61.In the past, the games industry was built on a ‘premium’ model: people paid upfront for a hard copy of a game and that was the only opportunity for monetisation. However, the growth of internet-connected games has enabled the industry to develop new ways of making money, including through ‘microtransactions’—small payments that players make throughout the process of playing a game, for example to acquire in-game skills or items or to progress more quickly through levels. We shall explore the contribution of these business models to the industry in Chapter 4, after detailing the effect that this monetisation strategy can have for the minority of players who might struggle to maintain control over how much they are spending on gaming.
62.We were contacted by a member of the public whose adult son built up considerable debts, reported to be in excess of £50,000, through spending on microtransactions in British company Jagex’s online game RuneScape. For example, bank statements showed that in one day the individual spent £247.95 by making five separate payments to the company. The resulting debt caused significant financial harm for both the player and his parents, whose evidence attributed the situation to the fact that Jagex has no limits on the amount of time or money players can spend on the game. This demonstrated to us that even companies with good policies to support some aspects of player wellbeing can fall short in other areas.
63.Jagex told us that it generates about one-third of its revenue from microtransactions, with two-thirds coming from an alternative subscription model. The company’s director of player experience Kelvin Plomer told us that players “can potentially spend up to £1,000 a week or £5,000 a month” in RuneScape, but that only one player had hit that limit in the previous 12 months. The company’s reasoning for setting this limit seemed to stem from fraud prevention, rather than out of a duty of care to prevent people spending more than they are able. Jagex does allow players to “request deletion of the account or suspension of the account or a payment block”; however, crucially in the case of the parent who contacted us, for data protection reasons it can deal only with account holders and so was unable to take direct action in response to the parent’s concerns.
64.Other players have raised similar concerns about in-game spending in RuneScape, and have referenced the way the game is designed to encourage players to spend more. For example, one told us that in his five years playing RuneScape:
I’ve spent over $2000 CAD on in-game wealth and in-game progression. I am not proud of this. I’ve loaded my credit card with $500 CAD in one night to skip hundreds of hours of content I didn’t find enjoyable to unlock achievements because of a promotion they were running. I’ve spent $300 CAD in 20 minutes legally on in-game wealth I lost instantly gambling because I wanted more in-game wealth to purchase better gear.
65.With so many games featuring microtransactions, we asked other leading games companies what they do to prevent people from spending more than they are able to in a game. Epic Games, the makers of Fortnite: Battle Royale, said that although it has no specific cap on how much people can spend, purchases are limited to what is in the virtual shop and so the maximum anyone could spend would be about $200 a day. The company’s director of marketing suggested that it is an individuals’ responsibility to manage spending and that the company provides tools, including parental controls, “to allow players to make those decisions for themselves.” However, as we have already examined, parental controls are of limited effectiveness when it comes to adult players.
66.The games companies we spoke to were generally reluctant to accept that they might have a role or responsibility to intervene proactively if a player’s spending fell outside of normal patterns. Moreover, they said that it would be too difficult to determine what level of spending might be harmful. Alex Dale, senior vice president for King, the makers of Candy Crush Saga, told us that while it used to alert users when they reached a certain spending threshold, it had stopped doing so because of player feedback. He told us:
we would send an e-mail out when a player’s spend was $250 in a week for the first time. It was an e-mail that said, “We notice you are enjoying the game a lot at the moment. Are you sure you are happy with this?” […] We got back, “I wouldn’t spend the money if I didn’t have it” and things like, “I’m fine, please leave me alone”. We felt it was too intrusive so we stopped doing that.
67.Dr Henrietta Bowden-Jones told us that the games industry might learn from approaches in the gambling industry, which has much clearer, industry-wide protocols enabling people to self-exclude from spending. She told us that “gaming is several years behind gambling in relation to protecting the vulnerable”, a situation that the then Minister described, if true, as “lamentable”, adding that “the industry has a job to do.” When we asked Neil McArthur, Chief Executive of the Gambling Commission, whether games companies should be obligated to monitor how much people spend, identify people with a problem and proactively support them, he responded that, “this is an area where progress needs to be made.” His colleague Brad Enright went on to tell us that self-exclusion measures to protect players “could be adopted by the video games industry voluntarily to address some of the concerns about excessive time [and] excessive expenditure.” Yet the Royal College of Psychiatrists goes further in suggesting that:
There should be no in game spending by children. Children are less prepared to deal with the potentially addictive nature of some modern computer games and are less able to make informed decisions about spending.
