on misleading actions and omissions (and also, as will be seen later, aggressive practices) the practice must cause or be likely to cause the (average) consumer to take a transactional decision they would not have taken otherwise. ‘Transactional decision’ is defined as meaning:Footnote ³⁹

Transactional decisions take a variety of forms. The most common of these will be the decision to make a purchase, and this will be the most relevant form in the context of loot boxes. But other examples could include the decision to click through a website following a commercial offer, to terminate a contract or switch to a new service provider.Footnote ⁴⁰ This demonstrates that it is not enough for a commercial practice to be unfair; it must be unfair in such a way that it is liable to change an average consumer's behaviour in relation to the product at issue.

(b) The limitations of information

It is clear that there will be situations where the marketing of loot boxes will involve a breach of Schedule 1, and others where there will be a misleading action or omission. A common rationale for intervening in the relationship between traders and consumers is market failure, and regulation aimed at the correction of one form of market failure, information asymmetry, forms the basis of a significant amount of consumer protection law.Footnote ⁴¹ Where traders provide information that is false, or present information in a deceptive manner, the provisions on misleading actions are likely to be relevant. Where consumers lack the ‘material information’ they need to make an informed transactional decision, those provisions on misleading omissions will apply.

However, there are good reasons to view reliance on these provisions as insufficient. First, the provision of information has long been regarded as an inadequate tool for protecting consumers.Footnote ⁴² Even where disclosure is prominent, it may be of information that is difficult for consumers to understand. It has been noted that loot box probabilities are more complex than gambling probabilities,Footnote ⁴³ suggesting that the critiques of disclosure in the context of gambling may be repeated in the context of loot box probability disclosures.Footnote ⁴⁴ Furthermore, there is ample evidence from behavioural sciences that consumers are prone to a host of cognitive biases, which cast significant doubt on the utility of expecting them to act rationally on the basis of information.Footnote ⁴⁵ In relation to gambling, over-optimism bias is particularly relevant, with research demonstrating that it may lead to disclosed odds being routinely ignored.Footnote ⁴⁶ The existence of other biases, from the gamblers’ fallacy to the near miss effect and the sunk cost fallacy also counsel against seeing disclosure as a panacea.

Second, to focus on information is to miss the essential character of the wrong. It is submitted that loot boxes are more typically unfair, not primarily because they are offered in circumstances of information asymmetry but because they involve the manipulation of the gamer by the trader through practices that are aggressive. Misleading conduct may form part of this, but the wrong goes beyond that. Furthermore, provisions on aggressive commercial practices will remain relevant even where there is no misleading action, and even if there is disclosure of ‘material information’ under the Regulations. As explained in detail below, the design decisions that surround loot boxes make this particularly likely in the context of gaming. In short, challenging the marketing of loot boxes as an aggressive commercial practice will better reflect the character of the unfair practice.

(a) Aggression, immersion and game design

To understand why the offering of loot boxes can be viewed as an aggressive practice it is important to think about the design of games and, in particular, the methods that are used to ‘immerse’ the consumer.

A variety of design features can be used to make games so compelling that gamers become immersed in them. A helpful taxonomy is provided by King et al.Footnote ⁵⁰ First, there are aesthetic features, such as the game's graphical and sound design. Secondly, there are story features, such as the main narrative and any side narratives. Thirdly, there are character features, such as the way a character develops over time. Finally, there will be gameplay features, such as the way that the game is controlled, whether multiplayer modes of gameplay are possible and the tasks that the player has to perform with the character.Footnote ⁵¹

The more attractive these design features are in aggregate, the more that a player is likely to become immersed in a game. Whilst immersion can sometimes be positive, in the context of loot boxes it has the potential to be highly damaging; this is largely because once immersed, a gamer may feel compelled to continue playing and may then make decisions that lead to detriment.Footnote ⁵² There are different forms of immersion that may operate. One example has been labelled narrative immersion, which reflects a player becoming so immersed in the story of the game that they feel compelled to complete it and so finish the tale.Footnote ⁵³ Another is challenge-based immersion, where the player becomes preoccupied by the challenge of the game and is driven by the compulsion to ‘win’.Footnote ⁵⁴ Where virtual reality (VR) and augmented reality (AR) technologies are utilised, so-called ‘system immersion’ may also arise; this occurs where technology functions to place the individual in the game environment.Footnote ⁵⁵ However, as VR and AR remain niche technologies, most games that utilise loot boxes do not exhibit system immersion as the primary immersive modality. The design of the game may therefore take the gamer through degrees of absorption: from engagement, through engrossment to – ultimately – total immersion.Footnote ⁵⁶

