As electronic trading went mainstream, it created an explosion of growth in the market. But has this growth run out of steam? Galen Stops takes a look.
The Big Bang Theory – in addition to being a popular American sitcom – is a well-known scientific thesis that posits the universe started with a small singularity before rapidly expanding, and indeed is still expanding to this day.
Although the FX market had obviously been around long before the advent of electronic trading, it was the introduction of this new technology that likewise caused a dramatic expansion of this market, the impact of which, much like the Big Bang, is still being felt now.
Just consider the numbers: in the Bank for International Settlements’ (BIS) 2001 triennial OTC FX turnover survey – the first one following the launch of Profit & Loss in 1999 – the size of this market was put at $1.2 trillion, whilst the latest survey in 2016 puts it at $5.1 trillion and the survey results due later this year are expected to inflate this figure further. As it stands, that’s a 325% increase in the size of the market over a 15-year period, a remarkable increase by any standard.
Although there were numerous factors that caused this growth, such as the rise of prime brokerage (PB) and the launch of CLS, all of these were underscored by technology, and specifically by e-trading.
“If you look back over the last 20 years, it’s clear that technology changed everything, it just underscores all the major developments over this entire period of time,” says Stephen Flanagan, executive director, global FX e-commerce risk manager at JP Morgan. “If you look at technology developments around PB and the API, you can see how this gave so many participants the opportunity to get into the FX market which, prior to this last 20-year period, was only really open to banks. It’s also driven so much more transparency on pricing, which has led to better choice for clients.”
Whereas electronic trading used to be bank-to-bank and was predominantly just for spot trading, it has evolved over the last 20 years, giving buy side firms access to a wider range of liquidity sources and a greater breadth of products. On the sell side, it allowed firms to reach a much broader universe of market participants and become more efficient, meaning that the same number of staff could handle a much wider client base.
In addition, electronic trading allowed the FX market itself to become globalised very quickly.
“At the beginning of the millennium, markets tended to trade locally,” says Tod Van Name, head of FX electronic trading at Bloomberg in New York. “But the advent of technology meant that all of a sudden you could trade any asset, anywhere, at any time and in any jurisdiction. So portfolio managers or hedge funds were able to change their asset allocations or exposures very quickly, and the more they were able to do that, the more they needed to hedge the currency risk that would occur as a result of this. That’s why you’ve seen this continual growth in the markets – because FX has moved from being primarily trade weighted, where it was driven by physical trade volume, to capital market weighted, where it’s driven by shifting assets overseas. This really fuelled the growth in the marketplace.”
Dwindling Number of LPs
While volumes exploded and a much broader swathe of buy side firms were able to access the wholesale FX market following this technological Big Bang, however, on the sell side the number of liquidity providers in the market has continued to shrink.
“There’s been a tremendous amount of consolidation in the banking sector; there aren’t as many banks as there once were and there aren’t as many market makers as there once were,” says the head of a trading platform in the US. Although high frequency trading (HFT) firms, which subsequently were able to rather successfully rebrand themselves as “non-bank liquidity providers”, did come into the FX market in a big way after the financial crisis, they still have not countered this trend of a shrinking number of firms providing liquidity into the wholesale FX market. It’s worth pointing out though, that this trend hasn’t necessarily been a bad thing for buy side firms.
“The number of counterparties that we have has come down dramatically in the past 20 years,” says Mike Harris, president of Campbell and Company. “Back then, we probably had 40 banks on our board and you were pinging them to try and figure out who was good in what pairs in what time zone. And you had to do that on a regular basis in order to have a good idea of who the best person to call was but, at the end of the day, you didn’t know how good their quote was relative to the marketplace because you didn’t even have dealable quotes – the quote you were looking at was in large part an indicative quote on Bloomberg or Reuters, it was stale.”
He adds: “The science of dealing and trying to get best execution in a very non-transparent world was very challenging, to say the least, compared to today.”
The concentration of liquidity providers has occurred in part because the up-front fixed costs associated with market making in FX are very high and it is challenging to build the critical scale necessary to be effective with pricing and distribution in this market. The question is: have we reached an equilibrium or is there more consolidation to come in this area?
While people generally appear reluctant to make any concrete predictions in answer to this question, which frankly is fair enough, there does seem to be a suggestion from a number of sources that the number of liquidity providers in FX is set to reduce further still.