68.In considering the way that some games facilitate gambling-like behaviours among players, it is important to acknowledge the distinction between licensed online gambling, social casino-style games that “have the look and feel of traditional gambling” but may not be licensed as such, and games containing features akin to gambling as one aspect of the overall product or game experience rather than the predominant quality. Our inquiry has focused on the latter, although the other two are both important issues that merit further consideration.
69.Many games contain features that are highly similar to conventional gambling products, without gambling being the primary aim of the game. However, there are concerns that being exposed to such features from a young age might normalise gambling. One parent expressed concern that the game Bricky Farm, which is rated suitable for children, contains a gambling-like feature. He told us:
Most worrying for me is a roulette style wheel mini-game whereby differing amounts of gems can be won for further advancement. This is where the game could become addictive to someone with a susceptibility but more than that it is introducing children as young as 4 to the ‘thrill’ of gambling.
70.The parent’s concern is supported by Dr David Zendle’s acknowledgment “that a really good predictor of problem gambling is the social acceptance and availability” of it. Indeed, the Gambling Commission told us that the Advisory Board for Safer Gambling expressed concerns in response to the Online Harms White Paper about the associations between “gambling lite behaviours and children’s behaviours”. Furthermore, Brad Enright from the Gambling Commission told us that even when games do not meet the regulatory threshold for gambling, but contain gambling-like features, the regulator does:
not think the current age ratings are in line with public expectation, so that should not be available for four-plus or even 12-plus.
71.However, others have argued that there is not yet enough evidence on the psychosocial effects of gambling-like mechanics, particularly on children. We have been told by academics that:
This is largely because academic scientists do not have access to the proprietary industry data needed to provide an answer to this question. Scientific research needs to study which gaming mechanisms are the most problematic and needing of both monitoring and regulation.
Dr Mark Griffiths from Nottingham Trent University’s International Gaming Research Unit recommended in written evidence that:
A scientific working group should be set up under the DCMS to collate the latest evidence relating to the effects of gambling-like gaming. This could then inform an evidence-based paper on gambling-like gaming in order to: (i) provide clarity regarding the evidence and the recommendations, (ii) be shared as a common guideline and practice by relevant UK organizations (i.e. UK Gambling Commission, UK Council for Child Internet Safety, Parent Zone, Childnet) that deal with support and advice provision towards parents and the community.
72.We believe that any gambling-related harms associated with gaming should be recognised under the online harms framework. To inform this work, the Department for Digital, Culture, Media and Sport should immediately establish a scientific working group to collate the latest evidence relating to the effects of gambling-like mechanics in games. The group should produce an evidence-based review of the effects of gambling-like game mechanics, including loot boxes and other emerging trends, to provide clarity and advice. This should be done within a timescale that enables it to inform the Government’s forthcoming online harms legislation.
73.One of the most prominent features in the debate about the potential links between game mechanics and gambling is loot boxes. Loot boxes are “items in video games that may be bought for real-world money, but which provide players with a randomised reward of uncertain value.” Those rewards will be virtual items for use in the game, such as tools, outfits and weapons, or characters with particular skills, all of which will be of variable benefit within the game. They are a common form of microtransaction, with a 2018 Gambling Commission survey finding that 31% of 11–16 year olds have paid money or used in-game items to open loot boxes. Although some games (including, notably, a version of Fortnite) reveal the contents of a loot box to the player before they decide whether to pay for it, usually the contents of loot boxes are unknown to the player at the point of purchase—what a player gets for their money is therefore based on chance.
74.At the 2018 ‘4C International Game Developers Conference’ in Prague, Ben Lewis-Evans, a user experience researcher at Epic Games gave a presentation called ‘Reward Psychology—Throwing out the Neurotrash.’ In this he set out some of the reward mechanics behind devices like loot boxes in games such as Fortnite. He said that, “The reason these are famous is they are quite often the ratios used in gambling. The response they tend to get is a very constant and high level response and that’s because you don’t know how many times you have to respond before you get the reward, so people tend to keep it up.” He also added in his presentation that, “We do have to be careful about controlling people, but we do need to make sure that they feel in control”.
75.Much of the evidence that we received from gamers critiqued the loot box mechanics in Electronic Arts’s FIFA series. In the game’s ‘Ultimate Team’ mode, players build their own football team by buying virtual ‘packs’ containing a randomised selection of footballers. To purchase these packs, players use ‘coins’ that are earned in the game and ‘points’ purchased for real-world money. In 2016, EA’s CFO Blake Jorgensen told investors that the Ultimate Team mode accounted for roughly half of the $1.3 billion the company makes from extra digital content—that is equivalent to $650 million in annual revenue, in addition to that generated by units sales of the full games.
76.Electronic Arts releases a new game in the FIFA series every year. However, with each new release, players’ teams are not transferred over and therefore they must rebuild their teams by purchasing more packs to acquire the best players. One gamer told us that this cycle resulted in them spending “almost £800 to £1000 a year annually on FIFA”. Another gamer told us that because a pack’s contents “directly affects gameplay because some players are not as good as others”, it incentivises people to keep buying packs in the hope of getting better players and, therefore, performing better in the game. They told us that:
in order to compete, players feel like they need to buy hundreds, if not thousands, of £s worth of packs in order to get the best players. Children are especially vulnerable because they lack the maturity to understand that these purchases are manipulative, and their parents may not understand that these purchases are entirely unnecessary.
77.Moreover, because the probability of securing some footballers in a pack is less than 1%, some gamers have expressed frustration at spending vast amounts without receiving the desired reward. To tackle the issue of speculative spending on loot boxes, one player recommended that:
the sale of items in game, cosmetic or otherwise, should only be allowed via direct purchase or via achievements unlocked through in game actions. In a loot box type system the player, typically, has a very small chance of unlocking the item they actually want. This encourages further spending on loot boxes in order to get the item they want.
78.We put some of these concerns to Kerry Hopkins from Electronic Arts, who responded that the way they have implemented this mechanic in FIFA “is quite ethical and quite fun”. Yet this is noticeably out of step with the attitude of many of the gamers who contacted us following our evidence session, including those who vehemently rejected her characterisation of packs not as loot boxes but as “surprise mechanics”. One gamer called the company’s testimony to us “a bare face lie”, and another told us that the company has:
heavily marketed and referred to their systems as ‘loot boxes’ for several years and […] the mechanics of the system are exactly the same no matter what they choose to call it.
79.We recommend that loot boxes that contain the element of chance should not be sold to children playing games, and instead in-game credits should be earned through rewards won through playing the games. In the absence of research which proves that no harm is being done by exposing children to gambling through the purchasing of loot boxes then we believe the precautionary principle should apply and they are not permitted in games played by children until the evidence proves otherwise.
80.We have heard concerns about the “structural and psychological similarities” between loot boxes and gambling. Dr Aaron Drummond and Dr James Sauer told us in written evidence that the random delivery of loot box rewards is akin to conventional gambling products and:
designed to exploit potent psychological mechanisms associated with the development and maintenance of gambling-like behaviours.
81.Dr Drummond and Dr Sauer argue that “it is plausible that engaging with these loot box systems could have short-term consequences (e.g., over-spending on accessing loot box systems) and longer-term consequences (e.g., facilitating migration to more conventional forms of gambling)”. However, academics broadly acknowledge that there is not yet enough evidence to reliably conclude that loot boxes cause problem gambling. This was echoed by the then Minister’s observation to us that:
If evidence does emerge that loot boxes can be a gateway to problem gambling, then we need to take that seriously and we need to take some action. But the evidence is not there yet. There are not many studies.