(b) Undue influence and coercion

The significant impairment of choice or conduct is at the heart of an aggressive commercial practice and, in the case of loot boxes, this might arise through undue influence or coercion.Footnote ⁵⁷ Undue influence is defined in regulation 7(3)(b) as meaning:

By contrast, the Regulations do not define coercion; they merely state (somewhat unnecessarily) that the concept ‘includes the use of physical force’.Footnote ⁵⁹ The inclusion of physical force was never in doubt. More significant is the extent to which a trader can be found to have coerced a consumer in ways that do not include such force. The deployment of psychological pressure is far more common than the use of physical pressure in commercial contexts, and psychological pressure is particularly significant where technology is concerned.

It is vital here to try to identify what it is that makes the offering of loot boxes within a game an aggressive commercial practice, whether by undue influence or coercion. There are different types of loot box to consider.

When a gamer is immersed in a story or gameplay, loot boxes may function in two main ways. First, they may increase the probability of a consumer's success in a game. These are so-called ‘pay to win’ loot boxes. Secondly, they may enable a consumer to avoid what is called ‘grinding’.Footnote ⁶⁰ Grinding occurs where consumers continually engage in repetitive gameplay in order to enhance their characters’ abilities or obtain particular items necessary for progression to the next area of the game. ‘Pay-to-skip’ loot boxes can help to overcome this.

Both pay to win and pay to skip loot boxes are potentially aggressive for the purposes of regulation 7. The pressure that the game's design creates on the part of the gamer results from a combination of factors. These include (but may not be limited to) the immersion in the game, the difficulty of the game and, where relevant, the multiplayer nature of the game. When combined with the need to take a transactional decision quickly, it is more readily apparent why this is aggressive. But it is worth delving a little more deeply into the nature of that aggression.

First, the conduct might be viewed as involving undue influence. This case might be constructed as follows. Gamers may make a very significant investment (of money, time and effort) in the game. The trader is aware of this and, as already seen, may have used a variety of techniques to draw the gamer in and immerse them. The psychological pressure that results creates a significant position of power to the benefit of the trader. This power asymmetry can readily be exploited in a way that significantly limits the gamer's ability to make an informed decision. The gamer is likely to be deeply immersed at the point that they have to decide whether to make the relevant transactional decision (in the form of purchasing a loot box). The dynamics of the game will be highly relevant here. Part of the pressure placed upon the consumer arises from the difficulty of the game, and the trader may have developed it such that advancement is difficult, improbable or impossible without a further purchase.Footnote ⁶¹ The time-limited nature of communications is liable to increase this pressure further. As the CAP has observed, when in-game purchasing occurs within immersive gameplay ‘there is a natural sense of urgency that may accompany decision-making that is uncommon in other forms of media’.Footnote ⁶²

It is important to reflect further here on the relationship between aggressive practices and information. As noted above, it is sometimes argued that any harm from loot boxes can be tackled through the disclosure of information. However, this is not persuasive. If the consumer has been informed in advance (for example when purchasing the game) of the probability of winning a particular item, it is unlikely that they will be in a position to use that in an informed way at the point when they have to make their (later) transactional decision. Furthermore, even if the odds were provided at that later point when the consumer has to make the transactional decision (ie in the game), it seems fanciful to conclude that the consumer is truly making an informed choice because of the pressure of the situation in which they find themselves.

The provision of information is clearly relevant to the assessment of whether a practice constitutes undue influence. Undue influence involves the significant limiting of the consumer's ability to make an informed decision. It cannot be argued that a practice inevitably involves undue influence simply on the basis that a consumer's choice was not fully informed. Indeed, the recent case of Orange Polska Footnote ⁶³ appears relatively restrictive in this regard. That case raised the question of whether a practice involved undue influence (the language of the Polish provision was ‘impermissible influence’) when a courier insisted that a consumer sign a contract. The Court concluded that:

It is correct that a practice will not automatically be aggressive merely because consumers do not have the opportunity to peruse information at their leisure. However, there will be situations where factors combine to mean that the lack of time a consumer has to make an informed choice can be strong evidence of aggression. In Wind Tre the CJEU emphasised that the concept of an aggressive commercial practice was defined by the fact that it impairs or is likely to significantly impair the average consumer's freedom of choice or conduct with regard to the product.Footnote ⁶⁵ It concluded that ‘it follows that for a service to be solicited the consumer must have made a free choice. That supposes, in particular, that the information provided by the trader to the consumer is clear and adequate’. Footnote ⁶⁶

Lack of information will, therefore, be a factor to be considered in the assessment of whether there is undue influence. In Orange Polska, it was found that undue influence will exist where the commercial practice is ‘liable to make that consumer feel uncomfortable and thus to confuse his thinking in relation to the transactional decision’.Footnote ⁶⁷ While this is most likely to occur where trader and consumer are face to face, such discomfort and confusion might arise within the high pressure environment of gaming.

