Lee Paul Gibson v TSE Malta LP (t/a Betfair)

The main question in this appeal is whether the defendant (“Betfair”) knew or ought to have known that the claimant, Mr Gibson, was a problem gambler in the ten-year period from 2009 to 2019 when he was gambling on a betting exchange hosted by Betfair or its predecessors. In that period, Mr Gibson, who owned a large property portfolio of tenanted properties generating substantial rental income, lost a great deal of money, over £1.4 million. In this action, Mr Gibson seeks to recover the greater part of that sum from Betfair on the ground, he says, that it did know (or ought to have known) about his problem and therefore should have taken appropriate steps to stop him. In failing to do this, Mr Gibson contends that Betfair breached its obligations set out in the regulatory code issued by the Gambling Commission, which is called the Licence Conditions and Code of Practice (the LCCP).

Before this court, Mr Gibson’s claim for damages is brought solely as a claim in negligence, with previous claims for breach of statutory duty and breach of contract having been dropped. Mr Gibson’s case is that: (i) Betfair owed him a duty of care to prevent him from incurring the losses that he did; (ii) Betfair breached the LCCP; (iii) because Betfair breached the LCCP, it therefore also breached its duty of care; and (iv) this breach caused his loss.

The trial judge, HHJ Bird, sitting as a judge of the London Circuit Commercial Court, dismissed the claim. The judge found that although the expert psychiatrists at trial agreed that Mr Gibson had a gambling disorder, at least from 2015, Betfair did not know he was a problem gambler and rejected the argument that they ought to have known. An important finding which forms part of the basis for this conclusion was that Betfair had made enquiries at the time about whether Mr Gibson could afford his losses and Mr Gibson had shown them that he was wealthy, with a large property portfolio, and had satisfied Betfair at the time that he could afford it (see the whole passage addressing his dealings with Betfair, starting at [77] and culminating in the findings at [128]-[132]). Therefore, there was no breach of the LCCP (judgment [132]). The judge also made a number of further relevant findings, summarised at [133]/[134], that it would have been very difficult to identify a problem gambler from data analysis alone; that Betfair’s approach to compliance evolved over time and was driven by a desire to root out problem gamblers; that the relevant LCCP provisions in place until 2019 were relatively undemanding; and that Betfair had appropriate polices in place and applied them.

The judge also held that Betfair did not owe a relevant duty of care ([156]-[169]), nor was he satisfied that causation of loss had been established even if there was a breach of duty, because if Betfair had stopped him gambling at the relevant time, Mr Gibson would have gambled elsewhere and to the same extent (cf Calvert v William Hill [2008] EWHC 454 (Ch) (Briggs J) and on appeal at [2008] EWCA Civ 1427).

Mr Gibson also advanced an alternative claim, for a lesser amount, based on the argument that the effect of s33 of the Gambling Act 2005 is that if Betfair’s policies and procedures were not compliant with the LCCP then that means all the gambling contracts entered into on its exchange were illegal and void and so Mr Gibson has a claim against Betfair in unjust enrichment. The judge rejected this submission about the effect of s33 of the Act, holding that it did not operate to void gambling contracts in this way ([176]-[182]).

Mr Gibson appeals on five grounds. Ground 1 challenges the conclusion that there was no breach of the licence conditions, ground 2(a) challenges the finding of no duty of care, ground 2(b) challenges the finding of no negligence, and ground 2(c) challenges causation. Finally, the fifth ground, numbered ground 3, challenges the conclusion about s33 of the Gambling Act that gambling contracts entered into in breach of the licensing conditions are not void.

What I have called the main point on appeal arises on ground 1. This is the question of what Betfair knew or ought to have known about Mr Gibson. It was recognised that unless this point on the judge’s factual findings succeeds, the other grounds cannot assist Mr Gibson. Both ways of putting Mr Gibson’s case – in negligence (grounds 2(a), (b) and (c)) and in terms of void contracts (ground 3) - depend on overturning the finding that Betfair did not have actual or constructive knowledge that Mr Gibson was a problem gambler and did not breach its licensing conditions.

Permission to appeal was given by Males LJ on the basis that the proposed appeal raised issues of some general importance as to the responsibilities of gambling organisations, which merited consideration by the Court of Appeal despite the judge's factual findings.

