Regulus Partners sets the record straight on how betting incumbents should represent black market threats, where the emphasis should be placed on understanding player habits and protecting integrity over market size impacts.
Whenever a regulator or government seeks to make ‘illiberal’ changes to gambling policy, the dangers of a black market will almost inevitably be wheeled out in evidence against. These dangers are routinely exaggerated, misrepresented or misunderstood by many stakeholders who should know better… from all sides of the debate.
Black market activity is opaque and its size anywhere is almost impossible to determine. However, three things are true: black markets do exist; domestic regulations are the biggest single lever in determining their size, and; a relatively small number of engaged players can create a meaningful black market. However, by examining the underlying drivers of black markets we make the case that focus should be on understanding player behaviour, not second-guessing size: policy decisions which encourage black market engagement can be clearly identified and they are nearly always counter-productive.
Along with almost everything else in gambling, ‘black market’ risk tends to be politicised and weaponised rather than properly understood. It is easy to argue that it is, or could be, “a very big thing” and therefore is a failure of policy. Given the systemic opacity of ‘black markets’, however, it is equally easy to argue that the scale, the risk or both can be overstated. We would suggest that both of these positions are more designed to win arguments than help policy. Elements of the gambling industry have undoubtedly ‘cried wolf’, over-egged the risks and misrepresented the causes of black markets since the dawn of regulated gambling. The advent of Point of Consumption online markets has caused many to double down on this practice. However, bad lobbying does not mean the problems should be dismissed while assuming a small black market is less of a problem than a big black market is perhaps the most dangerous policy misdirection of all.
In this blog, we examine the underlying causes of black markets, what they mean for consumers, what they mean for commercial dynamics and what the policy implications are. There are essentially three drivers of black markets once a domestically regulated regime is in place and one red herring:
- Product restrictions (eg, banning certain products or product categories)
- Value distortions (eg, taxing customers or causing prices to be adjusted)
- User experience restrictions (eg, getting customers to do things they don’t want to do)
- Visibility restrictions (eg, banning advertising), which impact share much more than shape
It is important to stress that black market risk cannot and does not mean that restrictive regulations are ‘bad’ or ‘unworkable’; only that the black market flip side needs to be factored in from an efficacy standpoint. Critically, it is not just a question of what you channel, but who you channel.
By far the most obvious way to create a black market is to ban certain products, especially if they are popular. While there are some regional and cultural variations to product consumption, online gamblers are surprisingly consistent across markets (with a few inevitable outliers):
- Horseracingtends to be 5-20% of revenue where there is a domestic market and no structural protections; racing is probably the most variable given huge differences in the presence and/or quality of domestic supply so the figures below exclude horseracing
- Sportsbettingtends to be c. 35-45% of revenue, with in-play representing c. 70% of turnover and c. 50% of net revenue
- Slots represent c. 30-40% of revenue, with higher value slots (€2+ per spin) representing c. 10% of sessions but c. 50% of slots value
- Poker represents 2-10% of revenue (without distortions)
- Other gaming represents 10-30%, including live, table games, bingo etc
When a product is banned, it can have a clear substitution benefit to other products. In France, for example, where online casino gaming is banned (slots, other gaming), poker is 16% of revenue and per capita, product expenditure is 50% greater than the European average. However, France’s total online spend per capita is only half the European average, putting it with Southern and Eastern European countries despite broader ecommerce adoption that is c. 20ppts higher. One of three things are going on here:
- The French don’t like online casino much: we know this isn’t true because it was a material .com market before regulation shut it off (eg, France was the top market for the .com assets bought by WH in 2008)
- French consumers have decided to gamble far less online than European peers
- A material proportion of French online gambling is not captured by domestic regulation
Point two may have some validity true because of restrictions, but France’s regulator recognises that this is only partially true (see ARJEL 2018-19 Activity Report, which estimates 300-500k black market online casino consumers but then makes the mistake of assuming engaged casino players who seek out black market sites spend the same per capita as domestically captured players, many of whom are mass-market sports betters: 85% of whom spend less than €500 pa). We can be pretty confident that it is mostly point three, therefore.
Our own estimates put the gaming only element for the French black market to be c. €1.1bn; ie, the poker market has captured only c. €60m of gaming demand. This would put online casino spend per capita at €2,200 pa, compared to a UK figure of c. 3m people spending €1,300 pa: in France the ‘long tail’ of customers is missing or substituted (critically including the cohort actively cross-sold from legal betting) but the engaged core is intact but not captured. To put this spend per capita into French domestically regulated sports and racing context, the same report reveals that c. 5% of players or c. 140k people spend more than €2,000 pa, so our €1.1bn casino figure triangulates to wider patterns of play. Big distortions can therefore have big black market revenue impacts.