As noted above, the offering of loot boxes might be described either as involving undue influence or as involving coercion. Willett suggests that as a result of the definition of the former, the consumer's restriction of choice in undue influence ‘must specifically be a restriction of choice that results from an information problem’.Footnote ⁶⁸ Although it is submitted that decision-making in relation to loot boxes will seldom be fully informed, coercion will in some respects be a broader and more flexible concept. Despite lacking a clear legislative definition, ‘coercion’ might be viewed as better reflecting the nature of the wrongdoing than undue influence. For example, there is no need for a position of power for a finding of coercion (although the presence of one is likely); the assessment is centred more on the precise nature of the trader's behaviour. Coercion seems to involve compelling or inducing someone to act in a particular way by improper means. Like undue influence, it can only be assessed by taking account of all the circumstances. The pressure (which is created in large part by the trader and will certainly be known to them) is thus central to the coercive character of an invitation to purchase a loot box. In both forms of aggressive practice, there is a significant compromising of the gamer's autonomy. In the context of consumer decision-making, autonomy may be viewed as ‘the absence of vulnerability or the capacity to act upon the market in the consumer's self-interest’.Footnote ⁶⁹ The circumstances in which loot boxes are commonly offered are those which cast doubt on autonomous, self-interested decision-making. Looking back to the definition of aggressive commercial practice, the reference to the restriction of freedom of choice stands out. The consumer who does not have the opportunity to reflect adequately on decision making as a result of pressure generated by the trader lacks the requisite freedom of choice and may be said to have been coerced.Footnote ⁷⁰

The argument that the offering of loot boxes may be an aggressive commercial practice, whether through undue influence or coercion, is strengthened by the wording of regulation 7(2). This provides that in determining whether a commercial practice uses harassment, coercion or undue influence, account shall be taken of a range of factors. Some are particularly relevant in the context of loot boxes. For example, the provisions refer to the ‘timing, location, nature or persistence’ of the practice. Timing may be especially apt in this context. Loot boxes are typically offered to consumers at particular points in a game, and it will be important to consider the dynamics of the individual game in assessing whether that offer is aggressive. It is recognised that some practices may not be aggressive when they take place at certain times, but aggressive at other times. This could relate to times of the day (such as a telephone call at night) or, as in this case, at key points within a commercial activity. The loot box will commonly be offered when a consumer is at a crucial point in a game and thus most susceptible to making a purchase. This may be when the game is grinding, the gamer is frustrated, and it seems that the only way to advance is to take the chance on a loot box. This pressure will be the greater the more immersed the gamer is. The coercive character of the practice may be further increased by manipulative messaging. The OFT Principles offer the example of a message telling players to ‘feed [a game character] ice cream or [the character] will be unhappy’.Footnote ⁷¹ Similarly, the Advertising Standards Authority held that direct exhortions to purchase paid content forming part of freemium online games were in breach of the Code of Advertising Practice, particularly where paid membership was presented to children as necessary to improve popularity and satisfy in-game characters. In the Mind Candy adjudication,Footnote ⁷² which was concerned with the game Moshi Monsters, invitations to join the paid premium version of the game were accompanied by statements that ‘The Super Moshis [a type of in-game character] need YOU’ and ‘Members are going to be super popular’. Such statements, when accompanied by a ‘Join Now’ command were likely to ‘put pressure on young players to purchase the subscription that would allow them to take part in this aspect of the game’. The more time-critical the decision is, the more likely the exhortation is to be aggressive.

