Wednesday, December 17, 2014

High Frequency Trading 101

I recently participated in an institutional investor round table where one of the topics of the day was high frequency trading. Although embarrassed to do so, I will admit that I had to do some serious groundwork on this topic because I had heretofore largely avoided it in any substantive way. If you (or your students) are in the position I was in just a few weeks ago, this post may be a good starting point to understanding a very complex and interested set of issues.

Being new to the high frequency trading debate, I needed to build a basic understanding of the issues. If you haven't read Michael Lewis' Flash Boys (or anything other than this delightful synopsis courtesy of the NYT Magazine) check out Forbes' explanation of high frequency trading.  Even if YOU don't need it, this is a great reference for students interested in the topic.  

Of course, another starting point was the flash crash of 2010, where the Dow Jones Industrial Average fell over 1000 points in a matter of minutes.  The flash crash wasn't the start of high frequency trading, but it was an event that highlighted the role it plays in the markets.  You can read the SEC's report on the Flash Crash here. The publicity raised awareness and scrutiny of the practice.  For example, the NY Attorney General indicted financial institutions in June 2014 for practices related to high frequency trading.

The debate on the pros and cons of high frequency trading can be boiled down to two very simple points, and both relate to efficiency.  High frequency trading promotes efficient market pricing by  relaying information across markets and reducing buy-sell price spreads.  Eric Budish and John Shim both at the University of Chicago and Peter Cramton at University of Maryland published a 2013 paper studying S&P 500 trading data where futures and exchange-traded funds are correlated at the minute intervals.  Their study found that the correlation disappears at the 250 millisecond interval.  At the 250 millisecond interval the prices are discordant, but by 1 second the price has smoothed.  This is how the bid/ask spread shrinks, a trend of efficiency that has reduced the bid/ask spread from 90 basis points 20 years ago to 3 points today.

The second argument is that high frequency trading is bad for small investors because it is not value oriented (buy and sell decisions have no relationship to the underlying value of the assets) and capitalizes on the supply/demand problems posed by large institutional investor buy/sell orders.  The fear is that high frequency trading distorts the long term value proposition of stocks.  This argument suggests that high frequency trading is not efficient but rather is superfluous financial intermediation because it isn't connecting buyers and seller of securities (making markets), but is jumping in between buyers and sellers who would otherwise find each other.  

A newly posted article on SSRN by Jonathan Broggard et al. using Nasdaq data finds that high frequency traders provide liquidity (and therefore stability) in times of high financial stress, and thus may perform a protective market function.  The same study also observes that in normal market conditions, high frequency traders demand more liquidity than they create.  These findings suggest the validity of both arguments summarized above.

If you have any suggestions for must read articles-- academic or popular press--on high frequency trading, please respond in the comments.

AT

https://lawprofessors.typepad.com/business_law/2014/12/high-frequency-trading-101.html

| Permalink

Comments

I would suggest: Dark Pools: The Rise of the Machine Traders and the Rigging of the U.S. Stock Market, by Scott Patterson; an article by Yesha Yadav, Beyond Efficiency in Securities Regulation. I don't think it's published yet, but it's available on SSRN at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2400527.

Posted by: Steve Bradford | Dec 17, 2014 5:43:14 AM

I agree about Dark Pools. I think it is a better book than Flash Boys although it clearly took someone of Lewis' stature to kick this issue into the spotlight.

Albert Menkveld (VU Amsterdam) has done some of the most useful empirical work on HFTs. Donald MacKenzie (Edinburgh) has written extensively on them from a more critical position (including one long but accessible piece in the LRB).

I think the main thing is to keep in mind that essentially all traders are now HFT's. The question is whether the post-Reg. NMS market structure has created an opportunity for predatory or opportunistic behavior. The evidence my co-author (Jenny Kuan) and I have put together suggests the answer is yes, as we find wider bid-ask spreads in the wake of NMS. (The spread is, in part, a price for the risk of adverse selection.) Even where authors have found narrower spreads the transaction costs of out-strategizing predatory HFTs are not usually accounted for. We are finalizing the paper for submission to finance journals but earlier versions are on SSRN.

Posted by: Steve Diamond | Dec 19, 2014 2:23:47 PM

Post a comment