Gio Pillitteri, head of e-FX market making at HC Technologies. says: “I think the answer is somewhere in between super concentration and complete dilution, but the key differentiator will be which firms are willing to invest in technology and which ones will be forced to rent technology from the top players because they either don’t have the appetite or the budget to invest themselves.”
The platform head in the US, meanwhile, sees the possibility of banks beginning to view the non-bank liquidity providers as potential acquisition targets, observing, “There could be some consolidation, I think you could see banks buying non-banks and I think you’ll see some movement of teams within the non-banks.”
One area of the FX industry where the expansion of trading volumes over the past 20 years has been mirrored is in the platform space.
In the ‘90s it was really just EBS and Reuters that offered wholesale FX trading, and even then only the sell side banks were able to access them. This began to change as the banks started launching their own FX trading platforms.
“The early iterations of single bank portals were about distribution, there was no trade or risk management tools – banks would literally get a position from a client that had just traded on the platform and they didn’t even know that the price was out to them, a position would just drop into the blotter!” says the US platform head.
Because their pricing wasn’t in competition with other banks on these platforms and the FX market itself was still very opaque, the banks were able to build in some cushion around their pricing in order to protect themselves. Hence, market participants indicate that the launch of these single bank portals did little to nothing in terms of improving the price at which buy side firms could execute their FX transactions, but did improve the efficiency of their trading.
Then, however, platforms like Currenex, Atriax, FXall, 360T, Hotspot, Gain Capital and later Lava FX came to market – and State Street’s FX Connect opened up to other banks – putting the banks in competition with one another, and this did lead to a significant compression of spreads.
According to the US platform head, it also led to a critical shift in the mentality of the banks. “Although some banks like UBS really embraced electronic trading, for most banks it was a project on the side, there wasn’t a strategy around e-trading. It was when the multibank platforms came out that I think everyone realised that this was going to become the future and it wasn’t just a project anymore,” the source says.
Of course, in the early days, the proliferation of multibank platforms did present some challenges for buy side firms wanting to trade on them.
“We had all of their GUIs. I remember at one point we had a trader who had four different monitors with each of the different platforms on them and he would sit there trying to get his mouse onto the best bid and offer which, as you can imagine, was very challenging in a fast moving market,” recalls Harris.
Aggregation helped solve this problem, while API trading and the growth of FXPB led to a further expansion of the number of market participants able to access these platforms, which in turn led to a raft of new multibank platforms being introduced to the market – such as GTX, LMAX Exchange, FXSpotStream and FastMatch – between 2010 and 2012.
Since then, the trend has seemingly been towards consolidation with Cboe acquiring Hotspot, Deutsche Boerse Group buying 360T and then GTX (an offshoot of Gain), Euronext getting FastMatch, CME buying EBS as part of NEX Group and FXall bought first by Thomson Reuters (now Refinitiv) which was in turn bought out by Blackstone and now set to become part of the London Stock Exchange Group (LSEG).
Moving the Chess Pieces
“Consolidation” is perhaps not the correct word here though, as Jill Sigelbaum, head of FXall, points out. “We’re not seeing consolidation of the platforms themselves, we’re seeing the consolidation of the companies that own the platforms, which is very different. These purchases have more to do with the value of the companies and the synergies across platforms, and it’s worth noting that the companies that have acquired these platforms are still running them independently,” she says.
Sigelbaum continues: “For example, here at Refinitiv we offer multiple liquidity sources and trading protocols, from trading on FXall, which supports multidealer RFQ, streaming spot prices, as well as a secondary market central limit order book (CLOB), to Matching, our primary CLOB. While as a company we strive to maintain a seamless and consistent workflow across all of our venues, as well as ensure a single point-of-access through our FXY desktop, our venues tend to have distinct user communities and to operate independently.”
Alan Schwarz, CEO of FXSpotStream, agrees that there has been commercial consolidation amongst the OTC FX trading venues, but predicts that exchanges buying into this space means that, in the long-term, there will be some structural change in the OTC platform space.
“We have seen consolidation amongst OTC venues but there hasn’t been any structural change in the FX market yet because of the M&A activity. All that’s happened thus far is that the chess pieces on the board have been moved around. I think that structural change is coming, however, and one of the more interesting things to watch in the next five to 10 years will be: what is the result of the consolidation today? Change is not necessarily bad, but the FX market works well today and is efficient, so participants need to ensure their voices are heard as we move forward. Lesson 101 of acquiring a new business is that when you’ve paid a lot of money for an asset, don’t start immediately changing everything, however, these exchanges are buying these businesses to enter the asset class and to drive efficiencies and at some point this will mean making changes,” he says.