As the judge explained at [2] and [18]/[19] of his judgment, the Gambling Act 2005 represented a watershed in gambling regulation. It was intended to recognise the economic benefits that safe gambling can bring whilst at the same time to introduce protections to safeguard vulnerable people against its dangers. The Act came after a review by the Gambling Review Body in 2001, which made various recommendations on regulation, and the Government’s response to that review, in a paper called “a Safe Bet for Success”, which recognised a need to modernise the law but acknowledged the need to protect children and the vulnerable from the dangers presented by gambling.

The judge accurately explained the scheme of the Act at [20]-[21]. In summary, the scheme is as follows.

Section 20 establishes the Gambling Commission and section 24 requires the Commission to issue codes of practice “about the manner in which facilities for gambling are provided”. By s24(2), codes of practice must describe arrangements that should be made by a person providing facilities for gambling for the purposes of, amongst other things, protecting children and other vulnerable persons from being harmed or exploited by gambling.

Section 24(8) provides essentially that a failure to comply with the code is not itself an offence, nor does it make a person liable to a civil claim. The provision in full is:

“A failure to comply with a provision of a code shall not of itself make a person liable to criminal or civil proceedings; but this subsection is subject to any provision of or by virtue of this Act making an exception to an offence dependent on compliance with a code”.

Section 33 provides that a person commits an offence if he provides facilities for gambling unless certain statutory defences apply. An important defence is set out by s33(2) – namely that the person (a) holds an operating licence authorising the activity, and (b) the activity is carried on in accordance with the terms and conditions of the licence. “Facilities for gambling” is a term defined in s5(1), and s13 then provides that a betting intermediary is someone providing facilities for betting. Thus there is no dispute that Betfair provides facilities for gambling nor any dispute that since 1 November 2014 Betfair has been required to hold and comply with an operating licence issued by the Commission, even though it is a company registered abroad.

Section 82 makes clear that any operating licence is subject to the condition that the operator ensures compliance with any relevant social responsibility provision contained in a code of practice issued under s24.

The relevant code of practice issued by the Commission under s24 of the Act is the LCCP. The relevant part of the LCCP in the May 2015 version is provision 3.4.1 concerning Social Responsibility, which provides:

Social Responsibility Code Provision 3.4.1

Customer Interaction

All licensees, except gaming machine technical and gambling software licenses.

1.

Licensees must put into effect policies and procedures for customer interaction where they have concerns that a customer’s behaviour may indicate problem gambling. The policies must include:

[ … ]

c.

The circumstances in which consideration should be given to refusing service to customers and/or barring them from the operator’s gambling premises.

[ … ]

e.

Specific provision for making use of all relevant sources of information to ensure effective decision making, and to guide and deliver effective customer interactions.

As it says, paragraph 1 of this provision requires the licensees to put policies and procedures into effect for customer interactions in certain circumstances. We were taken to various documents which contained or reflected aspects of Betfair’s policies and procedures over time. However, save for one point about refusing service, which I will address separately below, the detail of these documents did not matter. That is because the premise of the appellant’s case based on paragraph 1 of provision 3.4.1 of the LCCP was an issue of fact about the state of Betfair’s knowledge whether Mr Gibson was likely to be a problem gambler. Mr Gibson’s case was approached on the basis that since, as he alleged, Betfair did know or ought to have known he was a problem gambler, Betfair should have refused service or barred him for using their platform.

The single point about the procedure and policy documents was a submission about what provision 3.4.1 required of them and a contention that Betfair’s procedures and policies did not in fact include a provision that Betfair would ultimately refuse service to a problem gambler. The point arose I think as a kind of fallback argument. There is no dispute that Betfair did not in fact refuse to serve Mr Gibson for any appreciable time in the relevant period. There were temporary suspensions due to anti money laundering (AML) checks (see below) but for this purpose they are irrelevant. I will come back to that briefly later.

There are also two preliminary points to make about knowledge. First, the concept of a “problem gambler” and “problem gambling” could, in a different case, raise questions about the criteria which have to be satisfied and how they relate to clinical definitions of gambling disorders, including for example the DSM criteria. The judge referred at [120] to a definition offered in the British Gambling Prevalence Survey 2010 that “problem gambling is gambling to a degree that compromises, disrupts or damages family, personal or recreational pursuits” and approached the matter that way. The only submission on this appeal, which suggested that the judge applied the wrong test for identifying a problem gambler or problem gambling, related to a point of detail about capital and income in the context of the affordability of losses. Aside from that, which I will address in context, the wider ambit of the definition of problem gambling does not arise and I will not examine it.