Yet, even if there is not enough evidence at this stage to establish a causal link between loot boxes and problem gambling, other research suggests that they may still be causing harm.
82.A study by Dr David Zendle and Dr Paul Cairns identified a link between the amount that gamers spend on loot boxes and their score on the Problem Gambling Severity Index (PGSI). Moreover, the large-scale study of more than 7,000 gamers suggested “that the gambling-like features of loot boxes are specifically responsible for the observed relationship between problem gambling and spending on loot boxes” as other forms of microtransaction did not display such a strong link. A further study found the same link among adolescents—in fact, the link between loot box spending and problem gambling among adolescents was more than twice as strong as the relationship observed in adults.
83.Dr Zendle told us in oral evidence that although his studies have not identified a causal link between loot boxes and problem gambling:
Something very different might be happening here where people who are already problem gamblers, people who already have a disordered and excessive relationship with gambling-related activities that may to some extent be beyond their control, are now going into their favourite games and saying, “Oh look, it is something that looks an awful lot like this thing I have a disordered and excessive relationship with”. That is why they are spending more money on loot boxes. It is not that it is a gateway; it is that it is a way that video games companies may, accidentally or incidentally, be profiting from problem gambling among their consumers.
84.Dr Zendle and Dr Cairns, among others, therefore make the case for enhanced regulation of loot boxes, such as ensuring games containing loot boxes carry parental advisories or descriptors outlining that they feature gambling content, and propose that “serious consideration is given to restricting games with loot boxes to players of legal gambling age”. Brad Enright told us that these “seemed like very sensible recommendations” which “the video games industry should probably take stock of.”
85.Games regulated under the PEGI system can feature content labels alerting users to the fact they contain in-game purchases. However, there is no specific content descriptor for loot boxes, despite Dr Zendle telling us that “they are formally very different to other microtransactions”. Moreover, PEGI has a content label alerting users to the fact a game contains actual, or simulated, gambling; however, again, this does not apply to games with loot boxes. The reason for this has been stated by a PEGI representative who said:
The main reason for this is that we cannot define what constitutes gambling. That is the responsibility of a national gambling commission. […] If a gambling commission would state that loot boxes are a form of gambling, then we would have to adjust our criteria to that.
Yet Dr Mark Griffiths argues that PEGI’s position appears to be:
somewhat hard line given that PEGI’s descriptor of gambling content is used whenever any videogame “teaches or encourages” gambling. Such a descriptor would arguably cover gambling-like games or activities and the buying of loot boxes is ‘gambling-like’ at the very least.
86.Loot box mechanics are integral to major games companies’ revenues and evidence that they facilitate profiting from problem gamblers should be of serious concern to the industry. We recommend that working through the PEGI Council and all other relevant channels, the UK Government advises PEGI to apply the existing ‘gambling’ content labelling, and corresponding age limits, to games containing loot boxes that can be purchased for real-world money and do not reveal their contents before purchase.
87.At present, the Gambling Commission states that purchasing loot boxes does not meet the regulatory definition of licensable gambling under the Gambling Act 2005 because the in-game items have no real-world monetary value outside the games. That position is despite the fact the regulator acknowledges that:
The payment of a stake (key) for the opportunity to win a prize (in-game items) determined (or presented as determined) at random bears a close resemblance to the playing of a game of chance. The playing of a game of chance for a prize of money/money’s worth is gambling under UK law.
88.We have been told that the regulator’s position centres on the definition of ‘money’s worth’. Neil McArthur told us:
There is quite a lot of case law around what money or money’s worth means, the difference between how you would assess value, the fact that value needs to be objectively assessed rather than in the eye of the beholder, so it is quite a complicated area.
However, this is arguably out-of-step with the digital economies in the games industry. Indeed, Brad Enright acknowledged that the definition of ‘money’s worth’ is an area where the Gambling Commission is “constrained […] by the current legislation.” He told us that that a broader definition of value:
would need a change in legislation, but also in doing that you would have to be very careful not to catch lots of other activities, which may also incorporate expenditure, chance and a prize of value or worth to the player.
