FX Survey 2012: Size isn't everything for FX banks' profitability
Every investment bank has spent huge investment dollars on FX since 2008. Now a shake out seems to be occurring. Banks with scale and budget are winning more share, but there are decent returns to be had for institutions of all sizes if they are focused.
Since 2008, foreign exchange has been a business to be in. That was also the year that industry-wide FX revenues peaked at a record high. Nonetheless, investment from banks to build flow-centric FX franchises continued apace for another two years. The prize? The most attractive return on equity of any global markets business on the street. Amid the backwash of the financial crisis, which perversely exaggerated FX revenues due to the extraordinary capital flow and counterparty risk that prevailed, it became clear that the days of liberal capital allocation were numbered. And as FX revenues skyrocketed, so did the return on capital. It didn’t take a genius to figure out that capital-lite businesses such as FX needed to be core, rather than ancillary, to the overall banking franchise, as they had been.
Suddenly, banks on the periphery of the Euromoney FX survey wanted to be in the top 10-ranked counterparties. Some set their sights still higher, to be top-three players. Before that, a small group that included Deutsche Bank, UBS, Citi and Barclays dominated the market. Between 2006 and 2008, the banks that ranked in the top five of the Euromoney survey held more than a 60% market share between them. Then came the financial crisis.
By the end of 2009, banks with FX ambitions had mobilized resources, a hiring spree ensued, technology budgets soared, and the incumbent’s market share was eroded. In Euromoney’s 2011 survey – which accounted for 2010 volume – the market share of the top five banks fell to 52.3%.
"I said to my staff at the time : ‘This is unsustainable’," says Jeff Feig, global head of G10 FX at Citi. "The returns of 2007 to 2009 couldn’t continue because not every participant could get the desired return on investment. We could see that a shake-out was required, and that’s happening now."
Indeed, times have changed, and while it is hard to quantify the level of investment that has been poured into the FX business, the revenue side suggests Feig’s conclusion is close to the mark. According to consultancy firm Coalition, annualized global FX revenues have declined 29% since 2008, to about $32 billion. For G10 currencies, revenues are down more than 50%.
The results of this year''s FX survey seem to suggest that appetite for the market-share fight is being tested. Market concentration is on the rise again, with the top five banks now holding 55% of market share. The largest FX flow houses are quick to jump on this as a crucial signal.
"It’s an inflection point," says Zar Amrolia, global head of FX at Deutsche Bank. "The industry is fundamentally going to change in the next three to five years. What we do now, this year, is going to determine the winners of the next five to 10 years."
Deutsche Bank, which topped this year’s survey for an eighth consecutive year, has become the unquestioned market leader in FX during the past decade with its astute combination of superior trading technology and market making. That competitive advantage allowed it to open up a big lead on its closest rivals in the latter part of the last decade.
However, it, too, has had to face the headwinds of competition. In 2008, amid the financial crisis, it commanded almost 22% of total market share. That has now been whittled away to 14.55%, and the gap between itself and the second spot is now 2.25%, compared with 6% in 2008. Amrolia thinks that gap will begin to widen again. As he sees it, the so-called "flow monster" business model is evolving.
It is being shaped by two main influences. Firstly, by regulatory changes – although not directly related to FX, with the exception of options and non-deliverable forwards – that will create a need for new services beyond just execution, such as processing and clearing that customers will demand across all products, including FX.
And secondly, by a new generation of electronic trading tools that delivers ever-more efficient pricing and which mitigates margin-eroding predatory client behaviour, while boosting their capacity to drive greater volumes over their platforms.
"The competitors, the services that we will offer, will fundamentally change driven by regulatory and technology changes," says Amrolia. "Anyone who isn’t thinking that way is fighting the last war."
War is perhaps the operative word here, more in the present than past. While competition has always been healthy, the level of investment required has brought a new intensity to the fight for market share. Both in the drive to capture more volume in an ever-increasing low-margin game and to be the first to solve the ever-present arbitraging by tech savvy high-frequency flow, which fits into the broader definition of latency.
Deutsche is upping the ante. Putting that in the context of this year’s survey – which showed aggregated volume of $208 trillion, with Deutsche’s share of that at about $30 trillion – Amrolia predicts that Deutsche will be trading volumes of $1 trillion a week by the end of 2013. It is a bold statement.
