Winning Post

 

Portugal: online gambling – the curious case of Portuguese growth   The Portuguese online gambling market has a habit of defying expectations. The regulatory regime that many commentators thought would ‘never work’ back in 2016 due to high and distortive taxes has proved resilient and attractive to its domestically licensed operators (see WP’s passim). Indeed, even in 2019 revenue per capita in Portugal of €17.46 (net of estimated bonuses) was 10% higher than in ‘liberal’ neighbouring Spain. After a relatively stable start to 2020 and an anticipated Q320 boost to betting due to Covid-19 policy disruption, Q420 revenue leapt forward far more strongly than anywhere else of similar market maturity (+77% YoY, +37% QoQ; Spain comps: +25%% YoY +20% QoQ), underpinned by a surge in active customers (see chart below). Portugal’s Q121 results demonstrate that this is not just a flash in the pan (albeit a quarterly customer-driven surge usually lasts the following quarter vs. a margin-driven reversal). Portugal’s run-rate revenue is now almost double that of Spain on a per capita basis (annualised €37.57 vs. €19.70; Q4 on Q4 as Spain has not yet published Q121 results). Entain’s previously rather bullish expectation of the market doubling in three years suddenly looks more credible. So what is happening? The first point that we would make is to clarify what is not happening. Portugal’s active base of 770,500 customers represents 7.5% of the population. While this is a significant increase on the 2.6% captured at the end of 2018, it is still not in mass market territory. For example, the UK equivalent figure is 15% and the majority of customers are solus bettors. In Portugal, the number of domestically captured customers who are solus bettors has actually declined from 50% at the beginning of 2019 to 35% now. Domestically regulated players who use both betting and gaming has been a key driver of growth: from a relatively stable c. 17% Q119-Q220 to 36% in Q121, or +6ppts per quarter. Portugal’s growth spurt has therefore been driven by attracting more active gamblers rather than engaging a large number of more occasional bettors. This behaviour also gives a clue as to where the growth may be coming from, as we discuss below.   We believe that three things have combined to make the growth acceleration happen quite suddenly from the second half of 2020. The first and most obvious is continuing Covid-19 policy disruption, which is driving strong volumes in a number of Southern European countries with a greater lag effect vs. initial lockdown performance than perhaps initially expected (probably because underlying ecommerce also had more catching up to do: vide Italy ‘despite’ the advertising ban). The second is that Portugal changed its fiscal policies in 2020 to remove the higher rates of tax for both betting and gaming, giving the larger betting-led operators especially an impetus to turn on the marketing taps (to start capturing revenue at the previous higher rate tax bracket of 16% of turnover would have been self-defeating).  Third, after a period of limited domestic competition, the Portuguese market is now better served by a broader range of high quality domestically licensed operators, including Betano, 888 Betway and SkillOnNet for example.   These three factors are undoubtedly creating the conditions for higher secular growth. However, the last two are also likely to be materially improving channelling from a previously strong black market. The immediate higher spending and multi-product nature of the new customers also supports this hypothesis. Due to an 8% turnover tax on betting, a material and structural black market is baked into the system, but Portugal’s continued growth demonstrates just how small the cohort of price sensitive customers is on a country-by-country basis (in terms of actives if not turnover).   There is a sting in the tail to all this growth, however. Notwithstanding Portugal’s gambling tax reforms, the blended tax rate on revenue still averages over 40% (with material volatility caused by the betting turnover tax). With more brands now fighting for a local share of voice shown to be very lucrative by the previous oligopoly of Betclic, bet.pt (now Entain), Estoril Sol, and Flutter (PokerStars), marketing costs are likely to be increasing materially for all .pt operators. While overall ARPUs are up 12% YoY, the number of domestically regulated accounts per active is likely to be increasing at an even faster rate given that the number of credible licensees has roughly doubled. Competitive growth is therefore likely to be putting pressure on operating margins, with local media companies and the Portuguese government being the biggest winners. Further, if our assumption is correct and the Q420-Q121 growth is driven largely by taking a chunk out of the black market rather than a burst of secular growth, then the structural inefficiencies of the Portuguese market will likely reassert themselves. Growing from a 1.9x per capita performance gap from its nearest neighbour will be an increasing challenge, even with the perverse ‘help’ (in reported revenue terms) of a customer tax being baked into Portugal’s betting revenue.    