BackPound flash crash puts spotlight on robot trades
Analysts say that the sell-off in the pound was probably exacerbated by computer-driven traders reacting at speeds faster than any human could muster
Premium
Making sense of the foreign-exchange market is Derek Mumford’s bread and butter, but he couldn’t explain this.
In the span of just two minutes in early Asia trading on Friday, the British pound had plunged more than 6 percent, sending the fourth most-traded currency on the planet to the lowest level in 31 years. Mumford, who advises companies on foreign-exchange and interest-rate risks, scrambled to find out why. There was talk of France’s president pushing for a hard-line approach on Britain’s exit from the European Union, and recycled rumors of a “fat finger" trade, but nothing that would justify a drop of this magnitude.
“It was out of proportion to the supposed trigger," said Mumford, a director at Rochford Capital Pty in Sydney.
While he may never be able to pin down the catalyst for Friday’s drop, Mumford and many of his peers agreed that the sell-off was probably exacerbated by computer-driven traders reacting at speeds faster than any human could muster. So-called algorithmic transactions in the foreign-exchange market have more than tripled over the last three years, accounting for almost $200 billion of daily turnover, according to Aite Group, a consultant in Boston.
That one of the planet’s oldest mediums of exchange -- and the dominant reserve currency until the first half of the 20th century -- could move like the legal tender of a frontier country is almost certainly to fuel debate over computerized traders’ role in the $5.1-trillion-a-day global currency market. Friday’s move in the pound follows a flash crash in the South African rand in January and a similarly unexplained move in New Zealand’s dollar last year.
“It’s one of these moves that we’re seeing more regularly," said Hugh Killen, the Sydney-based head of foreign exchange, fixed income and commodities trading at Westpac Banking Corp., Australia’s second-biggest lender. “People are caught in a flash crash that does seem to be algorithmic driven."
Fragility in foreign-exchange markets has been increasing even as “phantom liquidity" creates an illusion of stability, Bank of America Merrill Lynch strategists Chris Xiao and Vadim Iaralov wrote in an Oct. 5 report. While noting that traditional liquidity indicators including bid-ask spreads have narrowed, they said one measure of the market impact for any given trade has climbed by 60 percent since 2014. The frequency and magnitude of “outsized volatility events" has also increased, the strategists wrote.
Perfect Storm
While the pound’s intraday drop of 11 percent after Britain’s surprise vote in June to leave the EU was bigger than Friday’s slump, traders said this one was more jarring. Not only did the crash happen in just two minutes, but nobody was expecting it. The move reminded several dealers of the reaction to the Swiss National Bank’s shock decision to abandon the franc’s cap against the euro in January 2015, which sent the exchange rate surging more than 40 percent.
“This is the most volatile move seen from sterling since Brexit, yet it can be argued that relative to Brexit, this sell-off was more dramatic," said Matt Simpson, a senior market analyst at ThinkMarkets in Singapore. “Nobody was prepared for it."
Volumes in sterling are typically lower during Asian hours, but may have been even smaller on Friday given a dearth of market-moving news and a reticence to take positions before the monthly U.S. employment report, Simpson said, calling it a “perfect storm for a price shock." Some traders pointed to several large pound positions in the over-the-counter options market at levels including $1.25 and $1.20, which could have exacerbated the currency’s move as it sank through those levels.
It’s too early to come to a conclusion about the role of algorithmic traders in Friday’s rout, said Ralph Achkar, capital markets product director at Colt, which provides trading infrastructure to electronic dealers.
“We have come across several market incidents where the initial finger was first pointed at algos, only for it to turn out to be human error or otherwise," Achkar said.
A Bank of England spokesman said the monetary authority is looking into the cause of the crash.
There was some confusion over how low the pound actually fell. Bloomberg’s composite price -- which takes the median contribution from a range of dealers -- showed the currency dropping as low as $1.1841, the weakest level since March 1985. But some banks quoted sterling at even lower levels, with at least one electronic trading platform recording a transaction as weak as $1.1378, according to traders familiar with the transactions who asked not to be identified because they aren’t authorized to speak publicly.
The discrepancy left some dealers in Asia waiting for client mediation teams to decide whether some stop-loss orders were filled, the traders said. One dealer said his foreign-exchange pricing aggregator of eight contributors blacked out for 30 seconds amid an absence of bids.
The size of the move blindsided derivatives traders who had in recent months been paring expectations for swings in the pound. A measure of one-month implied volatility in the currency had dropped 70 percent from a high in mid-June through the start of this month. It surged 17 percent on Friday.
While the flash crash may heighten concerns over the potential risks in markets where computers can execute hundreds of trades in an instant, there were few signs of contagion on Friday. The yen, a traditional haven during times of turmoil, was little changed. Other markets took the episode in stride, with Asia’s benchmark stock index and futures on the S&P 500 limiting losses to 0.3 percent and 0.2 percent, respectively. The pound pared losses to 1.4 percent as of 10:49 a.m. in New York.
For Westpac’s Killen, the flash crash refocuses attention on the changing structure of foreign-exchange markets as global investment banks pull back from dealing. Currency sales and trading headcount at 12 of the largest banks shrank 28 percent worldwide between 2010 and 2015, according to Coalition Development Ltd., a London-based research firm. Under the so-called Volcker Rule, lenders are restricted from trading for profit with their own money.
“Clearly the market is much more electronic, and as a result it is more much driven by algorithms," Killen said. “When you combine this with the fact the major traditional players in the FX markets carry a lot less risk than they might have done a few years ago, we’re getting greater liquidity gaps." Bloomberg
With assistance from Mark Cranfield Jake Ulick Lananh Nguyen Michael Wilson David Finnerty and Andrea Tan
Unlock a world of Benefits! From insightful newsletters to real-time stock tracking, breaking news and a personalized newsfeed – it's all here, just a click away! Login Now!