Another factor that regulation 7 mentions is ‘the exploitation by the trader of any specific misfortune or circumstance of such gravity as to impair the consumer's judgement, of which he is aware, to influence the consumer's decision with regard to the product’. This provision was designed to tackle conduct where traders exploit misfortune that they know to have befallen the consumer and which impairs the consumer's judgement. One example might be a funeral director who persuades a recently bereaved client to purchase an expensive coffin by associating such expenditure with the consumer's devotion to the bereaved. However, the circumstances within a game where a loot box is offered for purchase should be taken into account. Compared with the funeral example, the offer of a loot box within a game may appear somewhat trivial, but this underplays the extent to which consumers become immersed in games and the extent to which they become deeply attached to characters. This point should not be pressed too far, but the connection that gamers can feel towards their characters (which are frequently viewed as extensions of themselves) should not be underestimated. As will be seen below, this is particularly pertinent where more vulnerable individuals are concerned.

As an aggressive commercial practice ‘significantly impairs or is likely significantly to impair the average consumer's freedom of choice or conduct in relation to the product concerned’. It is important to consider what is meant by ‘freedom of choice or conduct’. In one sense, of course, consumers are ‘free’ to make a choice when offered loot boxes. They can decide that it is not so important to win the game after all. They can ‘choose’ to resist the temptation to take a chance, remaining mired in grinding loops and conclude that progressing to the next chapter of the game's story or to a new challenge is not worth it. However, there must be a strong argument that the deep frustration generated by being unable to progress in particular ensures that the freedom to exercise choice is substantially impaired.Footnote ⁷³

A final point to reiterate is that the aggressive practice must cause or be likely to cause the (average) consumer to take a transactional decision they would not have taken otherwise. As noted above, ‘transactional decision’ means:Footnote ⁷⁴

The transactional decision requirement appears to permit a trader to argue that while there was undue influence, coercion or harassment, and while this significantly impaired or was likely significantly to impair, the average consumer's freedom of choice or conduct, it was nevertheless unlikely cause him to take a transactional decision he would not have taken otherwise.Footnote ⁷⁵ However, once it is established that the average consumer's freedom of choice or conduct is substantially impaired, it surely follows that the average consumer might act differently. The transactional decision test appears superfluous and should not prove a barrier to enforcement action in practice once the other elements are demonstrated.

(a) Standard average consumers

Where the commercial practice in question ‘reaches or is addressed to a consumer or consumers’ the Regulations state that ‘account shall be taken of the material characteristics of such an average consumer including his being reasonably well informed, reasonably observant and circumspect’. This ‘standard average consumer’ benchmark is the default position and will typically be applied where commercial practices are not targeted at particular groups of consumer.Footnote ⁷⁷ It is the test that the Court of Justice developed in a series of cases culminating in Gut Springenheide.Footnote ⁷⁸ According to OFT v Purely Creative it reflects ‘the common sense proposition that the Unfair Commercial Practices Directive exists to protect from being misled consumers who take reasonable care of themselves, rather than the ignorant, the careless or the over-hasty consumer’.Footnote ⁷⁹ The wording follows EU case law on trademarks and implies that a relatively objective test should be applied as a benchmark.Footnote ⁸⁰ The Commission's Guidance views the (standard) average consumer as ‘a reasonably critical person, conscious and circumspect in his or her market behaviour’.Footnote ⁸¹ Precisely how strict this test is remains debateable. Abbamonte argues that it assumes that consumers should behave ‘like rational economic operators’ in that they ‘should inform themselves about the quality and price or products and make efficient choices’.Footnote ⁸² Incardona and Poncibo share Abbamonte's interpretation, but express regret that it departs from ‘the unpredictable realities of individual human behaviour’.Footnote ⁸³ By contrast, Weatherill suggests that the test is appropriately (and more) sophisticated and flexible in its vision of the consumer. He concludes that the standard average consumer ‘is smart enough to (for example) process disclosed information, but he or she is no perfect rational actor’.Footnote ⁸⁴ It is true that national courts appear disinclined to adopt the strict standard that Abbamonte posited. Those courts have some discretion when applying the test. Recital 18 to the Directive makes clear that the average consumer test ‘is not a statistical test’ and that national courts and authorities have to exercise their own judgement to determine what an average consumer would understand and how he or she would behave. Case law reveals that this discretion is frequently exercised in a relatively consumer-friendly manner.Footnote ⁸⁵ Duivenvoorde compared the application of the benchmarks in Member States to the case law of the CJEU and concluded that: ‘none of the Member States researched illustrate the emphasis of the CJEU's case law on the consumer's responsibility to beware of potentially unfair commercial practices’. Footnote ⁸⁶

Of particular concern from the perspective of consumer protection is the offering of loot boxes to consumers who may be less well-informed, observant and circumspect than the average. The Regulations allow the interests of such consumers to be considered through the average targeted benchmark and the average vulnerable benchmark.

