Stocks quoted in this article:
Everybody's talking about high-frequency trading these days. That's, of course, because Michael Lewis is talking about high-frequency trading.
The criticisms really have nothing to do with the book itself. I haven't had a chance to read it in the two days or so since its release, and I'm guessing 99% of the people commenting on it haven't read more than the cover or an excerpt either. Rather, the complaints as I gather are about why something that's been known around Wall Street for a while now is suddenly generating interest.
I mean, the "flash crash" took place almost four years ago. And if memory serves me correctly, we pretty much blamed the high-speed traders. Specifically, we blamed them for essentially shutting the machines down and eliminating any illusions of massive liquidity that they provided.
Broken Markets: How High Frequency Trading and Predatory Practices on Wall Street are Destroying Investor Confidence and Your Portfolio by Joe Saluzzi and Sal Arnuk came out nearly two years ago and provided a pretty comprehensive look at the issue (it's a great read, by the way).
In other words, high-frequency trading is hardly a new issue; rather, it's just one that's suddenly getting massive publicity. We're really criticizing ourselves, I suppose, for not being loud enough about it. It's hard to blame Michael Lewis for basically doing what he does quite successfully. If you're of the mind that we need changes in how the exchanges handle the algos, then you should be thankful.
That, of course, leads to the question of what changes we can actually effect? The goal is to somehow level the playing the field. But of course, the playing field is never level. If we were entering orders via Dixie cup and string, Goldman would find a faster type of string and a better Dixie cup. We can't stop technology from developing, and whoever has more resources will always figure out an edge. We can take steps to help, though.
One of the ideas is to prohibit "co-location" on the exchanges. Basically, the proximity cuts precious nanoseconds off the process of seeing the orders and executing the trades. As Matt Levine notes on Bloomberg View, though, the machines would just move across the street and still garner the same edge. I'd still favor the proposal. I mean, it's essentially an institutionalized kickback to the exchanges. But he's probably correct in that they'll find a locational advantage anyway and it won't change the system much.
Another proposal is to literally change how trades are matched.
Currently, securities are traded continuously, so that orders are accepted and matched by price, with ties broken by which order arrives first. This system emphasizes speed over price, rewarding high-frequency traders for flooding the market with orders. One detailed proposal would seek to correct this imbalance by processing orders in batches in frequent intervals, to ensure that price -- not speed -- is the deciding factor in who obtains a trade.
The idea is to only allow execution every second or two. Getting in a nanosecond early has less of an advantage that way. I realize there are probably unforeseen consequences for a fairly radical market change like this, but it's certainly a concept that holds promise.
Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.