It’s been a busy 20 years since Profit & Loss launched. Colin Lambert and Galen Stops have picked out 20 key events or trends during that time and asked senior industry figures for their perspective on them – here’s numbers 20-16.
20. A New Currency for a New Era
While the big event of 1999 was undoubtedly the launch of Profit & Loss, the year also marked the launch of Europe’s much vaunted single currency, the euro. Gone were the multitude of European crosses for traders, instead the world was looking at what could easily become its second reserve currency after the dollar.
“The launch of the euro back in 1999 was a watershed event,” says Adrian Lee, president and CIO of Adrian Lee & Partners in London. “Leading up to it, no one ever believed that it would work, and some still don’t believe it has.”
A recent poll indicated that around three-quarters of Europeans like the euro, however there are still many voices of dissent. Even 20 years on there are rumblings from Italy suggesting if not completely abandoning the euro, then creating a parallel currency, so economic problems persist.
There is also the exchange rate itself and the psychological impact of its movements. “When the euro came in there was a big depreciation, it went more than 20% to 80 cents after starting at 1.17 and I think the reason was that people just didn’t believe in the concept,” suggests a senior figure at one large asset manager. “Then at some point, I think Asian central banks in particular bought into the concept because they saw a huge opportunity to diversify their growing reserves at 86 cents or 90 cents. There was a huge appreciation followed by a pullback which has in turn led to the current period of stability.”
The stability isn’t welcomed by everyone, for as one voice trader at a bank in Asia says, it had reduced limited opportunities even further. “I hated the thought of the euro as a trader when it was introduced, we lost so many opportunities in the crosses. Twenty years on I still hate it, we had a big move down, a big move up and then a correction and the net effect has been something like five or six per cent down over two decades. I don’t even bother trying to trade or position EUR/USD anymore, it bores me.”
So, love it or hate it, you can’t ignore the euro. It may not have delivered a strong stable economy across the European Union, but it has endured, and as one analyst argued earlier this year, the euro has been an enduring success, but a fundamental failure.
19. An Old Fashioned Scandal
Bogus options contracts, a prime brokerage account, and off market trades – could be the plot to a TV series, but they were actually behind the first FX scandal of the 21st century and the P&L era when Allied Irish Banks’ US subsidiary AllFirst Financial revealed a rogue FX trader, John Rusnak, had run up, and hidden, $750 million in losses.
Aside from the fundamental actions of the trader concerned, which were nothing new, the episode highlighted the fragile nature of post-trade systems in the banking world – the fintech revolution was not yet in full swing. An unusual twist to the story was that Rusnak himself had approached a risk management vendor two years before his activities were uncovered with a view to upgrading the bank’s systems because it was “insufficient for its needs”.
Another plot twist then emerged when AllFirst claimed that the losses had been built up over five years, leading one market participant to observe, “So he approached a vendor three years into what he must have known was fraudulent activity? How does that work?”
“So, how did it all turn out? Well Rusnak served a jail sentence, Allied Irish offloaded AllFirst the year after and then, 13 years on it finally settled a lawsuit it brought against Citi”
There is no doubt that better risk systems would have helped uncover the activity, which – and this may seem incredible to younger market participants used to USD/JPY being stuck in a very small range – occurred in that currency pair, because Rusnak created false trades, false confirmations from fake third parties and when he did actually trade, he sold far into-the-money options to raise premium.
So, how did it all turn out? Well Rusnak served a jail sentence, Allied Irish offloaded AllFirst the year after and then, 13 years on it finally settled a lawsuit it brought against Citi (which had a countersuit going) arguing that the bank, as one of Rusnak’s prime brokers, had aided his activities.
One final, and cruel, twist to the story. Sadly, or luckily perhaps, we cannot discover the trade publication concerned, but just two years before the activity was discovered, AIB was awarded ‘Trading and Risk Systems IT Manager of the Year’.
