Supplementary memorandum from the British
Horseracing Board (DGB 163)
COMMENTS ON
THE BETFAIR
REPORT "THE
FUNDING QUESTION"
(NOVEMBER 2003)
Introduction
1. In November 2003, Betfair produced a
paper which purports to deflect the charge that the shortfall
in Levy income for the 41st Scheme can in any substantive way
be attributed to betting exchanges and their impact on bookmaker
margins. Instead, Betfair claims that the shortfall is created
by changes in punter behaviour towards short priced favourites
and migration to other products most notably fixed odds betting
terminals. The paper was widely but briefly reported on in the
major racing papers.
2. The paper is long on theory but short
on evidence. Critically, Betfair have no evidence to support their
claims of shifts in punter behaviour. Without this, the foundation
stone of their arguments, Betfair's attribution of the shortfall
to other causes has no basis in fact.
3. Ladbroke subsequently commissioned a
leading expert witness consultancy, LECG, to review the Betfair
report. It is a rebuttal of Betfair's arguments.
4. The LECG review largely reconfirms the
conclusions of an earlier analysis presented to the Board, "The
Impact of Betting Exchanges", which was based on bookmaker
data and which demonstrated that while gross margin had migrated
to other products, the primary cause of the shortfall was not
a shift in the weight of money towards short priced favourites
(although this was acknowledged as a secondary, minor factor)
but the inflation of starting prices brought about by the ability
of the on course market to trade into betting exchanges which
operate to far lower overrounds.
5. LECG have had access to the same data
and have come to the same conclusions as the BHB. It is understood
that the LECG report has been submitted to the Joint Scrutiny
Committee.
The Context and Relevance of the Betfair Report
6. Betfair claim to have produced the report
in response to the BHB Chairman's widely reported IBC speech which
referred to the negative impact of betting exchanges on the funding
of horseracing. For many, the paper appeared to have been well
thought through and therefore has had success from a PR point
of view.
7. The shortfall is not however directly
relevant to the execution of the charging policy which BHB now
believes is appropriate for betting exchanges and their users.
The relevant questions which have to be answered in relation to
betting exchanges, in so far as funding is concerned, are:
8. The question of the shortfall is secondary
to and dependent on the answers to these questions which form
the necessary guide to a distinct charging policy for betting
exchanges and their users and to which BHB maintains a clear focus.
9. This paper does not address the issue
of the charging mechanism which BHB proposes to adopt in relation
to betting exchanges.
Overview of the Betfair Report
10. In its executive summary, the Betfair
report makes five assertions. A more detailed review of these
is made in the Appendix.
(a)
"A flawed basis of calculation. Horseracing
has based its funding model on a study whose assumptions and calculations
are fundamentally flawed. It has been neither understood nor accepted
that the calculations by OCP report failed to factor in the effects
of changed punter behaviour that might be brought about by the
new basis of tax".
11. It is self evident that the full recycling
assumption, upon which estimates if levy yield were made, were
wrong. Betfair produce no evidence, however, on changed punter
behaviour but instead focus on questionable theories of presumed
rational punter behaviour to support their arguments, which are
then dressed as conclusions.
12. The evidence from Ladbroke, seen by
BHB and LECG, on weight of money and singles betting is that there
was no material change and that the principal cause of the greater
than expected margin percentage decline can only therefore be
attributable to margin as driven by changes in starting prices
or the expected win rate of horses. This answers the key question
not of where did the money go, but why did the greater than expected
decline in margin percentage happen in the first place.
13. Betfair admit that margins on non favourites
have shortened as a result of on course bookmakers' activities
on exchanges. As they do not have access to the data they do not
appreciate the material impact this has on overall margins. Bookmakers
make over 42 per cent of their gross win from non favourites.
Betfair dismiss this as if it were not material. It is very material.
14. BHB and LECG dispute Betfair's analysis
of starting prices on favourites. Actual margins on favourites
for Ladbroke between the 38th Levy Scheme (the base year for the
forecast of the 41st Levy Scheme yield) and the 41st Levy Scheme
declined by 5 per cent. BHB does however agree with Betfair when
they assert that small changes in SPs can make large changes to
margins.
15. The calculations made by OCP are not
fundamentally flawed. Betfair's are. This is referred to below.