89.The existing legislative framework has been critiqued by Dr Aaron Drummond and Dr James Sauer, who argue that real-world monetary value is too narrow a definition of value when it comes to loot boxes:
It ignores the subjective value created for players from the combination of scarcity of, and competitive advantage provided by, in-game items in the gaming environment. These in-game rewards can have value for players—and influence players’ behaviour (i.e., motivate them to engage with loot box mechanisms)—without being converted into real currency. Perhaps more importantly, it ignores the fact that players are demonstrably willing to pay real money for the chance to acquire these items; implying they have a monetary value even in the absence of the ability to convert them back in to currency.
90.It is also widely acknowledged that the virtual contents of many loot boxes can be ‘cashed out’ for real-world monetary value—a process known as ‘real-world trading’, which brings them closer in line with the legal definition of gambling. The games industry has taken a hard line against the practice, which is usually prohibited within a game’s terms and conditions. Kerry Hopkins told us that Electronic Arts has “24/7 security staff monitoring activity to be sure that people are not engaging in this type of behaviour”, and yet still bans thousands of people a year for violating its rules prohibiting people from selling in-game items outside the game.
91.The prevalence of real-world trading suggests the industry does not have as secure a hold on it as it might argue, and Dr Mark Griffiths told us that the Gambling Commission’s position “appears to be a case of the law struggling to keep pace with technology.” Indeed, the Gambling Commission’s written evidence states that the problem is more widespread than the industry acknowledges:
Based on open source research, the volume, variety and sophistication of websites advertising opportunities to exchange in-game items for cash, indicates that to term such circumvention of regulation as ‘occasional’ understates the extent of this issue for certain games.
It goes on to conclude:
In particular, Brad Enright cited US-company Valve’s Counter-Strike: Global Offensive as an example of a company failing to proactively enforce real-world trading. He also told us that research here and in North America suggests “that understanding of gambling law is pretty poor in the video games industry” and called for the industry to do more to tackle the problem, stating:
We think that they have the creativity and the talent to come up with solutions that would make that far more proactive, so that individual gambling commissions around the world are not dealing with this issue one by one.
92.Other countries have taken different regulatory approaches to loot boxes, which we believe the UK could learn from. For example, Belgium’s Gaming Commission concluded that loot boxes constituted “games of chance” subject to the country’s gambling laws. Brad Enright told us that this was because “Belgian gambling legislation has a broader definition of ‘prize’”, which “has allowed the authorities to threaten legal action against some of the larger operators”. The Netherlands Gambling Authority also recently concluded that some loot boxes it studied—including in FIFA 18—contravened the country’s Betting and Gaming Act. We were told by the Gambling Commission that the Netherlands’s “view is pretty much the same as ours”, as the contravention arises from the fact players can exchange in-game items obtained through loot boxes for real-world money.
93.The experiences of other regulators also indicates what could happen if loot boxes were determined to be gambling under UK law. We heard that if legislation was changed, any games company using loot boxes would need to obtain an operating licence, which are accompanied by a range of regulations around transparency, duty of care and age restrictions. Yet, the way in which games companies have withdrawn mechanics in response to the ruling in Belgium suggests that:
In reality, it would withdraw all the mechanics. It presumably would not want to be subject to all the rules and regulations.
94.In China, games that feature loot boxes are legally required to state the odds of obtaining each item. However, Brad Enright was clear that even if companies voluntarily introduce transparency in this way, if loot boxes were still legally defined as gambling, “that is not a substitute for a licence” or all the conditions they impose.
95.There are other ways in which games have become linked to betting. For example, ‘skin betting’ or ‘skin gambling’ is the use of virtual items acquired in a game as a method of payment for a stake in external, unlicensed gambling. A recent Gambling Commission survey found that 3% of 11–16 year olds had bet with in-game items on websites outside of video games or privately. Since 2015, the Gambling Commission has investigated a number of websites that provide facilities for this form of unlicensed gambling, including to children. That included bringing a criminal prosecution against the operators and advertisers of a website called FutGalaxy.
