At this point it is widely known in the FX market that there
is an initiative underway, being spearheaded by FastMatch, to build a
consolidated tape for FX, similar to the one that exists in equities.
But given the differences in market structure between these
two asset classes, what does an FX tape look like and can an effective one ever
realistically be built?
The main driver for wanting to build a tape is to improve
the FX data available to market participants. The last Bank for International
Settlements (BIS) survey in 2013 put the average daily spot volume of the FX
market at $2 trillion, only a fraction of which is disclosed to the market.
This means that market participants are relying on data from
only a small subset of the overall trades that take place to build algos or
“There is a need for more transparency with order
placements, so is there need for a tape? Do clients really know what they’re
getting without a tape? There needs to be some disclosure against a benchmark
that isn’t WMR,” says a senior figure at one major bank in London.
At a high level, the tape currently being proposed would
function by having dealers report transactions into a central service and then
that service will accumulate the transactions from everyone reporting,
normalize the data, implement surveillance tools to ensure there are no
outliers in the data and then distribute those transactions to subscribers of
The plan is currently to create a new and independent
company to operate the tape. This company will be owned by the contributors to
the service but will have its own CEO, compliance officer, technology head and
Firms contributing trades to the tape will be able to choose
whether those trades are disseminated in real time or on a historical basis.
However, because the real time data will cost more for market participants to
access, and therefore is more valuable, contributors to the tape will earn more
credit for allowing trades to be distributed in real time than they will if they
elect to have it distributed on a historical basis.
Additionally, firms will get more credit the more trades
that they contribute to the tape, to create a financial incentive for larger
firms to start reporting. The way that this “credit” works is that subscribers
to the tape are charged a fee to access the data, one price for the real-time
data and a lower price for the historical data.
Dmitri Galinov, CEO of FastMatch, says that the actual cost
for producing the tape is relatively low and therefore there will be plenty of
excess revenue that will be distributed back to the firms that contribute
information to the tape. So firms that report more trades and in real time will
get more “credit” in the form of revenue from the service.
In addition, the founding members of the tape will have
equity in the company that is formed to operate it and the more flow that they
contribute, the more equity they will apparently earn in that company.
The main difficulty, says Galinov, is getting these founding
members in place.
“The biggest challenge is to create critical mass,” he says.
“It’s not a big technological challenge, all we’re talking about is reporting
trades so it would probably only take about three months to get this live and running
once we have the initial participants in place.
“It’s a chicken and egg problem. You need to have a
diversified set of participants to have a meaningful tape and so the way that
we’ve approached this is to divide the entire institutional FX market into four
broad categories: banks, non-banks, ECNs and brokers – both institutional and
retail for the brokers.
“What we’ve said is that for a tape to be successful, we
need to have at least three or four major players in each of these categories
contributing to it,” he comments.
Galinov predicts that if commitments can be secured from
this number of diverse market participants, then the tape could be launched
with somewhere in the region of $300-400 billion in average daily volume being
printed on it, which he describes as “an excellent start”.
There have been ongoing phone calls over the past few months
between various market participants involved in the initiative to try and agree
the rules of the tape. Now Galinov is trying to solicit verbal commitments from
some of these firms until the group reaches this target figure.
Currently, such commitments have apparently been secured
from four firms, one in each of the categories outlined by Galinov, while
MarketFactory has been lined up as a potential technology partner for the
Unsurprisingly at this stage, there are some rules regarding
how the tape will function that still need to be ironed out.
For example, large trades will have an exemption, meaning
that contributors can wait until the end of the day to report a large trade to
reduce market impact for their clients. But what size trade constitutes a
“large order”? And will this vary by currency pair? A $5 million trade in
EUR/USD might not be considered a large trade, but what about in much less
liquid currency pairs?
Another example is whether reporting certain currency pairs
have more value than others? Is the print of $1 million of voice in EUR/USD
worth the same as, say, $1 million of EUR/PLN?
These kind of questions, says Galinov, are ones for the
tape’s executive team and board to decide once the project is underway.