As Citi’s Feig tells Euromoney, his firm’s bid to overtake Deutsche has only just begun, after a huge effort in 2011, where its jump in volumes represents the second-largest swing in the survey’s 34-year history.
All of this costs money. And finding people to build the systems isn’t an easy task either. Engineers had to be lured away from Silicon Valley, exchanges and the equities markets to build the new generation of technology. Late last year Citi opened a technology research and development centre in Israel, joining Barclays, which had done the same earlier that year.
Coalition breaks FX revenues into three tiers to form its Coalition Index of 10 of the largest investment banks globally. These include Barclays, Citi, Credit Suisse, Deutsche Bank, Goldman Sachs, JPMorgan, Morgan Stanley, Royal Bank of Scotland, UBS and Bank of America Merrill Lynch.
Average annual revenues for top-tier firms were $2.6 billion to $3.4 billion between 2008 and 2011, second-tier revenues were $1.4 billion to $2.5 billion, and third-tier revenues amounted to $0.9 billion to $1.2 billion during that period, according to Coalition’s IndexPlus report. HSBC, which isn’t tracked by Coalition, reported $3.3 billion in revenues in 2011, excluding what it calls "pay away" to its other divisions for shared revenue.
Although Coalition doesn’t specify the banks in each tier, it is generally thought that Citi and Deutsche are the most profitable FX firms by revenue.
These higher revenues are a reflection of their revenue weightings with corporate clients. Citi, HSBC and Deutsche rank as the top three banks with non-financial corporates in this year’s survey. While representing the most dispersed client group among the index participants, corporates make up one-third of client revenues, while asset and wealth managers make up around one quarter.
Coalition’s analysis also shows that the contribution of FX as a proportion of total fixed income, currencies and commodities (FICC) income is rising. In 2009, it represented 16% of total FICC revenues; at the end of 2011, it had risen to 27%. This is instructive of the changing revenue mix within FICC, driven by regulatory change and capital charges faced by banks.
The business case for an increased weighting in FX remains compelling. In a report published in September, consultants McKinsey calculated that ROE might fall by 80% to 85% across structured credit and rates businesses, to potentially below the cost of equity. For FX, ROEs were only expected to fall by 40% to 45%, to about 16%, twice the return of a flow rates business, the report said.
Therefore, there’s a lot up for grabs. Coalition says that the leading 10 investment banks are now securing a greater share of the revenue wallet. In G10 currencies, for instance, those banks now account for 55% of the total revenue pool, up from about 49% in 2008. The middle market appears to be getting squeezed.
According to other leading FX banks, the real test of market concentration will be whether those emerging FX banks of the past few years have a willingness to sink further investment into the business to compete with the large flow players.
"Instant results are difficult to achieve in foreign exchange," says George Athanasopoulos, UBS’s global co-head of FX. "Coming in, you need to invest for three to five years before you see results on the scale that the top handful of banks have been [having]."
He adds that those that have invested heavily in technology to get where they are will have to invest again just to maintain their position. "Fixed costs are going to go higher and you need to have revenue momentum to support it," he says.
As a result, banks that sit outside the main flow businesses might need to assess the viability of their business models and work out whether they have the stomach for the fight for market share.
"Those outside the top 15 will probably need to drop their ambition to be a high-volume FX bank," says Chris Vogelgesang, who is global co-head of FX at UBS, alongside Athanasopoulos. "The top five can do this, but positions six to 15 will need to make up their minds. Are they going to invest, or are they going to retrench and refocus?"
Evidence of a reassessment is mixed. Nomura is one firm that appears to be tweaking its business model. Its FX business, which emerged out of the ashes of the Lehman Brothers collapse, set about becoming a global flow player. It had gathered momentum, too, climbing 43 places in the Euromoney rankings between 2009 and 2011. That momentum stalled this year, however, staying unchanged in 14th place.
Last year, its global head of FX, Richard Gladwin, left the firm, as did several other key FX staff. Was this the unravelling of Nomura’s ambitions to be a global FX flow player after a large-scale build-out of its global platform?