96.The Gambling Commission argues that the games industry has a role to play in preventing such third-party activities. Its written evidence states:
The significant risk of harm posed by these unregulated gambling websites, whilst unintended, is nonetheless a by-product of the manner in which games have been developed and in-game economies incorporated for commercial benefit. Despite there being no evidence of any direct commercial relationship between games publishers and the illegal gambling facilities, it is reasonable to infer that there is an indirect benefit derived from these activities given they drive engagement with the game and it is the games publishers who are the ultimate source of in-game items acting as a de-facto central bank.
In oral evidence, Neil McArthur told us that games platforms and manufacturers need to be proactive and “make it as difficult as possible for people to create parasitic offerings based on their game platform that encourage people into gambling.”
97.We agree with the Gambling Commission that games companies should be doing more to prevent in-game items from being traded for real-world money, or being used in unlicensed gambling. These uses are a direct result of how games are designed and monetised, and their prevalence of undermines the argument that loot boxes are not a form of gambling. Moreover, we believe that the existing concept of ‘money’s worth’ in the context of gambling legislation does not adequately reflect people’s real-world experiences of spending in games.
98.We consider loot boxes that can be bought with real-world money and do not reveal their contents in advance to be games of chance played for money’s worth. The Government should bring forward regulations under section 6 of the Gambling Act 2005 in the next parliamentary session to specify that loot boxes are a game of chance. If it determines not to regulate loot boxes under the Act at this time, the Government should produce a paper clearly stating the reasons why it does not consider loot boxes paid for with real-world currency to be a game of chance played for money’s worth.
118 Anonymous ()
120 Qq440, 445
123 Ty Beauchamp ()
124 Q1029 ff.
125 Q1020 ff.
131 Royal College of Psychiatrists ()
132 Gambling Commission, “”, (March 2017), p 13
133 Scott Rooney ()
137 Professor Andrew Przybylski, Netta Weinstein, Pete Etchells and Amy Orben ()
138 International Gaming Research Unit, Nottingham Trent University ()
139 David Zendle, Rachel Meyer, Stuart Waters, Paul Cairns, “”, PsyArXiv. (19 May 2019), Web
140 Gambling Commission ()
141 We note that loot boxes vary considerably across games, for example some are not purchased for money but obtained through game-play. For the purposes of this discussion, however, we are referring to loot boxes that can be brought with real-world money and do not make the contents clear to the player before purchase.
142 “,” YouTube, published 13 February 2018, accessed 2 September 2019
143 “”, Gamesindustry.biz, 2 March 2016
144 Lathan Boyle ()
145 Cameron Graham ()
146 James Handford ()
147 Dave Gannon ()
150 Alexis Somerville (), Andrew Stormont ()
151 Dr Aaron Drummond and Dr James D. Sauer ()
152 Dr Aaron Drummond and Dr James D. Sauer ()
153 Dr Aaron Drummond and Dr James D. Sauer ()
156 David Zendle and Paul Cairns, , PLoS ONE, vol. 13.11, (21 November 2018)
157 Dr David Zendle and Dr Paul Cairns ()
159 Dr David Zendle and Dr Paul Cairns ()
162 “”, WCCFTech, 12 October 2017
163 International Gaming Research Unit, Nottingham Trent University ()
164 Gambling Commission, , (24 November 2017), accessed 18 February 2019
165 Gambling Commission ()
168 Dr Aaron Drummond and Dr James D Sauer ()
169 Gambling Commission ()
171 International Gaming Research Unit, Nottingham Trent University ()
172 Gambling Commission ()
173 Gambling Commission ()
175 Qq1557, 1558
178 Q1578 ff.
181 ‘Skins’ is a term for cosmetic enhancements, such as outfits for a player’s character to wear, in a game; however, a range of virtual items, including virtual currencies, may be used for unlicensed gambling in this way.
182 Gambling Commission ()
183 Gambling Commission ()
Published: 12 September 2019