(Letter published in Irish Field, 14 Jul 2012, p.A5 and, as `Confusion reigns over Betfair calculations', Racing Post, 15 Jul 2012, p.12.)
To: Editors, Racing Post, Irish Racing Post and Irish Field.
I would like to commend Paul Bittar for making peace between the British racing industry and Betfair so soon after his arrival at the helm of the British Horseracing Authority. I would also like to encourage his Irish counterpart Brian Kavanagh to agree a similar deal.
The continued widespread misunderstandings at the highest levels in Irish racing concerning the betting exchange model and its revenue-raising potential are highlighted by the stark differences between the figures which I have just read in this morning's (7 July 2012) Irish Field and Racing Post.
Mark Costello in the Irish Field (p.3) wrote that `Betfair has revealed that it is making EUR13.5 million from Irish customers per annum and a 15% gross profits tax on this sum would yield EUR2 million.'
Brian Graham in the Racing Post (p.24) wrote that Betfair had revenues of EUR10.7 million from their Irish-based clients; no period was specified, but he concluded that `Irish racing should be receiving EUR21.4 million per annum', more than ten times Mark Costello's estimate. Jonathan Mullin in the previous day's Racing Post (p.12) reported that Betfair had revenues of GBP10.7 million from Irish customers last year. It appears that Brian Graham, apart from neglecting to mention the time period to which he refers, has used an implausible exchange rate of EUR1=GBP1 to convert the figures, so that by `revenues' he probably does mean `2011 annual gross profits'.
Crucially, none of the three articles specifies whether these gross profits are those earned from Irish racing alone, from all racing worldwide, or from all sports (and non-sporting events).
Brian Graham next states that the average client pays five per cent commission to Betfair [on winning markets]. This is the MAXIMUM and not the AVERAGE commission rate; of course, some clients pay premium charges and other charges in addition to commission.
He then correctly deduces that if all clients paid commission at five per cent and there were no other charges contributing to Betfair's gross profits, then the clients paying that commission would have won aggregate commissionable profits of EUR214 million on their winning markets [in 2011]. He neglects to mention the near certainty that those same clients would have lost an identical amount on their losing markets (unless Irish punters are on average better or worse punters than Betfair clients living in other jurisdictions).
He next invents a ratio of ten per cent between aggregate commissionable profits on winning markets and a concept which he describes as turnover; he calls this ratio `profit margin,' and estimates turnover as EUR2.14 billion [in 2011].
He then compares this implied turnover to the turnover defined as the aggregate stakes taken by old-fashioned bookmakers who act as layers only and who are subject to a tax of one per cent on those aggregate stakes.
If old-fashioned bookmakers worked on the Betfair principle of betting to a 100% line and charging, say, 5% commission on net winnings (excluding backer stakes) and if each of their clients backed only one horse per race, then on a turnover of EUR2.14 billion, they would pay out aggregate gross returns (including backer stakes) of the same amount, EUR2.14 billion. For the bookmakers' gross profits from commission in this example to be only EUR10.7 million would require clients' winnings (net of stakes) to be only 10% of gross returns, or the average odds at which winning bets were struck to be 1/9. This might perhaps happen if Frankel ran in and won every race in the calendar, but it is clearly implausible.
As Brian Graham must have done, I have made some essential simplifying assumptions in my attempt to make any sense of his figures. However, I think my assumptions are much more plausible than the implications of his conclusions.
Brian Graham goes on to ask how much income tax is being paid by Betfair clients on the EUR214 million which he estimates that they earn on their winning markets. The answer is very little, for two perfectly good reasons.
Firstly and most basically, the same Irish Betfair clients in the aggregate must have lost roughly the same amount of EUR214 million on their losing markets, for the aggregate gross winnings of all exchange clients are zero by definition. Of course, within these aggregates, some clients win and others lose.
But, secondly, it has long been recognised that gambling winnings are not taxable, in the same way as gambling losses can not be offset against taxable earnings to reduce tax liability. Only licensed bookmakers who use Betfair as part of their business may pay some income tax on their aggregate annual Betfair winnings.
Perhaps you could invite Brian Graham to spell out the precise assumptions behind his calculations.
Alternatively, William Hill might invite him to appear as an expert witness in the present judicial review case and allow his assumptions to be subject to the probing analysis of the learned counsel representing the Levy Board.
55 Ard Coillte
Home: +353 61 374651 Mobile: +353 87 2547230