Not at all, says Nomura. After steadying the ship, it claims the business has refocused under the new leadership of Steve Ashley, the former head of rates at Royal Bank of Scotland and now global head of fixed income, which includes FX. While Nomura maintains it is still a global FX house, it now emphasizes its unique Asian flows as its edge.
"We’re focused on delivering a premium service in products where we have an edge," says Nigel Khakoo, Nomura’s co-head of EMEA FX trading and global head of options.
Although momentum might have slowed, FX revenues have been strong, say sources at the bank. Nomura doesn’t break out its FX revenues, but fixed-income revenues rose by 20% in 2011, with FX being a leading contributor.
Morgan Stanley, another bank to join the emerging group of flow players in 2008, appears to have made an impression after investing – as several market sources have suggested – $200 million to build its single-dealer platform Matrix. Stephen Glynn, the firm’s head of FX, tells Euromoney he intends to continue to forge ahead with Morgan Stanley’s ambition to become a leading flow-driven player. It climbed one spot to ninth in this year’s survey, leapfrogging its traditional rival Goldman Sachs.
Questions are now being asked about Goldman’s positioning in the FX markets. Despite its well-regarded electronic trading platform Redi Trader, its leading position in currency options has collapsed, from second place in 2011 to 10th this year. It remains unclear whether this marks just one bad year for the firm, or a waning commitment to the business. The firm declined to speak to Euromoney for this story. Few of its rivals are yet prepared to dismiss them as a spent force in FX.
Meanwhile, BNP Paribas, which has maintained its overall 11th position in the Euromoney survey for a second year, recently launched its single-dealer platform Cortex FX as it seeks to grow its franchise beyond its corporate-centric base. BNP’s global head of institutional sales Adrian Boehler maintains the bank is more than just a niche player, whether that be in corporates or in quantitatives and derivatives, which has long been the external perception of its core business model.
Boehler told Euromoney in February: "What we are not is niche, nor should we attempt to be niche, because we have the scale to have far greater ambitions. In practical terms, we can afford to have greater and more wholesale ambitions, as a result of our impressive footprint."
And so the jury is out.
|Global foreign exchange markets revenue pool|
|Source: Coalition Proprietary Investment Bank Revenue Pools|
At the other end of the spectrum, the regional or product niche model is now producing some healthy returns on equity. The same advancements in technology that have allowed the main flow players to capture market share is also enabling smaller niche banks to lock in healthy returns.
According to Justyn Trenner, principal of the advisory and analytics firm ClientKnowledge, the huge budgets that big banks dedicate to obtain the best technology can often leave the regional banks intimidated. However, it does not need to be that way, and he argues there is a growing plurality of vendors reducing the cost of key technologies, as well as a wide range of electronic liquidity venues making tight pricing accessible for all.
"There’s no doubt that to play at the top, highest frequency game there is a significant investment necessary and a significant commitment to ultra-low latent frequency infrastructure and complex analytics," he says. "But on the other side you’ve got the reducing marginal cost of what you might call ‘efficiency pay technology’ that will allow you to see best price and access a far a wider range of liquidity than you could see even two years ago."
From the work undertaken at regional banks by ClientKnowledge, that’s providing those banks with an uplift in revenue of between 10% and 20% of captured revenue.
Richard de Roos, head of FX at Standard Bank in Johannesburg, says the smart use of technology and a clear sense of what their business model is means smaller regional players can carve out profitable and effective franchises without having to make the big investments of a main flow bank.
De Roos says: "We’ve had to do the obvious things: we had to put efficiencies into our systems to optimize spread; we’ve had to build stronger relationships with the clients domestically; we had to globalize into the international market to the extent that we can; and we need to maximize segments that we can within our ability to reach those markets as a niche player."
In terms of optimizing spread, Standard Bank is implementing a liquidity aggregator for G10 currencies that allows it to access and trade away risk, while also building an algo that prices off the depth of liquidity available on the aggregator.
"It allows us to capture more revenue, from say a one-point spread down to half a point – if you can get away with half a point to wash through that non-core business, then you can make some real efficiency gains," says de Roos.
This then allows Standard Bank to start pursuing bilateral arrangements with bank counterparties to secure their smaller non-core types of transactions. For example, they can go to a large flow bank and offer them direct access through their aggregator by reciprocation to their own core liquidity in rand and other African currencies. In other words, the relationship becomes symbiotic between the two business models.