Another interesting facet of how the tape will function is
that by default, the tape will not reveal who contributed to it, but
contributors have the option to have the trades that they submit attributed to
“If you begin printing spot
to a tape, at least that’s a start. The value in doing so is obvious, but the
challenge in implementing it is very difficult. If it’s not going to be driven
by regulators then it needs to be industry driven and that’s very challenging,”
Instinctively, it seems hard to envision a major, large
contributing firm wanting to advertise to the market the actual trades that
they are reporting, but Galinov argues that the experience of the equity
markets suggests that firms might actually come to prefer this option in some
“In equities, we observed a phenomenon where firms that
reported transactions began to identify themselves and very quickly every
dealer started identifying themselves rather than preferring to remain
anonymous,” he says.
The logic behind this is that identifying themselves on the
tape enables these firms to advertise their liquidity to the entire market. The
contributors can choose to only identify themselves on certain trades so, for
example, if EBS BrokerTec wanted to advertise how much USD/CNH volume was going
through its platform, it could make this currency pair identifiable on the
tape. Likewise, if a dealer has a particular strength in a currency pair or
pairs, then it could effectively advertise this via the tape and potentially
attract more liquidity as a result.
Despite all this, the challenges in actually getting a
consolidated tape off the ground in FX are significant.
“If you begin printing spot to a tape, at least that’s a
start. The value in doing so is obvious, but the challenge in implementing it
is very difficult. If it’s not going to be driven by regulators then it needs
to be industry driven and that’s very challenging,” says the banker in London.
One of the first challenges facing this initiative is
getting buy-in from the major players in the market. Galinov talks about
needing to get three or four major players from each segment of the market on
board, but on the ECN side, the two biggest players that publicly report their
volumes – EBS BrokerTec and Thomson Reuters – each have their own market data
business that a tape could potentially undermine.
When looking at the multi-dealer space, it should be pointed
out that these two platforms are seeing substantially more volume than any of
the other platforms that publicly report volumes.
Why, when these firms have a successful and profitable
business selling market data feeds, would they support a consolidated data
Once again, Galinov points to the equity markets. In
equities, the exchanges sell their real time market data directly at one price
point and then created a central tape called the Central Information Processor
(CIP), which sells near real-time data from the exchanges at a much lower price
He claims that the number of firms that need real-time data
direct from the exchanges is smaller, but they are willing to pay more, and
that there is a much larger segment of market participants that would be
willing to pay less for data that is slightly delayed, perhaps by a few
So rather than cannibalising their existing market data
businesses, it will open up a new potential revenue source in addition, says
Galinov. Whether these firms will buy this argument remains to be seen.
On the bank side, one speaker at a recent Profit & Loss conference highlighted
one of the difficulties they’ve encountered when exploring the idea of a
consolidated FX tape.
“Conceptually the tape sounds like a good idea, but then we
go and talk to our clients and they say that it’s a good idea as long as you
don’t put our data on it. That’s going to be the difficult thing,” the speaker
said, adding that when a big buy side client puts through a large order, they
definitely don’t want that reported onto any tape.
“The NBBO drives price competition,” the non-bank source
says. “Because of the bilateral relationships in FX, yes, it’s not a
like-for-like comparison. But over time, the ability to see other prices that
existed will increase competitive pressures and force liquidity providers to
offer tighter prices.”
The response from Galinov is that there is no requirement
for banks to contribute 100% of their trades to the tape and therefore if a particular
client doesn’t want them to, then they don’t have to contribute their trades to
His argument as to why buy side firms would actually be
happy for their trades to appear on a tape is a little more vague and appears
to rest largely on anecdotal evidence.
“I’ve been talking to many buy side clients that are
completely supportive of the tape and the reason is because it would enhance
the TCA and algo offering from the banks that are presented to them.
“Also, there is a set of clients, who are large financial
institutions, that are very familiar with the tape, because they operate in
other asset classes where there is one and so they understand the value of it.
So there will be plenty of clients that will allow banks to report their
transactions to the service for transparency reasons,” he says.
