On M ichael Lewis's latest book and High Frequency Trading

I have not red Michael Lewis's new book (I am a fan of his, and have used some of his older books as texts in my courses before). However, I have read the recent articles in the press about the book.

I do not find any logical arguments there specifically against high frequency trading. On the other hand I do see valid arguments against a whole host of nefarious practices on Wall Street such as front-running, preferential access (at a fee) to flow of orders, investment banks running their own private exchanges (dark pools), adding extra decimal places to pricing, etc. I also see a bunch of bogus arguments such as investments in high tech (use of private networks with very low latencies), superior algorithms, etc.  They are bogus because the alternative is to use ancient technologies that can not support the trading volume on the market and can not facilitate liquidity, the main reason for the very existence of markets.

As long as the economy follows the power law distribution for everything there will always be some using high-powered computers and some, at the opposite end of the spectrum, using the abacus. The solution is not to slow down the adoption of newer technologies but to provide the incentive to adopt such technologies. The SEC has not been doing its homework well when it comes to market technologies. SEC should look at everything in terms of ensuring equitable access (not necessarily equal) to order data and an impartial market clearing system.

 

High frequency trading is a red herring. The culprit is some one else.

In the US, until recently, the stock trading was concentrated in a handful of exchanges, but the market was not fragmented in that VERY few companies listed themselves on more than one US stock exchange (exceptions include HP, and Charles Schwab). With the development of the dark pools, the market got fragmented. There is nothing wrong with such fragmentation per se because it does provide some benefits: increased competition, and innovations in trading. However, it does have certain problems: lack of transparency, greater search costs, etc. as long as there is no single national market information system.

It is the lack of transparency that makes the fragmented system inherently unfair. They are called "dark" because of this; others haven't a clue what you are doing. The main purpose of such pools is to keep what they do a secret while having full information about what others are doing.

In the old days, when a very large order arrived at NYSE and there was a reason to suspect lack of liquidity leading to adverse market reaction to the trade, the order would be sent upstairs (to the ornate board room above the exchange, for those who have seen NYSE) or elsewhere, where in smoky rooms decisions would be taken on how to execute the order without upsetting the markets. That too, in a sense, lacked transparency, but those who indulged in that had a personal reputation to protect.

To repeat, high frequency by itself does not cause harm. What does cause harm are all the aspects of trading and execution that lack transparency because of artificially created information asymmetries. SEC should be taking a closer look at those factors rather than letting us move back to the abacus age by banning HFT altogether.