Having included Michael Lewis’ “Flash Boys” in my reading on a recent trip to Florida, I was fascinated to discover that the FBI are investigating high-frequency traders (HFTs) to see if they have broken any laws. But surely, if there is any credence in his allegations, shouldn’t the institutional investor sector be up in arms?
Two core themes struck me in Michael Lewis’ book. On the one hand he talks of the iniquities of the millisecond timing advantages of the HFT community. But timing advantage has been around for expert investors from time immemorial. The most famous example was Nathan Rothschild’s alleged dealing in U.K. government bonds in 1815, after his messenger reached London with the news of the defeat of Napoleon ahead of the official envoy. Technology has definitely changed the structure of timing advantage with the speed of horses and boats being replaced by the length or latency of the fiber-optic, microwave or other communication link between trader and exchange.
The main area of concern, and surely potential criminality, in the process Lewis describes, has to be the allegation that exchanges structure their business to sell timing and information advantage to certain investor groups. That would allow them to act in favor of one group, primarily high-frequency traders, according to Lewis, and against the interest of all other investors. He also argues that the creation of multiple exchanges has in fact facilitated this process both by fragmenting liquidity and changing market dynamics.
I have to admit to having felt somewhat jaded at the end of the book, for the distinction between good and bad was too stark and simple; and some of the arguments appeared based on anecdotal rather than explicit evidence. But the book does raise serious questions for markets. Is the apparent liquidity provided by HFTs toxic or beneficial? Is the presence of multiple exchanges to the advantage of all their clients or just some of their clients? Are dark pools a sensible medium for large transactions or a surrogate for adding spread to the deal? And what will be the long-term impact of the spread, and growing dominance, of HFTs in market turnover?
The question of liquidity is highly relevant. The classical market maker in the days of “Big Bang” essentially took on positions, both short and long, in the knowledge that information on these would be kept private, thus allowing time to unwind them. Regulatory belief that market transparency was paramount has meant that the privacy of the information has suffered, and thus also a willingness to accept risk. The classical HFT does not provide liquidity to the end user. They trade against their perception, or understanding, of pending market trades and will normally net out around the trade rather than with the trade. If the estimate that such traders make around $10-30 billion a year is accurate, they are a costly addition to the market. If the assumption that they trade around the trade is correct, their contribution to liquidity is toxic and illusory rather than beneficial.
The presence of multiple exchanges and dark pools must by definition reduce liquidity as supply and demand for different stocks is spread over more trading venues. However, if the existence of multiple venues creates arbitrage opportunities, then they will also create incremental liquidity. There was a view, when several private exchanges were first created, that their prime purpose was to ensure that the main public exchanges trimmed their fees and adapted their business model to the 21st century. In fairness to the legacy exchanges, this is no simple task as they cover a wide range of shares with different liquidity and different earning potential, while the private exchanges tend to focus on the most liquid stocks. But the problem is more one of the pricing of exchange participation and the need to differentiate by stock according to the underlying value of the associated revenue flow to the exchange. This may not gel with public policy in the different countries where the legacy exchanges are based, but it appears mission critical to the survival of markets, with broad coverage of corporate capital issues, for cross-subsidies definitely work to their disadvantage.
The issue of whether more exchanges are advantageous was highlighted by the recent merger of BATS and Direct Edge Holdings. Logically one would have expected the company to create a single trading venue, which would be easily the third largest exchange in the U.S. However, as Michael Lewis notes in his book, although they will use identical technology, they will maintain separate markets. Intuitively, this is counter-productive and not cost effective. It is logical to assume that continued market fragmentation, given that shareholders of the two groups include GETCO and Citadel, is primarily to the advantage of the trading community. After all, given the multi-billion dollar annual value of HFT and the much lower capital value of the exchanges in question, one has to assume that it is the former that drives any decision rather than the latter.
But what will the long term impact of HFTs be? They do create risk. At the exchanges there is a serious level of technology risk as faster and greater capacity is always demanded. We have more and more examples of trading outages as exchanges move into new worlds in terms of transactions per millisecond. At the firms, there is technology and business risk. This was most graphically evidenced by the 2012 failure of Knight Capital. It accounted for around 17% of the volumes of both NYSE and NASDAQ, but it lost around $400 million on computer-generated erroneous trades and had to be rescued after its value on NASDAQ fell 75% in just two days. And for the investing public there is structural risk as HFTs focus on the most liquid stocks, which by number are far less than 5% of all stocks traded. This must lead to material transformation in the legacy exchanges’ business models and will undoubtedly harm new capital creation by small and mid-size companies.
Michael Lewis paints, perhaps, a too black-and-white picture. But the trading markets are seriously flawed. Technology has created a risk environment where a firm, with a 15-20% equity market share in the largest capital market in the globe, can run into financial trouble through a mere $400 million loss. Regulation may have destroyed the classical market making activity because of a desire for too much and too rapid transparency. Exchange fragmentation and dark pools are of debateable value to investors. Too much of the current apparent liquidity is toxic, and too many small and mid-sized companies risk marginalization in global capital markets.
An interesting point made by Lewis was the concern of some major institutions in the U.S. around exchange structures. Indeed, it is up to the institutions to demand change; after all, they are paying for the bulk of all the extra costs, primarily in spread, of any delinquencies in the current environment. Unfortunately, they have traditionally remained mainly silent on major market issues.