Second, no distinction was drawn between “having concerns that a customer’s behaviour may indicate problem gambling” (the words of the LCCP) and the test of knowledge applied by the judge. I do not believe there is any practical difference, on the facts of this case at least, between the concept described in the LCCP and the judge’s approach.

Recognising that this aspect of the appeal is a challenge to a finding of fact, Mr Gibson’s case is put on the basis of The Mayor and Burgesses of Haringey v Ahmed & Ahmed [2017] EWCA Civ 1861 at [29]-[31] that the finding was unsupported by the evidence and that the conclusion was one which no reasonable judge could have reached. Referring to In Re B [2013] UKSC 33 at [44], the submission is simply that the judge was wrong.

The judge’s factual conclusions were summarised at [128]/[129] as follows:

“[128] It cannot in my judgment be said that Betfair knew or ought to have known that Mr Gibson had a gambling problem at any time between 2009 and 2019.

[129] The difficulties faced by Mr Gibson in making good this point include

that he kept his gambling problem to himself,

he could (at least on the face of the information he gave to Betfair) afford to fund his gambling,

he misled Betfair about his gambling

and it is very difficult to identify a problem gambler who is not being honest.

Taken together these difficulties are in my judgment insuperable. In my view Mr Gibson did not simply fail to share information about his gambling problem, he took steps actively to hide it and to portray to the world at large, and to Betfair in particular, a wholly inaccurate picture.”

[I have separated out the points in [129] for ease of reference]

Then at [130] the judge held that Betfair had appropriate policies in place; at [131] he rejected Mr Gibson’s uncorroborated evidence that he (Mr Gibson) had shared his gambling problems with others (an unnamed man in a Betfair hospitality box at Old Trafford and the landlord of his local pub); and at [132] the judge concluded there was no breach of the LCCP.

Mr Gibson’s case was that these overall findings at [128]-[132] failed to take reasonable account of a number of facts the judge had found earlier in the judgment and a number of further facts which the judge ought to have found. The first group of points relied on are worth setting out in full, along with the paragraph references where the judge found these facts. They are as follows:

“(1)

Mr Gibson lost £1,480,728 across the 10 years he gambled with Betfair [1];

(2)

From June 2012 his account remained at all times in deficit [77];

(3)

Heavy losses can be a fairly obvious indicator of problem gambling [135];

(4)

Mr Gibson had lost £300,000 by 11 October 2014 [77];

(5)

Mr Gibson was clearly frustrated when his access to the Betfair exchange was temporarily suspended for [Anti Money Laundering] AML checks [87] [97];

(6)

Betfair knew he was selling properties and re-mortgaging them [110] (the Judge’s finding at [110] that there was nothing obvious in the transactions revealed to Betfair that would suggest he was doing other than acting broadly in the normal course of business is wrong and unsupported by the evidence);

(7)

Mr Gibson’s income before personal expenses and tax was, and was presented to Betfair as being, in the region of £20-£25k per month [86] [96];

(8)

In June 2018, the AML review conducted concluded that, even taking account of selling property to obtain funds to gamble, he did not appear to be able to fund his losses [95] [96].”

The second group are the following. I have renumbered them from the numbering used in the appellant’s skeleton at paragraph 14 to follow on from (8) above. They are:

(9)

Mr Gibson placed at least 20,000 individual bets in the six years prior to 22 January 2021, which is more than 5 per day;

(10)

Mr Rourke [of Betfair] had a detailed understanding of Mr Gibson’s betting. He received regular reports of the top 5 daily losers from his VIP cohort and his customers’ revenue generation and therefore knew that on some days, Mr Gibson had lost in excess of £30,000 and that by 7 November 2014 his betting activity had escalated to a lifetime high;

(11)

As part of AML checks, Betfair knew Mr Gibson’s home was valued at £214,000 which was much less than the level of his gambling losses;

(12)

Mr Gibson was placing bets on obscure football matches that may indicate a gambling problem;

(13)

Mr Gibson’s level of betting was very unusual and that if a person was staking £20,000 on a single correct score market, as Mr Gibson sometimes did, his bona fides should be established within a few days.

