The latest post from Themis trading to lobby against HFT (so as to protect Themis’ business) is up on the net( check out: http://www.ritholtz.com/blog/2011/03/hft-reverse-splits-and-hidden-signals/ at The Big Picture blog). Unfortunately, the post from Themis contains a number of cognitive biases and just plain errors.
They quote a paper from “two very distinguished Cornell professors” (a basic cognitive bias of appeal to authority, as well as a basic error, one of the professors is at Columbia, not Cornell – sloppy work from Themis).
The post refers to a study, saying “we finally found an academic study” where “the conclusion was not that HFT was great since it shrinks spreads and increases liquidity.” I find it remarkable that Themis make plain that they have been searching for a study to support their own lobbying, and that they “finally” found one. This is called confirmation bias, but it is rare that lobbyists admit to it so readily. Or foolishly?
The Themis lobbyist then goes on to try to discredit all of the other studies, saying the overwhelming plethora of research studies that conclude the benefits of HFT are “funded by an exchange or large brokerage firm”. Really, such childish appeals to the desire to find conspiracy theories is tiresome. What do you expect from a lobbyist, though?
The Themis lobbyist then goes on to selectively quote tiny portions of the research. What the Themis lobbyist doesn’t quote, though, is this line in the paper: “The existing empirical literature, although mixed, supports the conclusion that it improves market efficiency by reducing bid/ask spreads and market volatility while making markets more liquid.” Let’s read that again; HFT “improves market efficiency by reducing bid/ask spreads and market volatility while making markets more liquid.”
The paper quoted is a valuable one; it is just misused and misquoted by the Themis lobbyist. For example, in the paper the authors attribute some of the profitability of HFT to “the differential speed advantage of high frequency trading that causes the inequity. To the extent that the speed advantage is generated by preferential treatment in the execution of market orders, these can be eliminated.” This is a good point, preferential treatment should be removed if it helps to reduce the unevenness of the playing field.
I urge anyone wanting to look beyond the crude Themis lobbying efforts to read the whole paper, it can be downloaded here: