It is possible to be nostalgic about the era before computerized trading, and thus before high frequency trading (HFT), when the transmission of orders took place by means of a human voice over a landline telephone. But we must qualify any such nostalgia. Consider that the market makers at Nasdaq as recently as the 1990s were colluding to avoid odd-eighth quotes, protecting their own spreads in a way that could never happen in today’s environment. Economists William G. Christie of Vanderbilt University and Paul H. Schultz of the University of Notre Dame, in a 1999 paper in the Journal of Financial Economics, found that despite some then-quite recent changes, “the ability of multiple dealers to coordinate their quotes in ways that are not at all transparent [has] a significant impact on trading costs.”
How stand things today, in the era of the $0.02 commission?
Eric Hunsader, the developer of real-time trading and market-data software, expressed his objections to the current state of HFT and the associated algorithms in an analysis posted on the Nanex website on August 8. Hunsader contends that HFT is “killing the eMini,” an electronic S&P 500 futures contract.
Criticism of HFT is itself old news, but Hunsader’s analysis has some new twists. It begins with the simple quantity of information that a real-time market service as Nanex, covering U.S. equities, options, futures, and indexes, must process. Nanex processed 1 trillion bytes on August 5. A year ago, each day’s data was half of that, and two years ago, it was one-quarter what it is now: 250 billion bytes.
“There is no new beneficial information in this monstrous pile. … It is noise, subterfuge, manipulation.”
From there, Hunsader narrows his focus to the eMini, a contract introduced by the Chicago Mercantile Exchange in 1997. He presents charts showing that liquidity in these contracts, as measured by the average sum of bid depth sizes, has dropped markedly in recent days, so that a sale of 2,000 contracts “would have sliced right through the entire book” through most of the day August 5.
Further, Hunsader contends that he has identified a specific algorithm that has done a lot of this work of leaking liquidity out of the eMini market. He has dubbed the algorithm The Disrupter.
The Disrupter works on two fronts. On one, it buys (or sells) a lot of eMini contracts, “enough to cause a disruption.” On the other hand, it intervenes in a big way in many of the exchange traded funds (ETFs) that are used to hedge or arbitrage a position in the eMini market. It makes those markets “soft and sloppy” so that the arbs get poor fills.
The arbs lose money and, since they aren’t idiots, they conclude that they can’t compete with the speed of light and they drop out of the market. In effect, robots take over as humans drop out, so although all the HFT trading itself creates the “illusion of liquidity,” by driving out other types of trader it eliminates real liquidity.
What to Do
“We can’t understand why this is allowed to continue, because at the core, it is pure manipulation,” he wrote.
In a telephone interview August 9, Hunsader expanded somewhat. He said he did not believe that the Securities and Exchange Commission’s recent regulation on HFT, rule 15c3-5, which came into effect in July 2011, will have any effect on the matter.
“I don’t believe in any new regulation. They’ve got Reg NMS. They should enforce it.” This is a reference to Regulation National Market System, a rule created by the SEC in 2005 to prevent the execution in one market of a price that was inferior to the price displayed on the same security in another market.
He also said that the first time he observed The Disrupter in the data was April 27, 2010. The second time was May 6, the afternoon of the infamous Flash Crash.
But some observers disagree with Hunsader’s view that there is a problem requiring solution.
Eric Falkenstein, the author of Finding Alpha (2009) and who served as portfolio manager for both Deephaven and Telluride, said Tuesday that spreads have come down remarkably since 1990, and that this must be attributed to the waves of innovation that have hit the securities industry since that time. Innovation and efficiency, he said – which have made impossible the collusion of which Christie and Schultz wrote – involves profits and losses, and the only way to make the former and avoid the latter is by getting information into prices. The faster the former is transformed into the latter, the better.
Specifically, Falkenstein sees Hunsader’s thesis as, in a word, ridiculous. “High frequency trading has introduced a massive amount of competition, and competition is the only mechanism that can and will keep it honest. All it takes is ten smart people competing with each other to whittle profits almost down to nothing, all the while increasing the depth and lowering the spread to retail investors.”