The publication of Michael Lewis’s book Flash Boys in 2014 thrust the arcane world of ultra-high frequency trading firms into the popular consciousness, prompting debate over everything from the industry’s morality to the possible risks it posed to financial stability.
What was undeniable was how several of these companies were staggeringly profitable, generating hundreds of millions of dollars by trading at speeds made up of fractions of seconds. More recently, however, life inside these trading black boxes has taken on a less charmed appearance. Last November Teza Technologies, a large high-frequency trading shop, announced it was quitting the HFT game as it could no longer make any money. Rivals have also announced falls in their trading profits.
So are we witnessing a meaningful long-term decline of these high-frequency trading shops or has the last year simply been a blip?
Many HFT firms are privately held, meaning we are forced to examine the financial statements of the few publicly quoted ones to get a sense of what is going on inside a complex industry. What can be gauged from publicly available information is that the amount of money they make is closely related to market volatility.
Since going public in 2015, trading income at one of the leading HFT firms Virtu Financial, whose one time pit trader founder Vincent Viola joined the Trump administration, have displayed a close correlation with the level of the Vix index, the Chicago Board Options Exchange’s volatility index which takes as an input the implied volatility of options being priced on the S&P 500 over the coming 30 days.
When volatility is high, bid-ask spreads get wider and market makers, especially of the ultra-high speed variety, can profit. Yet with the Vix having remained subdued, Virtu’s earnings have fallen steadily since it went public, with its total adjusted net trading income falling every quarter of 2016 compared to the same period in 2015. Shares in Flow Traders, an Amsterdam-based outfit also listed in 2015, have languished below the price of their initial public offering.
The benign explanation for all this is that the very fact certain HFT outfits are remaining in profit while riding out this long bout of low volatility demonstrates the barriers to entry around their businesses. According to Virtu, the average intraday volatility of the S&P 500 in August of last year hit lows not seen in records going back to 1970, something the company said was to blame for its net trading income in US equities falling by a fifth from the second to the third quarter, and by more than a third compared to the year before. Following this logic means one can expect that when volatility returns to more normal historical levels, Virtu will return to to the heights it reached when the company listed in 2015.
The stock market currently appears to believe this to be the case. Virtu’s shares have rebounded since falling to an all-time low just before the US presidential election on hopes that a jump in Donald Trump-fuelled US trading activity will boost earnings. And Virtu, which reports results this week, certainly remains a profitable business, achieving an earnings before interest tax, depreciation and amortisation margin of 54 per cent in the third quarter, even if this was down from 66.8 per cent in the same period in 2015.
A far less encouraging explanation for investors in these businesses is that the sustainable trading edge that had enabled some of them to be so impressively profitable has for some reason quietly started to be eroded by competition.
Shareholders in listed trading and market-making firms have watched numerous times over history as seemingly robust trading businesses with apparently durable competitive advantages have lost their way. Knight Capital, which blew a $440m hole through its balance sheet in 30 minutes of trading in 2012 and had to be rescued, is one of the most infamous examples of failure. But many others have simply fallen into decline. Van Der Moolen was a Dutch market maker that by 2007 accounted for 11 per cent of NYSE trading volumes but found itself muscled out by new Algorithmic competition and filed for bankruptcy in 2009. Before it bought out Knight, the Chicago-based Getco suffered several years of significant erosion of its once elevated profit margins.
Investors in Virtu’s common stock must now ask themselves some frank questions. When Virtu first listed its seemingly unparalleled profitability convinced investors buying into its IPO to value it on a multiple of earnings, rather than a multiple based on its book values. Today, Virtu trades on more than 17 times its book value and it has a negative tangible book value.
If profitability continues to decline, then it will become increasingly difficult to convince the market that its shares should trade at such an extreme premium to book. Anyone interested in the health of the wider HFT industry will be tuning in when Virtu reports results on Thursday.
Sample the FT’s top stories for a week
You select the topic, we deliver the news.