(b) Average targeted consumers

The second type of ‘average consumer’ is identified in regulation 2(4). This provides that where the commercial practice is ‘directed to a particular group of consumers’ the average consumer is ‘the average member of that group’. This is the ‘average targeted’ standard.Footnote ⁸⁷ In some respects, this may be the most appropriate test to apply when considering how loot boxes should be assessed as a commercial practice. The purpose of the average targeted test is to ensure that where a trader directs a commercial practice towards a particular group, the practice is judged by the benchmark of the average person within that group. The commercial practice of most relevance is the offering of loot boxes, for example when a consumer is struggling to proceed.Footnote ⁸⁸ However, it is important to consider this against the background of the initial sale and marketing of the game in which the loot box is found. In most cases there will be no question of an aggressive practice at that earlier stage and it may be that there is no misleading practice. However, CAP Guidance suggests that the presence of loot boxes within a game may be material to some consumers, particularly those with specific vulnerabilities. For this reason any advertising should make clear that the game includes forms of random-item purchasing, as well as providing information of the type of item, in a prominent manner.Footnote ⁸⁹ Furthermore, the initial marketing remains relevant because it frames and contextualises the later commercial practice. The OFT Principles provide a useful, albeit non-exhaustive, list of factors that may be considered in determining whether a game is targeted at children. This includes: the inclusion of characters popular with or likely to appeal to children; cartoon-like graphics; bright colours; simplistic gameplay and/or language; that the game concerns an activity that is likely to appeal to or be popular with children; that the game is available to be downloaded, signed up to or purchased by anyone and is not age-restricted; and that the game is featured in a children's section of an app store. Footnote ⁹⁰ Where there is targeting, the practice in question will be assessed against the benchmark of the average member of that group.

Alternatively, a game may not be aimed at a particular group, but may be likely to be played by members of that group. This again forms important background to the argument that the subsequent offer of a loot box is aggressive. This is because that subsequent offer is likely to be aimed at, or to affect, a consumer who is a member of the relevant group.

(c) Average vulnerable consumers

Traders may argue that their games (and the loot boxes which they contain) are not ‘targeted’ at particular groups of consumers. They are available to all (subject to age classifications) and will be played by a heterogeneous range of individuals. However, a practice or product need not be aimed at a particular group of consumers for the courts to take account of any vulnerabilities that consumers who engage with it may be likely to possess. The Regulations state that:

1. (a) where a clearly identifiable group of consumers is particularly vulnerable to the practice or the underlying product because of their mental or physical infirmity, age or credulity in a way which the trader could reasonably be expected to foresee, and

2. (b) where the practice is likely to materially distort the economic behaviour only of that group, a reference to the average consumer shall be read as referring to the average member of that group.

This ‘average vulnerable’ consumer benchmark was one of the more controversial aspects of the Directive. An objective test based on the assumption that consumers could be expected to be reasonably well informed, reasonably observant and circumspect (which largely reflected the position in UK law) was unacceptable to some member states.Footnote ⁹¹ German courts, for example judged commercial practices from the perspective of the gullible consumer, the ‘flüchtigen und unkritischen durchschnittsverbraucher’. The final iteration of the average consumer test represented a trade-off which, notwithstanding its limitations, can help address the interests of consumers when it comes to loot boxes.

The average vulnerable benchmark only applies where there is a ‘clearly identifiable group of consumers’ who are particularly vulnerable to the practice or the underlying product because of specific causes, namely: mental or physical infirmity, age or credulity. In addition, the vulnerability must be something that the trader could reasonably be expected to foresee. It can apply both where no group is targeted (but the trader should have realised the vulnerability of a particular group) and where one group is targeted but the trader should have appreciated the vulnerability of another group.Footnote ⁹²

There are various ways of grouping consumers who may be particularly susceptible to the detriment arising from loot boxes and some fall outside those elements specified in the average vulnerable test.Footnote ⁹³ However, the elements therein provide a helpful framework for analysis. The final part of this paper draws out the main causes of vulnerability identified in unfair commercial practices law and considers the implications of applying these benchmarks to assess the offering of loot boxes. This is relevant both when loot boxes are targeted at particular groups of consumers, and when they are merely likely to be received by them.

