Loot boxes are items in video games that can be paid for with real-world money but contain randomized contents. Many games that feature loot boxes are played by adolescents. Similarities between loot boxes and gambling have led to concern that they are linked to the development of problem gambling in adolescents. Previous research has shown links between loot boxes and problem gambling in adult populations. However, thus far, there is no empirical evidence of either the size or existence of a link between loot box spending and problem gambling in adolescents. A large-scale survey of 16- to 18-year-olds (n = 1155) found evidence for such a link (η2 = 0.120). The link between loot box spending and problem gambling among these older adolescents was of moderate to large magnitude. It was stronger than relationships previously observed in adults. Qualitative analysis of text data showed that gamers bought loot boxes for a variety of reasons. Several of these motivations were similar to common reasons for engaging in gambling. Overall, these results suggest that loot boxes either cause problem gambling among older adolescents, allow game companies to profit from adolescents with gambling problems for massive monetary rewards, or both of the above. Possible strategies for regulation and restriction are given.
1.1. What are loot boxes?
Loot boxes are a relatively new way for players to spend money in video games. In recent years, loot boxes have mushroomed from a relatively obscure and unknown in-game mechanism to an industry that is predicted to generate up to $30 billion in 2018 alone . There is concern among both regulators and researchers that spending money on loot boxes may be linked to gambling-related harm among both adults and children.
Making ‘microtransactions’ of small amounts of real-world money for virtual items or other advantages has been common in video games for many years. For example, players of the 2011 action game Dynasty Warriors 7 can pay $0.99 to $1.99 to unlock exclusive in-game weapons; the role-playing game Tales of Xillia lets players pay $3 to buy ‘Level Up’ packs that make their characters stronger; players of the city-planning game SimCity Buildit can pay small amounts of money to increase the efficiency of their construction efforts. Even over a decade ago in 2005, players of the mobile game Puzzle Pirates could pay real-world money to buy ‘doubloons’, an in-game currency that could be spent on virtual items and services . Similarly, in 2006, players of the open-world game Oblivion were able to make a microtransaction to buy cosmetic armour for their in-game horses .
However, in recent years, a new kind of microtransaction has become increasingly prevalent in video games: the loot box. In the examples given above, players who had paid real-world money all knew what they would get in return for this expenditure: they were buying additional levels, or doubloons, or weapons, or costumes or horse armour. By contrast, when players buy a loot box, they are not paying for something specific—they are, instead, paying for something that appears to be randomly selected from a list.
For example, in the popular first-person shooter Counter-Strike: Global Offensive, players can pay $2.49 to open a sealed ‘weapon case’. Cases may contain extremely rare and valuable ‘skins’ that change the appearance of players' weapons. In fact, some of these skins are so prestigious and uncommon that they can be re-sold on secondary markets for many thousands of dollars . However, when paying to open a loot box, players of Counter-Strike also run the risk that the case they have paid to open contains an unappealing or common item, rather than a rare or desirable one. There is no way for them to tell what they will get when they pay their money. Similarly, players of the football game FIFA 19 pay real-world money to purchase ‘player packs’ that contain new footballers for their teams. These packs may contain rare and valuable players that improve their team's performance or they may not. In either case, players do not know what a pack contains when they pay real-world money for it.
1.2. Loot boxes and problem gambling
The behaviour outlined above involves staking real-world money on the chance outcome of a future event. Several international regulatory authorities have noted that there are striking similarities between this behaviour and gambling. This has led to various investigations across the globe of whether loot boxes in video games are in contravention of existing gambling legislation and therefore constitute an illegal and unlicensed form of gambling [5–7].
Connected to these arguments about the legal definition of loot boxes are questions about harm, and more specifically, harm to children. As the UK Gambling Commission themselves note in :
many parents are not interested in whether an activity meets a legal definition of ‘gambling’. Their main concern is whether there is a product out there that could present a risk to their children.
The specific form of gambling-related harm that is most commonly associated with loot boxes is problem gambling. Problem gambling refers to disordered and excessive gambling activities that are so extreme that they lead to significant problems in an individual's personal, family and professional lives. Problem gambling is linked to factors such as depression, anxiety, bankruptcy and suicide [9–11]. One key pathway to the development of problem gambling is via conditioning: the more individuals are exposed to the arousal associated with gambling activities, the more they come to expect and require this excitement, leading to the disordered and excessive patterns of gambling-related spending mentioned above . This pathway to problem gambling is considered to be a particular risk among adolescents. Indeed, exposure to gambling activities in childhood is an important predictor of problem gambling among adults .
There are good theoretical reasons to believe that loot boxes might be ‘psychologically akin’ to gambling themselves, and exposure to them might therefore cause problem gambling among children. In , Griffiths specifies five characteristics that differentiate gambling from other risk-related behaviours. These are:
In , Drummond & Sauer undertook a systematic analysis of 22 video games that feature loot boxes to determine the extent to which they fulfilled these characteristics. They found that 10 of the 22 video game loot boxes fulfilled all of the criteria listed above, and many more fulfilled most of them. They found that loot boxes not only shared ‘important structural and psychological similarities with gambling’, but that ‘100% allow for (if not actively encourage) underage players to engage with these systems'. Indeed, Drummond and Sauer concluded that the presence of loot boxes in video games might therefore be forming a ‘ripe breeding ground’ for the development of problem gambling among children.
Empirical research supports the existence of this effect in adult populations. In , Zendle & Cairns conducted a large-scale study on gamers aged 18 and over, and measured both their loot box spending and their problem gambling. Results of this study indicated that the more gamers spent on loot boxes, the more severe their problem gambling was. The effect size associated with this relationship was of magnitude η2 = 0.054—stronger than links between problem gambling and common risk factors in the gambling literature such as drug abuse.
However, while links between problem gambling and loot box spending seem robust among adults, no research has yet either examined whether loot box spending is linked to problem gambling among adolescents. More importantly, no research has yet examined what the magnitude of this relationship might be. There are good reasons why loot box spending might cause problem gambling among children—and there are also good reasons why the relationship between loot box spending and problem gambling may be stronger in children than it is in the adult population.
Adolescents as a group seem particularly susceptible to problem gambling [19,20]. Indeed, problem gambling is often estimated to be more prevalent among adolescents than it is in adult populations [21–23]. There are several explanations for why adolescents might be more likely to develop problem gambling than adults. For example, neurodevelopmental research suggests that the immaturity of various aspects of brain structure and function are linked to increased impulsivity among adolescents—and this may lead to increased vulnerability to problem gambling [24,25]. Similarly, research into coping strategies among adolescents suggests that this group may lack effective ways to cope with the ‘turbulent times’ that are associated with their time of life. They may therefore turn to gambling activities as a way to escape from painful states, leading to the development of problem gambling . When taken together, these results suggest that pathways to problem gambling via loot box spending would be particularly pernicious among adolescent populations.
1.3. Features of loot boxes
There are a broad variety of different video games on the market. There are a similarly broad variety of different ways that loot boxes are implemented in these games. There are concerns among both academics and regulators that some kinds of loot boxes may be more strongly linked to the development of problem gambling than others [15,28]. Several key differences between kinds of loot boxes are listed below.
1.3.1. Some loot box contents can be cashed out