France has an active anti-black market policy of prosecuting businesses targeting French consumers, ISP blocking and payments blocking. This is no doubt keeping the mass market out of online casinos. However, the mass market rarely engages with online casino anyway: betting is the only clearly mass-market product (eg, in France there are 3.5m online sports betting accounts vs. 1.2m for poker; in Britain, 3x as many people engage with online betting vs. online gaming). Even in more liberal markets, casino customers tend to be relatively few in number but enjoy what they do. We also know (a corollary of greater customer concentration) that ‘problem gambling’ prevalence is higher among those who participate in online gaming compared with online sports betting. Hence our confidence in a higher black market figure from a similar number of actives and the policy concern that product distortions are almost certainly failing the most vulnerable.
In summary, by banning an entire product category France has removed protections for a relatively small number of people (c. 500,000 or c. 15% of unique online gamblers in France). However, these people are also the highest spending customers in France, who are not being effectively captured by the regime. Many will be doing so safely, but not all, and none with any domestic protections. In terms of customer protection (setting aside landbased protection and tax), France is not achieving the goal of limiting exposure to online gambling, although it is arguably achieving its goal of limiting mass-market exposure to online gaming. This is at the expense of customers who want to use products that normally represent c. 50% of a functioning market, however. It is hard to judge this as an effective policy outcome by any measure.
France is an ‘easy’ example because of its stringent restrictions. However, it should be noted that high-value slots typically represents c. 15-20% of revenue and in-play represents c. 15-25%. Major restrictions to these products can affect a lot of revenue and cause people to seek supply elsewhere. Because these people are few in number demonstrates (relative to revenue generated) that they are the most engaged. A product restriction policy that pushes the most engaged customers away from domestically regulated markets is structurally bound to have weak customer protection outcomes.
The four most obvious fiscal-regulatory interventions that distort value are:
- Taxes on stakes (which as soon as they become material vs. payout have to be reflected in pricing)
- Taxes or significant regulatory limitations on promotions (which limit the incentives which can be offered vs. black market)
- Taxes on customer winnings
- Setting payout ratios
Relatively few customers are genuinely and actively ‘price-sensitive’, whether this is to the odds of a bet or the level of promotions offered. However, substantially all customers are ‘spend sensitive’, either consciously or otherwise and those that are not spend sensitive are by default less likely to be moderate in their gambling. The relatively few customers who are ‘price-sensitive’ (maybe c. 5% of an active customer base) are engaged and savvy players: they will have the wherewithal to hunt for the best value, even in the black market. Any regulatory intervention that means black market providers can offer better odds or significantly better promotions will mean that this customer cohort becomes a ‘flight risk’. Again, this is a relatively small number of customers which many operators would rather do without, but a significant proportion of ‘activity’ (probably representing c. 20% of turnover, c. 10-15% revenue). These customers may be savvy, but regular engagement with gambling may elevate risk and they are being channelled away from domestic (or any) customer protections. These customers are also likely to be most sensitive to ‘sharp’ practices (dishonest bonus terms, excuses not to pay out etc), and so the fairness of their gambling experience is also likely to be impacted.
Customers that are ‘spend sensitive’ but not ‘price-sensitive’ make up the vast bulk of regular recreational players. In absolute terms, they are still not a big cohort: the mass market which is driven by events rather than an amount to be gambled over a period is much bigger in more mature jurisdictions. However, these customers will represent c. 60-80% revenue, including substantially all High-Value Customers. Some of these customers might be attracted to a black market but they are surprisingly sticky so long as the customer experience is adequate (hence France, Germany, Portugal or Poland having a domestically taxed market at all). The customer protection issue for these customers is more subtle, not black market related but worth mentioning. By artificially increasing the cost of their pastime, the fiscal-regulatory framework is accidentally requiring that more or money be spent gambling to get to the same level of enjoyment. This may lead to more harmful play and is an area that should generate more study than it does, in our view.
There are also important policy choices here. For example, it may be viewed that excessive promotional offers are a bad thing. There is logic to this position being taken. However, it is hard to argue that tax is the most effective or targeted way of reducing excessive bonus offers: it is a blunt instrument at best and can cause material distortions. Similarly, there is clearly governance merit in keeping tax and regulation separate, but if taxes become part of the problem, then this separation can lead to some of the causes of gambling-related harms to be misunderstood or fall through the cracks.
One additional point we should make here is that increasing taxes on the NGR level has no impact on price and operators who threaten to increase prices to ‘offset’ higher net revenue taxes are simply threatening to cut off their noses to spite their face. There is no causal link, and a reasonably competitive market creates its own demand corrective. When adopted, this position is either disingenuous or economically illiterate; either way it does not serve the industry well to make the argument (see our September 2018 blog ‘sports margins – price wars?’).
Banning some or all forms of marketing and advertising is perhaps a cathartic way to remove one of the key areas of ‘noise’ in the political debate and ‘simply’ undo one of the more controversial elements of the ‘visible’ rise of online gambling (landbased gambling companies are rarely marketing-led and so channel shift increases gambling advertising almost by default).