There are also logical arguments to support the idea that there will be a thinning out of FX platforms in the coming years. For example, some suggest that the launch of so many platforms has in some cases divided up liquidity rather than actually increased it.
“There have been a lot of electronic trading platforms introduced over the years and I think some of them source liquidity from the same places and as a result it can get chopped up. I do think over time though that there will be some consolidation and there will be fewer venues to trade on,” says Joe Hoffman, CEO, currency management, at Mesirow Financial.
Meanwhile, Van Name questions whether having so many platforms is scalable for buy side firms. “The FX market has no shortage of firms that are looking to provide some kind of nuanced solution and there’s a lot of great technology out there but at some point, as a consumer of technology, how many different vendors can you actually interact with? It doesn’t become scalable, so it’s logical that the market will look to combine those rather bespoke solutions into something holistic that makes more sense in terms of what the ultimate delivery is to the client,” he says.
However, people have been predicting consolidation amongst OTC FX trading venues for years (one only has to go back and review any of Colin Lambert’s start-of-year predictions for evidence of this) and – in large part – they have been wrong. Indeed, James Sinclair, executive chairman of MarketFactory, argues that the trend will be towards more fragmentation and specialisation, even amongst trading venues.
“As the costs of technology have fallen, the barriers to entry have fallen and it’s become easier to launch new venues. In addition, people are now much more accustomed to trading across multiple venues. So I think that there will be more and more venues as the market becomes unbundled, networked and, in some cases, peer-to-peer,” he says.
Sinclair also points out that the continuing growth of active participants in the FX market, against a backdrop of a decade of economic expansion, means that there is still room for new venues to come in and service the needs of these participants. A good example to back-up Sinclair’s assertion is the planned launch by Dmitri Galinov, who founded FastMatch, of a new OTC platform this year that will initially be focused on NDF trading. Although, obviously, it’s much too early to tell if this new venture will have staying power.
In his analysis of the platform space, Campbell’s Harris seems to strike a balance between these opposing perspectives. “Competition is always a good thing, but you get to a point where you have 30 or 40 venues and there’s an opportunity cost associated with just doing all the work to certify them. How many venues is the optimal number to encourage competition while still being efficient? That’s a question for a Harvard MBA case study. What I would say is that I’m a big believer in disruptive technology and so if someone new comes into the market either forcing everyone to price lower or providing some different technological feature, we’ll look at that. Coming out with a copy of something that the market already has doesn’t do any good, but there’s a lot of different ways to come at the ECN model,” he says.
Faith in Technology
Amidst all of this, there is a broader question hanging over the FX industry which is: has the expansion of this market, which was powered by that technological “Big Bang” 20 years ago, run its course? After all, hidden behind the headline figure of dramatic growth in the BIS numbers is the fact that the 2016 survey represented the first time that the overall size of the FX market has decreased this side of the millennium. If one accepts the arguments that the number of liquidity providers and the number of platforms in the market are set to decrease, then could this all be an indication that, far from expanding, the FX universe is actually now contracting?
For Van Name, market conditions are the reason the growth of FX flows seems to have slowed or stopped lately. He points out that financial markets have gone from a zero interest rate environment to one where central banks were raising rates and economic growth was taking off. Now we are facing a potential recessionary environment, and central banks have started relaxing interest rate policies in order to manage that shift. Moreover, the central banks have not been doing this in tandem, which created a lot of volatility in Treasuries and the interest rate market. Yet in FX, which used to follow those markets very closely, volatility levels have remained relatively flat.
“The reason for this is asset managers debating how they want to deploy their assets and where they want to make their investments. So it’s a really unusual environment,” says Van Name.
He adds: “I think that economic factors are impacting FX flows. Will flows keep increasing at the same rate they have over the last 10 or 15 years? Probably not, but I think that there are economic drivers that might change that. For example, if the Chinese economy takes off again then you’ll see a dramatic pickup in currency trading.”
For his part, Flanagan has faith that – just as technology fuelled the increasing size of the FX market over the past 20 years – so too will it drive the next stage of growth for the industry.
“I wouldn’t disagree that we appear to have maybe hit a little bit of a stall versus where we were a few years back and I think that this underscores the cost to enter into this market as a liquidity provider now, which is very high,” says Flanagan. “But I think that this is just a stall before the next surge in technology enabled growth.”