Points (1), (2) and (4) are matters of primary fact which the judge clearly did have in mind. To grapple with the further points, I will focus on the difficulties the judge identified in [129] (set out above), taking them in turn.

The first difficulty identified by the judge in [129] was that Mr Gibson kept his gambling problem to himself. This was a conclusion amply supported by the evidence. For example, Mr Rourke who was Mr Gibson’s VIP manager at Betfair, and whom the judge found to be a reliable and impressive witness and of whom the judge rejected various allegations made against him (at [14]), described how Mr Gibson presented himself to him. Mr Rourke said Mr Gibson was someone who always gave the clear impression of enjoying his gambling and who never mentioned anything remotely concerning. Mr Rourke also said that Mr Gibson was calm, level-headed and rational.

The second difficulty identified by the judge in [129] was that Mr Gibson could (at least on the face of the information he gave to Betfair) afford to fund his gambling. The judgment deals with this in considerable – and appropriate – detail. There are two topics to cover on this aspect: AML and the test for affordability.

From [78] to [108], the judgment reviewed the contact between Mr Gibson and Betfair over time. A striking feature of these interactions is that many of them involve AML checks. Essentially what was happening was that if an AML check was triggered then information about how the account was funded would be sought from Mr Gibson by Betfair. The precise triggers were confidential but for whatever reason Mr Gibson’s activity did trigger a number of AML checks.

If information about how the account was funded was not supplied to Betfair’s satisfaction in time then Mr Gibson’s account would be suspended. Once information was supplied, essentially as to the source of funds, the suspension would be lifted. For example, the first recorded AML request was in 2015 and Mr Gibson was cleared to use his account on 13 August 2015 ([81]). Checks of this sort happened repeatedly over time and they are the reason why Betfair was given detailed information by Mr Gibson about his finances and about where the money he was gambling with came from.

At all times, Betfair had two relevant internal teams, one concerned with AML and the other, led in 2017/18 by Ms Lawson, concerned with responsible gambling. Ms Lawson gave evidence at trial and the judge found ([14]) that she took her role very seriously and was dedicated to her work. She was keen to ensure that appropriate steps were taken to identify and support problem gamblers. The judgment held that it was impossible to separate AML requests from responsible gambling interactions ([80]), again a finding well supported by the evidence.

Points (5) and (8) made by Mr Gibson (above) relate to AML. Mr Gibson certainly was frustrated when his account was suspended due to AML checks (point (5) above), but the judge clearly had that well in mind and took it into account – see [87] and the paragraphs leading up to it. Point (8) is accurate as far as it goes, but the paragraphs referred to are part of an overall passage in the judgment from [94] through to [100], which is revealing. The particular interaction starts on 19 June 2018, when Mr Gibson’s losses were around the £1m mark. Mr Rourke emailed Mr Gibson asking for AML information giving a full explanation as to how he was funding his account. His account was suspended. Later in June some information had been provided – which included references to income and to remortgaging properties – but Betfair was still not satisfied and Betfair was, as an internal email put it, “a long way off as being comfortable with his source of wealth”. This aspect is in [95]/[96] and is what is referred to at point (8) above. Mr Gibson was frustrated ([97]) and the interaction continued. Mr Gibson provided further information ([98]), which included references to selling three houses in 2016/17 and estimates of yearly profits. At [99], the judge noted that now Betfair were satisfied that they had sufficient evidence of his sources of wealth. The account was reopened ([100]) and Mr Rourke sent a message to Mr Gibson about responsible gambling, asking if he was comfortable with his level of spending. He replied that he was.

The information provided to Betfair as a result of the AML checks feeds directly into the next topic of affordability.

A final point is the suggestion that even if Betfair’s AML team assessed Mr Gibson’s losses as affordable, the responsible gambling team had a distinct obligation to assess the same question. However, whatever is the position today, at the relevant time there was no requirement that an operator’s responsible gambling team was required separately to assess affordability.

There was extensive evidence that Mr Gibson could afford his losses and that he presented evidence of this to Betfair. That provides important context for point (3), that heavy losses can be an indicator of problem gambling, as the judge recognised.