(b)
"Punter behaviour has changed.
As a group, punters now bet a higher proportion of their stakes
on low-margin favourites and singles. Less sophisticated bettors
now bet some of their money on FOBTs while those who have stayed
with racing are the more sophisticated players who by definition
are lower margin as a group".
16. The only possible source of data to
support the first claim is bookmakers. Betfair has no evidence.
The evidence from Ladbroke seen by LECG and BHB is that there
was some early limited change in behaviour, there has been no
material shift in either favourites betting or singles.
17. Betfair also has no evidence for the
latter claim. While there is no doubt that some of the additional
money won by punters went on FOBTs, the evidence from the published
results of Ladbrokes and William Hill is that the overwhelming
bulk of money on FOBTs during the period of most significant growth
(late 2002/early 2003) is incremental.
18. LECG point out that margin decline did
not take place until some five months after the change in GPT
which was well signalled. This casts further doubt on Betfair's
claims. LECG see it as much more likely that the fall in margins
from February 2002 was caused by the increasing growth in the
use of exchanges by the on course market. LECG also find it noteworthy
that the upward blip in over-rounds in March and April 2002 happened
when it was decided for a short period of time that the on-course
market could not hedge into exchanges. When this decision was
reversed, overrounds went back down again. Betfair make no reference
to this.
(c)
"Turnover target was too low.
When underlying structural changes are considered, the turnover
increase of 45 per cent that racing has experienced represents
a dramatic and significant underperformance. A 95 per cent increase
is what was needed to hit targets".
19. The key flaw in Betfair's re-calculation
of the turnover needed to produce a static level of gross margin
pre and post GPT is the assumption that the cash margin derived,
not only in total , but from each category of horse ( favourite,
non favourite, outsider etc) should stay the same. If this were
the case, the mix of betting on favourites would increase by 26
per cent which, we have seen, has not happened.
20. It is quite simple to re-format Betfair's
calculations to demonstrate that a 45 per cent increase is indeed
a much more likely expected outcome of the change in tax. This
re-working is contained in the Appendix.
21. Given the higher than expected margin
drop, we do agree, however, that turnover would have had to increase
by over 60 per cent for gross win to have then remained static.
(d)
"Over-round per runner is an irrelevant metric.
Although often cited OPR is an irrelevant metric for the funding
of horseracing, because margins on horses are not consistent.
Gross punter loss is the only driver of horserace funding. The
theoretical over-round is by definition not the actual gross loss".
22. OPR is on its own is merely a signpost
to potential moves in margin but it is not irrelevant. While a
downward change can, as Betfair claim, be benign it is far more
likely to be evidence of negative margin pressure. Indeed, LECG
in their review find OPR a good "surrogate" for margin.
BHB's analysis linked inflation in SPs to changes in OPR and final
margin in a way which is consistent with what is seen in the market.
This analysis has been validated by Ladbroke.
23. Unless the pattern of winning horses
changes markedly, the primary driver of the level of gross win
is price ie the SP. If this inflates, then margins decline. If
SP inflation accounts for the bulk of a margin shortfall, shift
in weight of money or betting patterns can only account for the
balance. This is what has been seen in the market and what the
evidence points to.
24. Going forward there is no market event
which would lead BHB to believe that there is likely to be any
material change to weight of money or betting patterns.
25. Any further decline in OPR is therefore
very likely to reduce margin and yield.
26. As on course OPR continue to exceed
OPRs on exchanges there is further capacity for the on course
OPR to decline.
(e)
"Incremental benefits of exchanges have been
ignored. An understanding of a betting
exchange's business model clearly demonstrates that the product
it offers is generating incremental money both in horseracing
and tax revenue".
27. BHB agrees that part of the betting
exchange turnover and therefore the yield that it produces is
likely to be incremental. Betfair make no attempt to analyse the
negative impact of migration from other channels most notably
telephone and internet betting.
28. BHB agrees that if integrity issues
can be adequately addressed, betting exchanges and horseracing
are almost uniquely suited to each other. Horseracing forms the
bulk of an exchange's business as other sports are less suited
to an exchange and an exchange cannot diversify into for instance
FOBTs. The question is not whether the income betting exchanges
produce from racing is incremental but if it is the right amount,
charged on the right basis.
|