"So, you’ve still got that niche type of relationship that gives what you’ve got in the region, while at the same time you’re able to use it to create a bilateral type of relationship," says de Roos.
Furthermore, a niche bank such as Standard Bank can also broaden its distribution even with small regional banks in Europe and the US on a unilateral basis, where they might regard African currencies as too small to manage themselves and where it’s easier to provide direct liquidity, so there’s no human intervention.
For its rand business, Standard Bank operates a variable netting algorithm, which allows it to lap up the smaller rand business and the smaller G10 business to a netting algo, and secures most of the flows that are coming through in small size, while at the same time providing useful analysis of data coming through to extract alpha.
Niche players, such as Standard Bank, face their own challenges too. Africa is the fastest growing region by volume in this year’s survey, at 65%, to just less than $1 trillion. Standard is the leader among the local banks and ranked seventh overall with 3.39% market share.
However, as interest in frontier currencies accelerates, will the likes of Standard Bank be able to defend their niches from the large global players?
"Competition from them is not new," says de Roos. "We’ve had it for 15 years and we’ve had to up our game." Where he sees Standard Bank’s strength is in the ability and familiarity to do business within Africa. "There is a lot of trade and investment going on within the continent itself, which we obviously have the ability to marry as well," he adds.
ClientKnowledge’s Trenner says that because niche banks can now obtain product very efficiently for their core customers, whatever the product is that they want, so they can efficiently price any instrument. That, in turn, makes the barrier to entry for an international bank, which is trying to break into a particular country, go up.
"Because the relationship value of the domestic bank remains and the ability to price remains, we’ll see a much more plural market, leaving aside the high-frequency space," he says.
However, while niche regional players might be able to maintain client loyalty, they have their limitations in comparison with the global emerging markets banks, such as HSBC, in capitalising on such dynamics as the South-South link.
Fred Boillereau, global head of FX and commodities at HSBC, says: "FX is about cross-border trade, one country trading with another. So, for instance, when you have to deal with a Chinese yuan versus Brazilian real cross-border trade, you need to stretch your legs to be in those two countries, and there are no local banks that have that capability."
Just how banks, and in particular, the flow monsters position themselves in the emerging markets will be something that plays out during the next five-years. Establishing an emerging markets franchise has been more about people and local knowledge, and less about being a flow monster, says Boillereau.
"Being a leader in emerging markets requires far more than just a price – it requires local knowledge, value-added research, close coordination between local and global networks and being able to manage liquidity/warehouse risk," he says.
However, the emerging markets model is becoming more crowded as the mix in revenues between the developed markets and the emerging world changes. Emerging markets FX now represents 50% of overall revenues, up from 28% in 2008.
This trend is also leading to a blending of business models too. Last year, Credit Suisse merged its FX division with its emerging markets fixed income division to form the global currencies and emerging markets group. The idea is to build on its existing niche in options and quantitative products within the fastest-growing regions.
As part of the merger, the bank has re-organized all of its trading centres and built e-solutions teams to offer clients a point of contact for all the bank’s products, such as its option platform Merlin, and quantitative tools, such as Locus.
"We definitely see ourselves as being a top player in FX," says Todd Sandoz, Credit Suisse’s global head of FX. "The only way to do that five years from now is to be a top player in EM FX."
However, as the increased competition and the emergence of EM-centric business models looks to proliferate, Deutsche’s Amrolia argues that the FX business model, as result of market structure regulation, will create revenue streams that are still underestimated.
"In the new world, we’re no longer just focused on execution and pricing," says Amrolia. "Capital, margin, total cost of trade, trade reporting, all of which were not previously an issue, are now going to be a really big issue."
The business model was split between the revenue-generating front end and the cost-accumulating back end, but the back end of the model will become a revenue generator in itself.
For instance, Deutsche has now combined dbOverlay, dbSelect, FXPB, OTC FX clearing, and listed products into an overall offering that offers clients cross-product margining and optimized collateral management.
Amrolia says all of these services can be viewed as a revenue model because clients will be willing to pay for these value-added services, such as capital optimization. "There are whole new revenue streams that are going to open up to this industry," he says.