Starting at [15], amongst other things the judge noted that Mr Gibson told Betfair at one point that he was a “multi-millionaire”. At [109], the judge noted that Mr Gibson had consistently and often “reassured Betfair he was able to fund his gambling (including his losses) and none of the information he provided Betfair painted a different picture”. The judge also noted ([110]) that Betfair knew Mr Gibson was selling and remortgaging properties, but that there was nothing obvious that would suggest he was doing anything other than acting broadly in the normal course of business. After all, Mr Gibson owned a large property portfolio, which he rented out. At [111], the judge noted that Mr Gibson was so determined to gamble that at one point in 2018 he deliberately misled Betfair about whether he had legal advice that what they were doing infringed his rights. At [121], the judge noted Mr Gibson’s evidence that his own home was valued at £300,000 and he held properties worth £5 million.

There was a suggestion that in truth Mr Gibson could not afford his losses. I can see no justification for overturning the judge’s conclusion that he could, but irrespective of that, the issue is not what Mr Gibson could in fact afford but what the information he presented to Betfair showed. As the judge dealt with in detail, the information from Mr Gibson did show that he could afford it.

There was also a point of detail about whether the dimension of the definition of problem gambling which concerned affordability of losses was such that an indication of problem gambling was an inability to fund losses from income (as opposed to capital) (see points (7) and (11)). So, it was said that even if losses could be funded from capital (i.e. Mr Gibson’s property transactions) such losses were to be treated differently and the fact they were not funded from income was an indication of problem gambling. The difficulty with this submission is that Mr Gibson’s business involved holding a property portfolio and involved buying and selling properties in that portfolio. The issue the judge had to decide was not a point of accounting treatment. Rather, it was a practical matter about Mr Gibson’s behaviour. I can see no ground for criticising the judge’s approach on these facts. The nature of Mr Gibson’s business was itself ample support for the judge’s finding at [110] that there was nothing obvious from Betfair’s point of view to suggest Mr Gibson was not acting broadly in the normal course of his business (contrary to point (6) above).

The finding that Mr Gibson could (at least on the face of the information he gave to Betfair) afford to fund his gambling was not only open to the judge but was in my judgment the right conclusion.

The third point made in [129] was that Mr Gibson misled Betfair about his gambling and that it is very difficult to identify a problem gambler who is not being honest. An aspect of the issue here is the suggestion that data about Mr Gibson’s pattern of betting, in the sense of the number of bets, their value, the size of the losses and the fact the bets were on obscure football matches, were indicative of problem gambling. This is essentially what points (9), (12) and (13) seek to address.

However, the finding that Mr Gibson misled Betfair ([111]) was clearly open to the judge. Related to it was the point that Mr Gibson repeatedly told Betfair he was comfortable with his gambling ([15]). Also relevant is the conclusion at [133] that it would have been very difficult to identify a problem gambler from data analysis alone, another finding for which there was clear evidence. We were shown a single document which suggested that data analysis could be used for this at the time but there was also substantial evidence before the judge in the other direction, and the document we were shown was an example of the now well-known phenomenon on appeal of “island hopping” (per Lewison LJ in Fage UK Ltd v Chobani UK Ltd [2014] EWCA Civ 5 [2014] FSR 29). Whether data analysis today can do this is a different issue which does not arise on this appeal.

The suggestion at point (10) seems to be that the judgment is lacking in some way in not setting out more detail on Mr Gibson’s betting pattern (points (9), (12) and (13)) and then in not explicitly finding that Mr Rourke had a detailed understanding of Mr Gibson’s betting and some or all of these features in particular. However, read as a whole, it is manifest that the judge had a clear understanding of what the level of Mr Gibson’s activity was and that Mr Rourke, as Mr Gibson’s VIP manager, had a detailed knowledge of it. The problem on this appeal is that, seen in context, these points do not assist Mr Gibson.

Standing back, the judge considered the relevant evidence, made no errors of principle in doing so and came to an entirely reasonable conclusion on that evidence. I can see no justification for allowing the appeal against the conclusion that Betfair neither knew nor ought to have known that Mr Gibson was a problem gambler.

The final aspect of ground 1 of the appeal was about what Betfair’s procedures and policies were, and a submission that this did not comply with the LCCP. Therefore, the judge’s conclusion at [130] and [134] that Betfair had appropriate policies in place and which followed his detailed judgment on compliance at [44] to [74], was wrong. I do not see how this point can assist Mr Gibson on appeal having failed to overturn the factual finding above but it can be dealt with shortly and so I will do so.