With regards to the ability for a tape to improve TCA for
clients, the nature of FX as a market that predominantly trades in an OTC
bilateral fashion means that its impact to provide data for effective TCA is
fundamentally limited compared to equities.
Whereas in equities the products trade on exchanges and the
National Best Bid and Offer (NBBO) means that there is a definitive best price
for each transaction, in FX the price that gets offered on a bilateral basis is
dependent on numerous factors, including that particular counterparty’s credit
profile and the nature of their flows.
Therefore, the prices printed to this hypothetical tape are
simply not available to everyone in the market, making execution comparisons
Despite this, a case can be made for arguing that although
TCA is intrinsically more complicated in FX than in equities, there is value in
having more data available when attempting to measure execution quality.
“Let me ask you this: when you’re trying to produce an
accurate chart of something, would you rather use three points that you can
connect or would you rather have 100 points that you can connect? Yes, when you
have 100 points there will be more anomalous points in your chart, but overall
it will be more accurate,” says Galinov.
Essentially the point here is that, although some of the
price points on the tape are likely be outside the band available to certain
market participants, which is the case now, it’s ultimately better to have more
data than less when it comes to TCA.
“I’m really impressed by how
much the FX market has changed. If we had gone to the banks and the dealers two
years ago with this concept we would have got a very negative response. I’ve
been surprised that just about every bank that we’ve talked to about the tape
has responded very positively, because they want increased transparency for
their clients and themselves.”
In addition, one senior figure at a non-bank market maker
claims that having a price comparison – even one that isn’t prefect – could
help make FX pricing more competitive.
“The NBBO drives price competition,” the non-bank source says.
“Because of the bilateral relationships in FX, yes, it’s not a like-for-like
comparison. But over time, the ability to see other prices that existed will
increase competitive pressures and force liquidity providers to offer tighter
Ultimately, the biggest challenge facing the firms trying to
build a consolidated tape for FX is the chicken and egg problem highlighted by
“One of the difficulties is that no one wants to be first.
The deal is that if you give your data, you then get free access to the tape
while other firms have to buy that data. But there’s a lot of value in these
prints and so unless there’s mass buy-in, there is a real disincentive to make
this data available to the market,” says the non-bank market maker.
Another source at a major US bank comments: “Clients want
indices and verification on trade pricing so [a tape] is something that we’ve
talked about before, but one of the difficulties is finding the right potential
partner to run it.”
This last issue is an important one, more than one source
opined to Profit & Loss that,
although the tape is supposed to be operated by an independent body, it is
still viewed by those outside the group of firms that are trying to create the
tape, as one designed purely to profit their own businesses.
This is not in and of itself problematic, the tape would not
be a charity organisation, nor is it in the equity markets. Indeed, data is an
increasingly big business in financial markets these days. But it has led to
some concerns – that may or may not be justified – about potential conflicts of
interest and differing agendas from those participating in the tape.
Even if this chicken and egg problem is resolved, a
meaningful FX tape will require the involvement of the incumbents in the FX
market. With the top five FX banks accounting for more than 50% of the global
trading volume in 2013, getting at least some of these banks on board will be
One market source Profit
& Loss spoke to tried to set up a pan-European consolidated tape for
equities, but said that although the idea was popular it failed to get
traction, at least in part, because the stakeholders in the market resisted the
Whether a consolidated tape for FX will appear any time in
the near future is still unclear, but there has been a noticeable shift in the
dialogue from some of the major players in the market, including the big banks,
regarding this issue.
This could be a result of the settlements that the banks
have been forced into, greater regulatory scrutiny and a subsequent renewed
focus on compliance or pressure from clients for more transparency or, perhaps
more probably, a combination of all three.
Either way, this is not a change that has gone unnoticed by
“I’m really impressed by how much the FX market has
changed,” he says. “If we had gone to the banks and the dealers two years ago
with this concept we would have got a very negative response. I’ve been
surprised that just about every bank that we’ve talked to about the tape has
responded very positively, because they want increased transparency for their
clients and themselves.”