The quest for more domestically regulated revenue continues to make strategic sense for a critical mass of operators. This in turn is driving both adoption and channelling, which then makes markets big enough to justify more competition. But what was once a cosy divide between profitable protected markets and profitable .com markets (with a few outliers) is now being eroded from both ends. The optics of topline growth improve, but regulatory risk and margin pressure both continue to increase, in our view. In this sense, Portuguese exceptionalism is more a matter of timing than outcome.   UK: Regulation – If the CAP Fits The Advertising Standards Authority this week provided an update on the exposure of children to TV advertisements for gambling and alcoholic drinks. While the report offers much of interest, it seems unlikely to help resolve the bickering over how far legislation should go to prevent children from being able to see commercials for betting and gaming.   The ASA revealed a slight increase in the exposure of children to gambling advertisements – from an average of 2.5 per week in 2019 to 2.8 a week in 2020 – against a backdrop of reduced exposure to TV commercials in general (defying the expectations of many lockdown prognosticators). The growth was largely attributable to marketing for lotteries and scratchcards (which constituted a little over one-third of all exposure). Advertisements for sports betting – which for some strange reason continue to attract the lion’s share of criticism - remained low at 0.3 per week. Excluding ads for lottery products, exposure was pretty flat (1.8 per week in 2020 compared with 1.7 in 2019 and 2.2 in 2018). Looked at in a slightly broader context, exposure was substantially lower than the peak of 4.4 advertisements per week recorded in 2013.   Predictably, whether the ASA results were considered positive or not depended upon the a priori views of the commentators – but attachment to opinion in the face of facts can present problems. Those in favour of censorship on the dubious pretext that advertising ‘normalises’ gambling (a lazily imprecise and deeply subjective concept) may struggle for credibility when their particular choice of ‘public health crisis’ displays such shrinking qualities. Meanwhile, those opposed to an advertising ban may struggle to claim virtue for a negative (if reduced exposure is positive, why not intervene more strongly?).   Those who participated in the latest Big Step event this week – an ‘eve of Euros’ walk from Scotland to Wembley Stadium – to call for an end to gambling sponsorship in football do have legitimate and specific concerns. Advertising (including sponsorship) can undermine abstinence and therefore be problematic for those in recovery from gambling disorder. It is unclear whether this constitutes sufficient grounds for a ban (and if it does whether such bans might also encompass beer, wine, pizza, banking and internet shopping) but it does highlight the need for higher standards of marketing responsibility and the swifter development of targeted interventions, such as adtech filtering.   Amidst all of this, the squabbling continued. The Betting & Gaming Council used the ASA report to push back against proponents of an advertising ban - labelling them ‘Sunday School prohibitionists’ in the process. Such ad hominem criticism is not particularly helpful – more suited to the playground than to public discourse – and we should expect better. And yet…and yet…   In a letter to The Times this week, a group of spiritual leaders, headed rather predictably by the Bishop of St Albans called on the expedition of legislation to curb gambling on the basis of NERA’s recent claim that to do so would benefit the economy. There is something unsettling in this attempt to shift the ground for legislative intervention from harm prevention to economic growth. The implication is that civil liberties may justifiably be curbed in order to boost taxation; and that real people doing real jobs may be put out of work, comfortable in the speculation that other jobs might be created elsewhere in the economy. This is all too clearly the logic of prohibition. Its promulgation by faith group leaders – some of whom at least might be reasonably referred to by the ‘Sunday School’ moniker – suggests thinly-veiled moralism behind their calls for stricter legislative controls. If the cap (or Mitre) fits….       US: horseracing – a retirement plan that could revitalise the sport The New York State Assembly has passed a bill relating to the care and tracing of racehorses and breeding stock. It introduces the compulsory microchipping and Jockey Club registration of horses to be eligible to race, creates a fund for the care of retired racehorses, and provides tax incentives for businesses and individuals to donate to the upkeep of retired racehorses. The bill also prohibits the commercial slaughter of racehorses and breeding stock (ie, sale for meat).   