The argument was essentially that the most Betfair’s policies and procedures at the time contemplated was suspension of an account, but never outright refusal of service. This is said to be a breach of the LCCP because provision 3.4.1 includes a requirement that there must be a policy that once an operator does have concerns that a customer is a problem gambler, they will ultimately refuse service to that person, which in turn is said to follow from the words in paragraph 1 coupled with sub-paragraph (c) of this provision, the relevant parts of which amount to the following:

“[…] The policies must include: […] The circumstances in which consideration should be given to refusing service to customers and/or barring them from the operator’s gambling premises.”

Betfair points out that the judge held at [133] that the relevant licence provisions in place until 2019 were relatively undemanding, and also that these terms in the LCCP have been superseded, but while these points are both true, the fact remains that Mr Gibson’s case is that, undemanding though the LCCP was, it did at least include the requirement identified.

The real difficulty with Mr Gibson’s submission here is that the premise of his case, that Betfair’s policies did not envisage exclusion, is not right, and so the point fails in any event.

The starting point on appeal is that the judge did not make a finding either way on this point. However, Ms Lawson gave clear oral evidence on this (transcript at Supplementary Bundle 3 or 5 p830). She was asked if Betfair “might” close an account and replied that if the player confirmed her team’s suspicions that they were struggling then: “we would close. There wouldn’t be a maybe. There would be complete closure at that time.” Therefore, since the premise of this submission is not well founded, there is no need to resolve the true construction of this part of the LCCP.

Therefore, I would dismiss ground 1 of this appeal.

Given the dismissal of ground 1, grounds 2(a) to 2(c) do not have to be decided to dispose of this appeal. These grounds raise a number of wider issues but as so often happens, once a full appeal is heard, the importance of the judge’s factual findings, as forming the context in which any wider issues fall to be considered, becomes even more evident than it could be at the permission stage. In Housden v The Conservators of Wimbledon and Putney Commons [2008] EWCA Civ 200; [2008] 1 WLR 1172 at [30] Mummery LJ said this:

"In general, it is unwise to deliver judgments on points that do not have to be decided. There is no point in cluttering up the law reports with obiter dicta, which could, in some cases, embarrass a court having to decide the issue later on."

Bearing in mind Mummery LJ’s wise observation, I will not consider grounds 2(a) to (c). They would be much better considered in a factual context in which they actually arise.

Ground 3 also does not have to be considered to decide this appeal and so anything I say will be obiter. Nevertheless, since it is a pure point of law and not so dependent on the factual context, I will address it briefly.

The argument relates to s33, which as mentioned above, provides that a person commits an offence if they provide facilities for gambling unless certain statutory defences apply, one of which (s33(2)) is that the person (a) holds an operating licence authorising the activity, and (b) the activity is carried on in accordance with the terms and conditions of the licence. The effect of s24 is that it is a licence condition to comply with any relevant social responsibility provisions of a code of practice issued by the Commission.

Mr Gibson argues that these terms of s33 of the Act have the effect that gambling contracts concluded where one party is in breach of its licensing conditions are void ab initio. Therefore, if he had succeeded in showing a breach of the LCCP, he would have submitted that this rendered the relevant betting contracts void. The judge also held at [149] that there was a single contract between Mr Gibson and Betfair over the period rather than individual contracts for each bet, but that only falls to be addressed if Mr Gibson’s case on s33 succeeds.

In law, a convenient starting point on the statutory illegality of a contract is Underhill LJ’s comprehensive judgment in Okedina v Chikale [2019] EWCA Civ 1393, with whom Davis LJ and Nicola Davies LJ agreed. In his judgment, Underhill LJ took into account the earlier authorities, some of which were also cited to this court, including Cornelius v Phillips [1918] AC 199, Phoenix Insurance v Halvanon Insurance [1988] 1 QB 216 and Hughes v Asset Managers plc [1995] 3 All ER 669, CA.

The question to be answered is whether the statute intends to deprive the contract of any legal effect with the result that it is unenforceable by either party, which depends purely on construction of the statute (Okedina [17]). The prohibition may be express or implied ([18]-[19]). When the issue is whether the statute contains an implied prohibition, normal principles of statutory construction apply, albeit there are points of particular relevance. One such point (citing Phoenix Insurance in particular) is that even if one party is prohibited from entering into such a contract, it does not follow that Parliament intended to render it unenforceable by either party. Whether that was the intention must depend on a consideration of all relevant factors including matters of public policy.

