Recently two books have appeared that highlight “dark pools” (i.e., new trading venues that permit one to keep trading activity relatively private, at least for a limited time), and “high-frequency trading” (i.e., trading performed by computer algorithms and keyed to very fine-grained time intervals):
Dark Pools
Dark Pools (2012). Scott Patterson, a staff reporter for the Wall Street Journal, introduces the reader to computerized trading algorithms, then recounts the history of the emergence and proliferation of independent trading venues and computerized trading. We learn about the many small firms that rose to prominence — e.g., Island, Instinet, Archipelago, Datek, Getco, Tradebot Systems — and also about many of the people behind these firms — e.g., Joshua Levine, Shelly Maschler, Jeff Citron, Jerry Rosen, Joe Cammarata and David Cummings.
Readers also learn about the early software that these pioneers wrote, which gave them the edge in their trading: Levine’s “Watcher” program, which permitted users to monitor literally hundreds of stocks or securities; the Datek “Monster Key,” which permitted Datek traders to jump to the top of the queue and thus outmaneuver competing traders; and Tradebot, one of the first fully algorithmic trading programs.
Finally, Patterson recounts the rise of high-frequency trading, where scrappy competing software programs reacted to generate trades in seconds, then in milliseconds and then in microseconds, and what this breathtaking development means for the future of market trading.
Flash Boys
Flash Boys (2014). Michael Lewis, a columnist for Bloomberg and the New York Times Magazine, introduces his book by recounting, in considerable detail, the construction of an 827-mile fiber-optic link from the Chicago Mercantile Exchange in Chicago to the Nasdaq stock exchange in Carteret, New Jersey. The cable did not merely go along conventional railroad or highway right-of-ways; instead it cut through mountains so as to minimize the distance. The point of this highly secret and very expensive construction project was to shave two or three milliseconds off the communication time between the two sites, to give a slight advantage to a select set of high-frequency trading firms.
When Dan Spivey, the mastermind of the project, approached prominent Wall Street investment banks with a very steep price (US$10 million) for a dedicated piece of the fiber in this cable, their response was, typically, “Can you double the price?” (so as to make it more difficult for competitors to have a comparably fast line).
Much of Lewis’ subsequent narrative centers on the adventures of Bradley Katsuyama, a Canadian market trader formerly employed at the Royal Bank of Canada, described by other trading firms in New York as “Canada nice.” When he was transferred to New York in 2002 and began to see first-hand the world high-frequency trading, and after he worked out why his own trades were failing to be executed, he became very concerned that some of these sophisticated operations were granting certain sectors of the industry an unfair advantage. Finally, in 2013, he and a colleague named Rob Park opened their own stock exchange, known as “IEX.” Unlike other exchanges, it is completely open, and has only a very limited selection of very well understood order types, thus thwarting numerous controversial practices.
In the wake of Lewis’ book and Katsuyama’s testimony, the U.S. Justice Department has launched an investigation of high-frequency trading, both for possible insider trading rules violations, and also to determine whether the computer trading places individual investors at a fundamental disadvantage.
Discussion
So which of these books is better? If one wants an exciting read and a high-level overview of the world of high-frequency trading, Lewis’ book is probably the better selection (and it’s somewhat more up-to-date). If one wants a more detailed look at the field and its history, then Patterson’s book would get the nod.
Inevitably neither book really goes into any technical depth, thus leaving many specific questions largely unanswered:
- What is the actual advantage, quantitatively speaking, of speed? On what types of trades?
- How are trades in the options and commodities markets affected?
- What types of algorithms, mathematically speaking, are used by various high-frequency trading operations?
- How has the advent of high-speed trading affected mathematical trading operations?
- What are the real issues for regulatory agencies?
- To what extent have the markets become dependent on high-frequency operations for efficient operation (e.g., to maintain optimal liquidity)?
- How do we quantify the benefits to the individual investor of high-frequency trading (e.g., to ensure the best price to the end user)?
While some of the issues are addressed in these books, in most cases the treatment is relatively superficial. But until such issues are deeply understood, it will be impossible to truly understand the brave new world of high-frequency finance, much less craft effective and non-counter-productive regulation. For example, both books contain many examples of how each new and well intentioned regulation gave rise to another opportunity to exploit and thwart the new environment, e.g., decimalization, the requirement to buy shares at the “best current prize,” and so on.
As both authors make clear, many of the answers to these questions remain poorly understood, even within the industry. Moreover, precise answers, even if available, would likely be highly technical and inappropriate in books such as these. Indeed, in Dark Pools, there is already a blizzard of acronyms and names to keep track of. One thing is for sure, namely that high-speed algorithmic trading is here to stay. The question is how to tame the beast. The amounts of money being made and lost are enormous, and the downside risks of unregulated dark trading are almost unimaginable.