FX is changing | FT Alphaville

FX is changing

Harpal Sandhu, chief executive officer of a systems firm called  Integral Development, is trying to change the way forex markets function.

Specifically, he wants technology to even-out the jagged playing field, where conflicts of interest abound and where rules apply to some market players but not to others.

But before we get to that, a little about how the FX market has been developing over the years.

Before the web really got going, interbank price-discovery was dominated by two key OTC players: Reuters and EBS (now owned by Icap). The bulk of the market was in the grasp of a select inter-dealer banking community. And, because participants knew who they were dealing with, prices were tight and liquidity was solid.

In the last decade, however, a multitude of different and cheaper broker platforms have sprung up to offer competing OTC services to both the retail and institutional FX market place.

But while the forex universe got increasingly fragmented, the key interbank market stuck faithfully to the Reuters and EBS systems for inter-bank business. That is to say, not much changed. The best prices were still to be found there — and the  shadow FX market remained as dependent as ever on interbank prices.

On that note, an article penned in August by Mark Warms, general manager of FXall’s European operations, draws on some interesting points from a Euromoney poll earlier in the year.

As Warms notes:

As the Euromoney FX poll showed this year, the market share of the top five banks remains remarkably steady at around 61.5%  even though several of them were high-profile losers from the subprime crisis. It is also worth noting that in the past year, the combined share of the top 10 banks rose from 76.3% to 79.7%.

Such figures, he says, underscore how FX banks have continued to dominate FX  despite rising competition and credit problems of their own — something which happens to go against the expectations set out in the 2007 BIS Triennial central bank survey. As Warms observes:

At the time of the last BIS Triennial central bank survey in April 2007, much was made about the emergence of a new breed of market-maker. In reviewing the turnover, BIS stated: “By counterparty, the expansion in turnover in the interbank market was comparable to growth over the previous three years, but was outpaced by the increase recorded in the non-financial customer and non-reporting financial institution segments, which more than doubled in size.”

But BIS makes another key observation: “Consolidation of the banking system was identified in the past as reducing turnover in the interbank market through channels such as efficiency gains and the ability to net trades across related parties within an organisation.”

Interestingly, however, Warms says the reason not much changed was down to FX banks beginning to internalise their order flow — or, as the BIS puts it, net their “trades across related parties within an organisation.” An occurrence which happens to be much more prevalent in FX than any other market, and one that appears to be transforming banks into a new breed of FX market maker. As Warms explains:

Internalisation is a much-talked about concept in other markets; in FX it is a reality. The ability to match-up trades means banks are no longer so reliant on — and perhaps are even independent of — the main external platforms for reference pricing. Banks have rediscovered the true art of market making, albeit with a new, electronic twist.

This has resulted in ever decreasing volumes going through Reuters, EBS, and other alternative platforms, while making banks’ own prices hard to beat.

But, as Warms notes, banks will always have business they cannot match and which needs offsetting externally. And here’s the interesting thing.

When the EBS system suffered a major outage back in March 2009, it was the shadow market that successfully stepped in to provide the additional liquidity needed. This led some to question if the days of Reuters and EBS dominance might actually be numbered.
As Euromoney said at the time:

Until recently such a suggestion would have been summarily dismissed. The consensus was that although numerous viable trading platforms exist, ultimately they all depend on Reuters and EBS. The two electronic interdealer brokers have been routinely described as the market’s hubs, with every other platform effectively spoking off them. Pricing on every other platform was little more than a variation on the rates shown by the two brokers.

Further still, it turns out the banks were also annoyed by another development. Reuters and EBS had turned to non-bank participants to compensate for lost flows to internalisation — among them high-frequency traders.  As Warms explains, banks were not particularly comfortable with the additional credit risk presented, making them more inclined than ever to test alternative liquidity sources. As Warms observed:

In FX, the issue is not about moving block trades, but rather it is about transacting smaller packets as efficiently and profitably as possible. Banks want to be able to meet with counterparties of natural interest, which will invariably be other banks.  Older market participants will remember how interbank brokers were able to meet many of these needs of the banks; the fact that non-bank players could not access their prices is believed to be a major reason why. On the whole, the brokers were trusted. And trust remains a crucial part of FX, just as it was when a bank used to tell a broker he was on the bid or offer. Electronic trading has not replaced this. Now banks are not only analysing how they price risk and credit, but also how they price trust.

Which leads us back to Integral’s Harpal Sandhu and his efforts to streamline and integrate the shadow FX system into the wider interbank system.

As Sandhu points out prices offered in the non-interbank market have always depended on who you are, how big your trade is, how creditworthy your profile is and what liquidity is like at the time of trading. The model essentially does account for credit risk.

The problem is, it’s hard to tell how fair the prices being quoted are — a) because they’re set at the discretion of the service providers and b) because the market’s opacity and fragmented nature makes it almost impossible to cross compare, especially if you’re a retail customer tied to one provider.

Meanwhile, there’s also no way of telling when bid/ask spreads are a function of liquidity conditions and when they’re a function of market-making fee structures (via the process of internalisation) .

Accordingly, Sandhu’s solution  — given banks’ own foray into the shadow system — is  interconnecting the whole market and centralising the data. His technology firm is attempting to do just that:

We believe that a key success factor for any FX facility will be its ability to accept output from all these various trading venues, support the different trading styles, correctly read all the various pieces of information that are associated with it, and normalize it.

Integral has spent hundreds of millions of dollars and several years building FX Grid to do just that. Among other things, we added multiple price discovery aggregation mechanisms, credit line management, netting, STP, verification of execution with QOS, and most importantly monitoring services.

FX Grid as a provider of DMA managed services, monitors and reports on what is happening, what happened already and what is expected to happen in the near future for trade executions. We do all of that as on “on demand” business service. Add to all of this a set of user-controlled algorithms for blending, splitting and spreading liquidity for internal and external users. As an added benefit, aggregating liquidity, on demand, from all possible sources dramatically reduces the danger of a system outage.  Designing FX Grid as a multi-sided trading network was the right choice because it doesn’t favor a particular trading style or venue.

We don’t pick ‘the best’ venue or execution style and force our customers to follow suit. We leave the decision to what is ‘the best anything’ to the actual stakeholders or the participants of the trades, while supporting them all the way.

According to Sandhu it’s a win-win situation. Banks get to offset flow with credit risk accounted for and attract retail flow; small brokers achieve improved execution; and medium-sized intermediaries are forced to compete more effectively or  lose business.

Meanwhile, because Sandhu is confident all market participants will benefit from his system he’s providing real-time prices to joe-public for free.

Retail clients that way have a chance to compare spreads with a reliable indicator– a move that, he says, should only really be a threat to those taking advantage of customers anyway.

As he states:

I call TrueFX a disruptive but benevolent move because it is hard to imagine strong criticism from anyone other than brokers that (ab)use the current system by acting as a market maker and broker in unison. As such, they are not acting as a neutral agent that is providing customers with best execution. The way markets are organised today, there can be a strong conflict of interest because brokers can influence currency rates to their advantage and, if there is no transparency, most customers will never know that they were taken advantage of by unfair business practices.

Related links:
Anna’s story –
FT Alphaville
A Twin Cities FX scandal – SEC freezes Pat Kiley assets
– FT Alphaville
The $100bn FX hustle
– FT Alphaville