From an academic standpoint, the efficacy of advertising restrictions is nuanced at best and broadly speaking gets less compelling the more swingeing the intervention. However, black market risk is also flagged as a reason to beware of marketing restrictions. On a common-sense level, this resonates: the lack of visibility of regulated sites reduces the likelihood of them channelling business vs. unregulated sites. This is a very compelling argument for the mass market, but this is not usually the focus area of regulatory restrictions. The vast majority of online gambling revenue is still generated by ‘lean forward’ customers, who will seek out supply and judge price, product, user experience and customer service. Whether an operator is domestically regulated or not is unlikely to feature highly (if at all) in the decision to deposit. In this context, marketing is about share rather than growth and much of it is generated because everyone else is doing it. A case is sometimes made that mass-market advertising ‘normalises’ gambling, but we are still unsure what is so abnormal about it if properly regulated than it cannot be advertised responsibly.
Advertising restrictions may therefore inhibit the growth of a mass market and may give advantages to omnichannel operators, those with established brands and/or those with other means of customer engagement. But the only binary engagement within that mix is the mass market, which is not a material black market risk. We do not therefore see many causalities between restricting visibility and growing a black market.
User Experience Restrictions
Until very recently user experience and customer service were decided by the operational and technical competence of the operator. Increasingly there is a regulatory dimension to this, however. Regulatory interventions that the customer does not notice will obviously have little or no effect. Interventions that the customer does notice will start to have an effect, eg:
- Mechanisms to prevent gambling
- Mechanisms to restrict the time or money spent gambling
- Invasive customer interactions
Some of these are undoubtedly good policy and customer inconvenience needs to be measured against protection. But for that measure to work a critical mass of customers must submit to the inconvenience or the result is the needless reduction in a safe leisure pursuit or leakage to the black market. With the exception of Czech in-person registration and some Covid-19 related deposit caps, Britain is leading the way in customer-centric interventions. We have already seen a significant reduction in high-value customer activity across a range of operators (WH, 888, LeoVegas, Flutter, Gamesys, bet365, Rank to name a few and cover most of the market). An important question is whether these customers have modified their behaviour or simply moved across to unregulated sites.
The big problem with answering this question is that we are dealing with comparably few customers: the top 1% of GB actives is likely to represent c. £1bn+ of revenue but is only 100k people. The 500k people one rung down is likely to represent c. £0.5bn revenue. The number of people who have signed up to GAMSTOP is an even smaller figure but there is already a specific underground avoidance industry. A cohort of people this small is hard to track: it can easily be lost in the search habits and even complaints of the c. 9m people in Britain who are very unlikely to consider any black market operators because they currently have no need to. Regulatory decisions that push adverse user experience into that 9m, or even become more onerous for the 1-2m more engaged gamblers may start to move that dial, however. For example, a recent survey conducted by Racing TV found that 83% of respondents thought proposed affordability checks would push them underground.
We would make three broad conclusions about what creates black markets. First, any material restrictions to popular products and/or price distortions give more engaged customers a clear need to find more attractive alternatives. Since these tend to be more engaged gamblers, they are usually a relatively small number of people but can represent very high levels of revenue and also are likely to be at elevated risk of excessive gambling. Second, the deeper the product impact, the more people will be captured and incentivised. Third, user experience related interventions are becoming a key battleground in terms of the drivers of black market engagement.
These conclusions may sound anodyne, but they are often lost in the debate. To look for problems in the black market we are not necessarily looking for big numbers but what causes certain customers to find what they are looking for elsewhere. Black markets are therefore most often caused by a small cohort of engaged customers. This leads to four more important policy conclusions:
- The fact that black markets are hard to see does not mean they are not there
- The most engaged customers are also the most likely to jump through enforcement hoops (using VPNs, alternative payment methods, spending time researching evasion methods)
- The most engaged customers are also likely to be the cohort most at risk of problematic gambling
- If a black market exists to solve a particular problem (eg, how to get around GAMSTOP in the UK) the fact that it is small hardly causes for encouragement: its purpose is to target the most vulnerable, which by default is a small market compared to overall market size
The debate about black markets should not be about guessing at scale either to inflame fears or dismiss them. Gambling policies impact player behaviour and at some point in the continuum of restriction and intervention, all players will either stop or seek alternative supply. The basic problem is that those most in need of protection are also far more likely to seek alternative supply rather than stop, while those who are less vulnerable are far more likely to just stop.
Policies which encourage black markets, therefore, have a doubly invidious outcome:
- They reduce the protections available to engaged players, including those most susceptible to harm
- They needlessly reduce the size, attractiveness and efficiency of the legitimate regulated market
If we are serious about making gambling safer it is time to spend more time understanding these behavioural drivers better and less time arguing about statistics.
Feature article edited by SBC from ‘Winning Post’ Thursday 21 January 2021 (click on the below logo to access a full unedited version)