(a) Loot boxes and younger consumers

Recital 18 to the Directive explains the rationale for the average targeted standard, stating that where a practice is aimed at a particular group of consumers ‘such as children’ it is ‘desirable that the impact of the commercial practice be assessed from the perspective of the average member of that group’. Games will frequently be targeted at children and the loot boxes they contain similarly so.Footnote ⁹⁴ Many games are, of course, not aimed at children. Some are demonstrably not, something which a trader may attempt consciously to demonstrate through labelling and age classification. However, as explained above, a commercial practice will be assessed from the benchmark of the average vulnerable consumer where there is a ‘clearly identifiable group of consumers’ who are particularly vulnerable to the practice or the underlying product because of specific causes, one of which is age. Products not aimed at younger consumers will frequently be foreseeably likely to be used by them. According to recent research by the Gambling Health Alliance, 83% of children and young people play video games when they are below the set Pan-European Game Information (PEGI) age classification.Footnote ⁹⁵ It will, therefore, potentially be relevant to assess a commercial practice by the benchmark of the average child. To be aggressive, the practice must at least be ‘likely significantly to impair the average consumer's [in this case average child's] freedom of choice or conduct in relation to the product concerned’. It must also cause or be likely to cause the average child to take a transactional decision they would not otherwise have taken.

There is a clear difficulty in using the term ‘average child’, as ‘children’ covers a range of ages. The court or other body deciding the matter will need to determine the appropriate benchmark on the basis of: (a) the age range of consumer targeted; and/or (b) the age range of consumers likely to be affected by a product. It may be easier to assess the benchmark for the former than the latter, although it will not always be straightforward. Applying an age range seems an appropriate way of capturing the wrong, with particular attention being paid to those within that range whose age makes them particularly vulnerable to the practice or product.

A key issue to consider is how age makes consumers more vulnerable and thus how the (likely) age or age range of the consumer affects whether a practice can be regarded as aggressive. There are several elements to this and it is perhaps most helpful to think of ‘younger’ consumers, covering both children and adolescents.Footnote ⁹⁶

First, it is well-established that children and adolescents are likely to perceive information differently from adults.Footnote ⁹⁷ While this is particularly relevant to whether a practice should be regarded as misleading, it is also important to whether it is aggressive, for the reasons considered above. One important factor is that children and adolescents are typically less experienced than older consumers. In relation to the definition of average consumer in the Regulations, they can expected to be less well-informed or observant than adults, and so less able to make informed transactional decisions.

Second, children may be more liable to feel and succumb to pressure than adults. Many videogames are clearly aimed at adolescents (as well as likely to be played by them). Adolescents may be especially susceptible to peer pressure.Footnote ⁹⁸ They are also more likely than many other groups to suffer from low self-esteem.Footnote ⁹⁹ These experiences and traits will be relevant in many contexts. However, they may be particularly pertinent where multiplayer and co-operative games are concerned. Younger consumers, and especially adolescents, may feel significant pressure to spend money in order not to let their team down, or to fit into the prevailing social structures of a game.Footnote ¹⁰⁰ Research has identified that motivations for purchasing loot boxes are varied, but include social factors, such as ‘gaining status and approval or as part of a group experience’ as well as fear of missing out.Footnote ¹⁰¹ While this pressure will apply to consumers of different ages, it may be particularly strongly felt by adolescents. According to a survey for the Gambling Commission, 52% of young people said that they had heard of in-game items and 44% of those said that they had paid money to open loot boxes in order to try to get such items.Footnote ¹⁰² Consumers in that age range may be particularly susceptible to pressure. Batat concludes that ‘inexperienced, unknowledgable and unconfident adolescents may lack self-belief to such a degree that they suppress their better judgement’.Footnote ¹⁰³ In the language of the average consumer benchmark, they can be expected to be less circumspect than older consumers. The Royal College of Psychiatrists has emphasised the connection between information and pressure, arguing that ‘children are less prepared to deal with the potentially addictive nature of some modern computer games and less able to make informed decisions about spending’.Footnote ¹⁰⁴

There is ample evidence that traders are aware of the attractiveness of loot boxes to children. A 2019 Federal Trade Commission workshop on loot boxes suggested that they involve ‘predatory tactics to encourage addictive consumer spending, particularly in children’.Footnote ¹⁰⁵ There are several ways in which this is done, as the OFT Principles attest. The Commission's Guidance on the UCPD similarly observes that ‘Online games or applications that are likely to concern children or teenagers, as a vulnerable group, are usually not solely targeted at children’ but recognises that ‘they often use cartoons or other features children or teenagers are typically attracted by’.Footnote ¹⁰⁶

A further concern arises from the tendency to use harmful techniques such as implying that characters (and by extension the gamers controlling those characters) will not be popular unless particular sartorial choices are made. The OFT Principles give the following example:

In some cases, the gamer will be offered the chance to buy such an item, but in others they will be offered the chance to win it (in the form of a loot box). Cosmetic items in loot boxes are commonly referred to as ‘skins’ and are considered below. They raise slightly different issues and involve slightly different pressures from loot boxes that relate to game play. But they are a particular concern for younger consumers in the context of cooperative games.