Turning to s33 itself, it was common ground that it does not contain an express prohibition which would operate to render contracts void. The issue is about what the Act implies. The judge held that s33 did not have the effect Mr Gibson contended for in [176] to [182] with the conclusion on the law at [181]. He held that the Act:

“ … does not ban gambling as an activity and cannot have been intended by Parliament to operate so as to void contracts. It simply imposes a penalty on one party. In my view, public policy overwhelmingly favours the enforceability of gambling contracts even when the operator is in breach of his licence. A successful gambler should not be deprived of the fruits of his bet, but equally in my judgment a losing gambler should not be able to escape the consequences of his decisions. [...]”

He went on to find support for that conclusion in s336 of the Act, which includes an express power for the Commission to void an unfair bet and provides that where a bet is voided “any contact or other arrangement in relation to the bet is void”. Since this only gives rise to a power to void a bet when it is unfair, that would be inconsistent with the interpretation of s33 which would void all bets if there was a breach of the LCCP. The judge also thought that Mr Gibson’s interpretation of s33 would lead to chaos because every unsuccessful gambler would potentially be able to claim lost stakes back on the basis of any breach of the LCCP. Finally, in terms of public policy, considering the position of a gambler who had placed a successful bet, the judge observed that it would (at least in the common case where the provider was a counter party to the bet) be contrary to public policy if that gambler was deprived of their winnings.

I agree with the judge, essentially for the reasons he gave. Looking at the words used in s33, they are focussed on penalties addressed to one side (the operator) rather than applying to both parties. In that sense, it is very like the situation in Phoenix concerning insurance contracts, and I refer to the points on public policy made there by Kerr LJ at 273 at A-E. The statutory prohibitions in that case (as here) were designed to protect the customer, but as Kerr LJ pointed out drawing a comparison with money lending contracts, the way the prohibitions work in policy terms depends on the context. Therefore, one can see why it would be good public policy to refuse to enforce the former (moneylending contracts) but bad policy in the case of the latter (insurance contracts). As Kerr LJ explained in the insurance context:

“… The statutory prohibitions are designed to protect the insured by seeking to ensure that undesirable persons are not authorised to carry on insurance business and that authorised insurers remain solvent. Good public policy and common sense therefore require that contracts of insurance, even if made by unauthorised insurers, should not be invalidated. To treat the contracts as prohibited would of course prevent the insured from claiming under the contract…

Although not exact, there is an analogy between this observation and the position of a gambler who placed a successful bet in this case. There is no policy reason why successful bets should be unenforceable (by the gambler against the operator) even if the operator is in breach of the LCCP.

In other words, reading s33 against the scheme of the Act as a whole, it would in my judgment be a very odd conclusion that this section was intended to render all gambling contracts void if the operator was in any breach of their licensing conditions. That conclusion does not sit easily with the power of the Commission in s336. Nor does it sit well with the terms of s24(8) (above), in which the legislator positively turned their mind to the consequences of a failure to comply with the code, but did not provide for the voiding of all contracts in such circumstances.

A submission made on Mr Gibson’s behalf was that the protection of vulnerable customers was “the” priority of the Act and so this was said to support the argument about implying this effect in s33. However, while protection of the vulnerable clearly was one of the most important priorities in the Act, it was not the only one. Modernising the law was also important, as well as setting up a new regulatory scheme. Furthermore, even if protecting the vulnerable could be said to be the primary priority amongst the Act’s overall purposes, that would not in my judgment go far enough to find an implied prohibition of the relevant kind in s33.

Mr Gibson’s argument is that any breach of the LCCP has the effect of rendering the contract void. That would therefore allow a losing gambler to avoid paying his gambling debts irrespective of any vulnerability and irrespective of whether the breach of the licence conditions was of any relevance to the bet in question. Such a result would be entirely contrary to the policy of the Act which, so it seems to me, is that in general gambling debts are enforceable.

Therefore for all these reasons, I would reject ground 3 of this appeal.

Finally, it may be worth mentioning that when this appeal was heard the Chancellor of the High Court was Sir Julian Flaux and I was Lord Justice Birss, whereas today we have the roles ascribed to us in this judgment.

I agree.

I also agree.