The USA has had some of the highest volumes of breeding and racing in the world, even factoring in population size. For example, there were nearly 80,000 races in 1990 with an average field size of 8.9 and average starts per runner of 7.9. In 2020, there 31,000 races (down 60%) with an average field size of 7.6 and average starts per runner of 5.1: a savage cut in supply. The number of horses in training which get at least one start to support this reduced fixture list and engagement per horse, has halved to 47,000, also implying a smaller welfare problem to deal with. In the same period, US horseracing betting handle has grown by 16% (+38% to its peak in 2004; -27% to 2011; then flat), which represents an absolute CAGR of 0.5% or a more meaningful inflation-adjusted decline of c. 2% pa. Halving the number of fixtures and horses in training has actually therefore allowed US horseracing to follow broadly the same long-term underlying revenue trajectory as GB (given that handle is a proxy for revenue in the US), even though GB has added a material number of fixtures over the same period.   The USA arguably still has too much low-quality racing even after a significant cut, a function of its divided ecosystem and generous funding that is not linked to racing-related productivity (racinos, ‘historical racing machines’ etc). However, a growing focus on the individual horse is likely to continue to reduce the appetite for ‘churn’ and further increase a focus on ‘compassionate efficiency’. Fewer horses which better cared for throughout their lives is not only sound from an ethical and sustainability standpoint, it also improves the ‘story’ of racing for racegoers, viewers and the betting public. Racehorse ownership too could be more broad-based if it were less commoditised. Less is more should be the motto of racing regulators and stakeholders nearly everywhere.     Global: boxing – avoid punching down Last week a 6’1’’ twenty-six-year-old YouTuber fought a 5’8’’ forty-four-year-old boxing promoter. The Mayweather - Paul fight drew 1m paying viewers, making it the 25th most watched pay-per-view boxing fight of all time, earning over US$50m in OTT and satellite media revenue. High profile boxing matches are huge crowd-pullers, and the addition of a social media celebrity has provided the potential to reach a broader audience. While boxing cannot compete with soccer for frequency or global reach it provides a first-rate mass-market touchpoint to male audiences especially.   It is unsurprising, therefore, that boxing is a popular product for betting companies to promote and one which is now likely to grow in value given the ability to directly reach so many US audiences (boxing is one of the few sports that has genuine global appeal). For pay-per-view media companies the opportunity remains obvious, the question is simply one of direct economic returns and entertainment cross-sell potential. For betting companies, the issue can be more nuanced. The ability to acquire new customers is key to growth, but the knowledge that a customer likes betting on high profile boxing matches, especially Exhibition matches, is not all that useful in terms of generating a return: there aren’t enough of them to justify any sort of CPA. For horseracing, soccer and tennis this is not a problem: there is plenty of similar content to offer. For boxing (and to a lesser extent US sports) this does present a problem, however. In order for a a large marketing acquisition cost to be made back, the customer is likely need to bet on more than that sport or be cross-sold into gaming. For many, a broad interest in several sports and possibly gaming solves this problem from a demand perspective. For some, operators may be keen to solve this problem with cross-sell. When done responsibly, backed up with appropriate (still evolving) safer gambling measures, and a view to sustainable customer enjoyment rather than short-term expenditure, cross-selling product presents a perfectly manageable shift in risk profile to enhance customer enjoyment. However, for others this may generate unwelcome and potentially harmful additional activity, especially if highly repetitive products (which high profile sporting events by default are not) are involved. Given the growing appetite to engage new audiences, the need to monetise US bettors especially more effectively, and the increasing sensitivity of advertising from a political standpoint, how operators handle this engagement and cross-sell process is likely to be a key regulatory pressure point going forward, in our view. Getting it right now is likely to reduce the likelihood of the default position for gambling reform being activated: political over-reaction triggered by commercial incompetence.

Disclaimer; The analysis provided in this report represents the opinions of the authors. Any assessment of trends and change is necessarily subjective. The information and opinions provided herein are not intended to provide legal, accounting, investment or policy advice, nor should they be used as a forecast. Regulus Partners may act, or have acted, for any of the companies and other stakeholders mentioned in this report.