(b) Mental or physical infirmity

Although it is possible that games are targeted at those with a mental or physical infirmity, it is more likely that the average vulnerable benchmark will come into play. That test only applies where the consumer is vulnerable ‘in a way which the trader could reasonably be expected to foresee’. The Commission Guidance points out that the foreseeability requirement ‘becomes relevant each time one needs to establish whether a given trader could have reasonable expected this practice to appeal in particular to vulnerable groups’.Footnote ¹⁰⁸ It is submitted that the probability of there being consumers whose vulnerability stems from mental or physical infirmity is sufficiently high that this should not be a difficult hurdle for the enforcer to overcome.

The relationship between gaming and health is complex and a detailed examination is beyond the scope of this paper. However, there will be consumers whose physical and mental infirmity is inextricably connected with gaming, and who are particularly susceptible as a consequence. According to the 11th Revision of the International Classification of Diseases, ‘gaming disorder’ is a pattern of behaviour which is characterised by:

A diagnosis of gaming disorder requires the behaviour to be sufficiently serious to result in ‘significant impairment in personal, family, social, educational, occupational or other important areas of functioning’. It should also have been evident for at least 12 months.Footnote ¹¹⁰ Conditions linked with gaming disorders include anxiety, depression, obesity, sleeping disorders, and stress.Footnote ¹¹¹

It is understandable (and appropriate) that a significant amount of attention has been paid to the effects of gaming on physical and mental health. As one commentator puts it: ‘[g]amers need to be educated on how to protect their thumbs, wrists, and elbows, their waistlines, their emotional state, their sleep, and their eyes’.Footnote ¹¹² However, for the purposes of the law on aggressive commercial practices, the key point is that consumers with gaming disorder are particularly vulnerable to the activities of traders and likely to take transactional decisions that they would not otherwise have taken. An inherent part of gaming disorder which is particularly important in this context is a diminished ability to control impulses and therefore to make informed and voluntary choices. The similarities between gaming and gambling have already been emphasised. Compelling evidence has recently been advanced of the close relationship between problem gambling and loot boxes.Footnote ¹¹³ Drummond stated that loot boxes are ‘designed to exploit potent psychological mechanisms associated with the development and maintenance of gambling-like behaviours’.Footnote ¹¹⁴ Most recently, Close and Lloyd describe them as having ‘structural and psychological similarities with gambling’.Footnote ¹¹⁵ The full extent to which loot boxes are a gateway to problem gambling – as opposed to problem gamblers being drawn to games with loot boxes – is not clear. While problem gamblers are likely to be drawn to loot boxes, and traders should foresee that, it has been suggested that there is now enough evidence to reliably conclude that loot boxes cause problem gambling Footnote ¹¹⁶ At the very least ‘relationships’ between engagement with loot boxes and problem gambling have been confirmed in a large number of studies.Footnote ¹¹⁷ In the words of Tristan Harns:Footnote ¹¹⁸

The probability and foresseability that loot boxes will be purchased by those with conditions such as gaming or gambling disorder and exploitation that such a relationship would involve suggests that the offering of such products may be aggressive and should be assessed by the average vulnerable benchmark.

(c) Credulity

The final element of vulnerability to consider is credulity. A difficulty with applying the credulity element of the benchmark (paraphrased by the EU's Guidance as ‘naivety’) concerns identifying who might comprise a clearly identifiable group of consumers that is particularly vulnerable through credibility (independent of age or infirmity). However, if it were found that the problem gambling or gaming falls short of mental or physical infirmity, it might still be argued that problem gamblers form clearly identifiable group who are particularly susceptible to manipulative practices on the basis of high levels of credulity.