18. Post-Trade Evolution
This is not so much an event but more, as the header says, an evolution; however, there is no doubt that the post-trade space has empowered the FX industry’s growth. Probably the headline maker is prime brokerage, which marked the first breach in what seemed the impenetrable wall of bank dominance of the FX market.
“The evolution of credit has been a huge thing in the FX markets,” says a principal at a non-bank market maker. “We could not have achieved what we have without it, especially once it was realised that pushing FX onto an exchange model was a pipe dream.”
The growth in prime brokerage could not have been achieved without the advances in technology seen over the past two decades. “When I first started in PB people didn’t really talk about the post-trade, it wasn’t a real focus, but that changed in the past five-to-seven years, largely driven by regulation,” says Kate Lowe, global head of trade services at State Street. “Before that change, the business was very manual and it was fraught with operational risk – trades were confirmed by emails, and there were so many of them flying around, it was incredibly hard to track.
“I know we can be quick as an industry to criticise the messaging infrastructure in FX, but it has served its purpose. Looking back now, people just don’t realise how bad it was”
“Automation has reduced that risk tremendously, I reckon that 15 years ago, a PB business would be facing about 75% operational risk, 20% settlement risk and 5% market risk, but there has been a massive reduction in operational risk – thanks to something as simple as the automation of messaging and matching,” she continues. “I know we can be quick as an industry to criticise the messaging infrastructure in FX, but it has served its purpose. Looking back now, people just don’t realise how bad it was. Now we have 99% or higher STP rates and it’s very rare for anyone to touch the trade – it wasn’t always like that – it used to be a huge challenge.”
That automation has also spread beyond the realms of prime brokerage of course. “Look at what UBS has done in the post-trade space,” a senior banker, who works for a rival organisation, says. “They have automated just about everything that can be – our clients tell us that even their monitoring is now automated and that we, and every other bank, lag behind them. It’s a tough thing to hear from clients, but it should be an ambition for all banks. Yes there is a start-up cost, but that will quickly be recouped. It’s hard to express exactly how expensive it is to run a manual operation.”
So while technology has undoubtedly changed the trading process, perhaps the biggest change, especially viewed through the prism of the large trading losses that were run up, largely undiscovered, in the early years of the century, has been in the post-trade space. Here, technology has undoubtedly made markets safer for participants, not just because rogue activities can be easily spotted, but more because the number of genuine errors has collapsed, and even those that do occur are quickly discovered and rectified.
17. The Growth of Retail
This really is a simple numbers matter; in 2001 estimates were that retail FX traders were responsible for about $12 billion per day in turnover – in 2016, according to the Bank for International Settlements’ Triennial Central Bank Survey, that number was $282 billion. So yes, the FX market generally has seen good growth over that time, something like 3.5 times, but retail FX volume has gone up more than 23 times its 2001 level.
It has not been a smooth ride, however, and the 2016 data actually reflects a small slowdown from the $313 billion it hit in 2010. There has also been a host of regulatory issues that continue to dog the industry. Estimates suggest that 75% of retail clients lose money when trading FX, but still some brokers on the edge of respectability offer 400:1 or even 500:1 leverage to these clients.
The role played by technology cannot be understated, especially in the years 2005-10 when the growth was at its strongest, suddenly, the man or woman on the street was able to access tight spreads and seemingly endless liquidity. Probably the biggest driver in that growth, however, has been Japan.
When that nation entered its zero or negative rate policy in the early years of the century, Japanese investors turned offshore for their yield and the already popular carry trade became even more in vogue.
Yes, there have been incidents in the retail space when the FX market has dislocated and yes, regulators have more work to do to deliver a coherent and global policy towards leverage, but probably the number one event indicating that retail FX has come of age and is influential was on January 4, 2019.
On that day, the FX market witnessed a flash-type event and as James Sinclair, chairman of MarketFactory, observes, “The January event was interesting because it does appear to be retail led. You only have to look at the currencies that moved and the currencies that didn’t. Those that did move were the Aussie, the peso, and the Turkish lira and the currency that didn’t was the Swiss franc. Usually if it’s an institutional move, you see the Swiss franc appreciate as part of a flight to safety.”