He continues: “Any time new technology comes in, it’s disruptive at first to the human being, but then as we adapt to it and incorporate it into our businesses it leads to more growth. When I look ahead, I see that data analytics – whether in the form of automated market colour out to clients or in the form of algos and artificial intelligence tools – are giving new breadth to order execution. I think that as these technologies develop and people become more comfortable with them, it will open up a whole new area of growth. People refer to it now as “data-as-a-service”, we’ve really only just begun to enter this business.”
Buy Side Becoming More Active
Such optimism could prove well placed, but it might not even be tools as sophisticated as AI and algo trading that prove to be the next big drivers of FX market growth. Multiple sources note that while spot FX is highly electronic there is still a lot more room for NDFs, and especially FX options and swaps, to shift in this direction.
“I think that the level of sophistication in the FX market will continually increase – the bar constantly rises. We now have a very mature spot market, but forwards are much less mature. Traders are still uploading curves manually and then they get published electronically. There are good strides being made on the options side but there’s still areas of this market that are not electronic whatsoever. If you look historically at the FX market, there’s still room to maneuver and expand in the areas of NDFs, forwards and options,” says Pillitteri.
The main impediment to this happening thus far, some of them contend, is an unwillingness from the banks to see the margins that they still make on these products eaten into as, much like it did in the spot market, increased electronic trading will lead to tighter spreads.
“Even in NDFs the banks were reluctant to go electronic because their traders didn’t want to give up the high margins. Still to this day, the banks don’t want Tullett Prebon, Icap, Tradition and BGC to quote to the buy side,” says one senior industry source. “Even though they’re legally supposed to, the IDBs will still not do it because they’re scared that the banks are going to cut them off, so it’s not a very efficient market. Once it goes fully electronic it’s a level playing field, but people will resist that because it’s all about hoarding as much margin as possible.”
Others see a much less nefarious reason why these products have lagged so far behind spot in terms of electronic trading. “It’s not by design, but it is by desire and there’s a difference,” says Schwarz. “Markets become electronic when there is a point at which the market believes it makes sense for it to do so. If someone today wanted to make every single product electronic, they could. It might not be cheap, it might not be quick, but they could do it, but the market doesn’t want to or need to at this particular point in time and there are many considerations that need to be taken into account when taking a product electronic.”
It may just be that regulations and an increased desire for transparency will provide market participants with the incentive that they need in this regard though. The increased demand amongst buy side firms for transaction cost analysis (TCA) tools in order to demonstrate best execution to investors and regulators has been widely documented. To effectively do this, firms need good data to show where the market was at and explain how trading decisions were made, and, as previously discussed, electronic trading will naturally lead to improved data around these products.
Of course, there’s no guarantees. The experience of Swap Execution Facilities (SEFs) highlights that sometimes even regulation will struggle to push market participants into an electronic trading environment before they are ready.
Yet another reason for optimism around FX market growth, however, is less around technology and more around the changing dynamics of the firms participating in this market.
Sinclair observes that “the world seems to be a little more passive now” and that this has impacted liquidity, particularly in Asia. He speculates that one reason for this might be that market data feeds mean that firms no longer have to trade in order to find out where the FX market is or assess the depth of liquidity available and, in the case of Asian markets, it might also be a side effect of all the Japanese bank mergers that have occurred in the past 20 years. However, despite this, Sinclair sees forces in the market that could counteract this change in liquidity.
“We’re seeing a long-term trend where more and more firms from the buy side are coming into the FX market and there are lots of previously passive hedgers who are becoming more active. This could help offset this other liquidity that is becoming more passive because someone like an Australian superannuation fund can’t just invest in Australian equities because so much of that market’s correlated with mining and is pretty limited. As a result, they have to invest in overseas securities and so you see FX flows coming from them as a result,” he says.
Sinclair adds: “You see the same thing in the Nordic region and ironically it’s in these smaller economies where equity managers invest cross-border where you start to see interesting flows coming from. So the market continues to expand and there are more and more participants coming in.”
Ultimately, despite the confluence of factors that have constrained the growth of the FX market in recent years, it seems impossible to imagine that it won’t continue to expand going forward. With the caveat that progress in any field is not always linear, the sophistication, availability and understanding of technology in FX only continues to increase, as does the market’s dependency on it. So while, yes, that initial “Big Bang” of electronification that occurred around the time that Profit & Loss was launched is unlikely to be replicated any time soon, the industry can be confident that its effects are still being felt now as this market continues to expand ever outwards.