Whether we argue that those particularly susceptible to the offer of loot boxes are suffering from infirmity or are particularly credulous in a way that falls short of that, it is important that conclusions are based upon evidence.Footnote ¹¹⁹ There is a significant body of literature within behavioural science that recognises how consumers perceive information and make decisions. In some cases the exploitation of cognitive biases will be at the heart of the wrongs associated with the offering of loot boxes. Several examples of these biases have been discussed in the context of loot boxes.Footnote ¹²⁰ One is the ‘gambler's fallacy’ where an individual believes that they are more likely to win following a long period of losses. Another is the ‘near miss effect’ where narrowly failing to win incentivises a person to continue. A third cognitive bias that is well-known is the ‘sunk cost fallacy’ where people feel compelled to continue spending having incurred expenditure so as not to feel or appear wasteful. These biases will be found in many people and while there has been discussion of the role of law in ‘debiasing’, achieving this will be difficult.Footnote ¹²¹ However, it will be possible to argue that problem gamers/gamblers, even if they cannot be said to possess a mental or physical infirmity, may form a clearly identifiable group who are particularly vulnerable to the techniques used to offer loot boxes. On this basis, those practices should be assessed by the standard of the average member of that group.

There might be other groups for whom the credulity test is applicable. Engagement with loot boxes is particularly linked with consumers who are male, young (which in relation to children and adolescents is considered above) and of lower educational achievement.Footnote ¹²² It does not seem a stretch to argue that such consumers are likely to be more ‘credulous’ than those of average educational achievement. Given that this is foreseeable (it is found in the literature) this may placer higher expectations on traders to ensure that their practices are not aggressive to such consumers.

(d) Vulnerability, loot boxes and skins

The focus above has been on loot boxes which contribute towards gameplay, either by helping the gamer to win or progress. An additional form of loot boxes are those which are cosmetic but highly valuable to gamers. The primary example is what are commonly referred to as ‘skins’. Skins are costumes or similar adornments that can be applied to a player's character to change their appearance.Footnote ¹²³ For example, Activision Blizzard's Overwatch includes the purchase of loot boxes which may include skins, emotes, victory poses, voice lines, sprays and highlight intros which do not affect the game, but which affect how the player presents themselves in the game.

These examples of loot boxes present some different challenges from those on which this paper focuses. However, it should not be concluded that merely because they do not present the same pressures as those examined, they cannot be aggressive. In such cases, the aggressive character of the commercial practice may stem from traders leveraging the immersion of the players in the game's milieu and the social importance of the prizes on offer. This may be particularly prominent in multiplayer games, and especially attractive to younger consumers and those lacking in self-esteem. It is not possible fully to justice to these items in this piece and it is an area where further research would be valuable.

(e) Individualised targeting and aggression

The assumption so far is that the trader's conduct should typically be judged by the average targeted or vulnerable standard because they will generally be aiming their practice at consumers who are likely to be vulnerable, or because they should realise that there is a group of consumers who are particularly likely to be vulnerable to that practice. There is, however, an additional and important element that deserves further investigation: that of individualised targeting.Footnote ¹²⁴ It is beyond the scope of this paper to tackle the subject in the detail it deserves, but it is important to identify it as an area that demands further examination.Footnote ¹²⁵

Individualised targeting will take place when the trader is able to ascertain details of the behaviour of a consumer in game and amend their offer accordingly. It is particularly prevalent where online gaming (especially ‘games as a service’) is concerned, with consumers ‘consenting’ to the collection of their data as a condition of playing the game.Footnote ¹²⁶

Where the trader is collecting data about a consumer and making offers on the basis of conclusions its algorithm draws about an individual consumer, there is a particularly strong argument that the trader is in a position power vis-à-vis the consumer, forming the basis for undue influence. As has been discussed, that position must be exploited ‘in a way which significantly limits the consumer's ability to make an informed decision’. The full extent to which traders are targeting consumers at an individual level remains unclear. However, it is known that patents have been awarded for algorithms which allow for this, for example adjusting the price of virtual items for individual users.Footnote ¹²⁷ This raises similar concerns to those raised by online behavioural advertising and personalised pricing. But it also reveals the extent to which consumers could be manipulated, particularly when they feel under great pressure to succeed for some of the reasons discussed above. As King et al argue:

In this way, the practice is not only unfair vis-à-vis the individual consumer using the product, but gathers data that could help to construct the ability to behave in a way that allows the manipulation of a broader range of consumers with similar characteristics and behaviours.

Further research is needed on the extent to which such microtargeting is being used, both in relation to loot boxes and in the context of other gaming. However it is clear already that the technology exists to allow for the manipulation of vulnerable consumers on the basis of their in-game behaviour.