16 Market Structure Shifts
Again, this is probably easily explained through the numbers. Since the BIS survey in 2001, the first to be taken post-euro, the following changes stand out:
• Spot volume between reporting dealers has declined from 56% of activity to 36%, while that with Other Financial Institutions has risen from 29% of activity to 56%. With Non-Financial Institutions, however, turnover has halved from 15% to 7.1% over the same period.
• In 2001, 13 banks in the US and 17 in the UK handled 75% of FX activity (this was itself a sharp drop from 1998). In 2019, the FX Committee surveys show that just under 75% of activity is handled by four or five banks.
• The average trade size in 2001 was $4 million for spot and $31 million for FX swaps. In 2019, in the UK it was $800,000 for spot and $46.5 million for swaps – in the US it was $850,000 and $37.5 million, respectively.
As banks in particular got better at pricing their clients, internalisation rates went significantly higher. No firm data is available, but one senior trading source reckons that the internalisation rate for the spot business at a top tier bank has risen from around 5% in 2000 to above 80% in 2019.
“In 2000, it was about typing in a price, getting a customer to hit it and then managing that risk on a trade-by trade basis,” the senior trading source says. “Trading was a manual process and we hadn’t even heard of the word ‘internalisation’. That probably changed around 2006 when the banks got better at pricing and monitoring their positions and they started looking for matches within their client franchise, although I don’t think there was that much analysis done – someone just decided to try it.”
“Top of book has become much tighter, but liquidity down the stack is much more fragile. Signalling risk is higher now than it has been at any time thanks to the technology and analytics tools being thrown at the business”
That change did lead to a much greater emphasis on client facing activities. As highlighted by the first statistic above, it can also be argued – as one industry figure does – that internalisation has helped statistic number two above – market consolidation. “I don’t think internalisation has been a good thing at all because it has taken a lot of flow out of the public market, making price discovery harder and exacerbating periods of illiquidity. It has also probably led to a few banks who either didn’t have the client franchise or the technology smarts pulling back, meaning we have four or five dominant players in the industry. Is that a good thing? I don’t think so.”
There is little doubt that the evolution of technology allied to the role played by prime brokerage has helped bring more market participants into the industry and levelled the playing field. The success of the high frequency firms in grabbing market share on the ECNs especially has seen trade sizes plummet, mainly because these firms did not have the balance sheet or risk appetite to trade in larger sizes.
“Top of book has become much tighter, but liquidity down the stack is much more fragile,” says Maxine Dennis, an FX trader with 35 years’ experience who recently left SocGen/Newedge in London. “Signalling risk is higher now than it has been at any time thanks to the technology and analytics tools being thrown at the business.”
Probably the biggest driver of market structure change, however, has been regulation.
“The wake-up call that was the 2007-08 financial crisis has seen regulation change the industry,” observes George Athanasopoulos, global head of FX, rates and credit at UBS in Zurich. “Institutional clients have grown tremendously in terms of the assets they are managing, but they are also facing constant pressure on their costs and that means their liquidity needs are much greater than they used to be – in fact, they are greater than the industry can supply – but they want to pay less for it.”
Dennis agrees, “A big change has been the shift of risk from the banks to the buy side,” she says. “Not all clients are happy with this, they still want their banks to take risk, but it is becoming harder for them to do so thanks to the regulations, especially around balance sheet management.
One other measure of market structure shift can be found in an old issue of Profit & Loss. In 2001, it reported that the trade time for a streaming price was one-to-two seconds. We have no idea where that benchmark sits now, but suspect it is probably in the nanoseconds.
Next week we’ll be publishing numbers 15-11 on our list – “Last Look”, “Icap Buys EBS”, “The Bitcoin Bull Run”, “In the Legal Firing Line” and “The Inexorable Rise of the Non-Bank Firm”