Ben Graham and High Frequency Trading

Just read Ben Graham’s book, “The Intelligent Investor” which taught me a lot about investing, specifically about the logical divide between speculation and value-investing. It was really insightful.

Armed with that knowledge, and having my curiosity stirred by all the opportunities being offered to software developers in High Frequency Trading, I started thinking about HFT field and how it relates to Graham’s principles. Trying to make a long story short: given the little I know about HFT, the two don’t jive. 

HFT is based on inefficiencies in the market place, and can only be profitable while those inefficiencies exist. It seems to me that at some point, as this pool dries up, there will be an overvaluation. But what’s more disconcerting, personally, is the rapidity and lack of human interaction to this hack of a field (I call it that because it’s using the market in a purpose for which it was not intended to be used). 

Value investing, at its core, involves a deep understanding of the companies you’re going to invest in. Certain securities (indexes, certain etfs, etc.) can be used to value invest with less of an understanding, but that’s besides the point. Pulling the trigger on a trade only happens after you’ve come to the conclusion that the investment is sound. The risk exists in the tail end of your understanding; the unpredictable noise that is the sum of the unaccounted-for, difficult to measure variables. For example, weather, sickness, related industry failure, or any of the number of things you couldn’t infer from the earnings report. But that’s why you diversify, hoping that for a group of sound companies, those variables don’t overlap, and that you’ll make money in the long run, no matter what.

But for HFT the trigger is getting pulled a bazillion times each second, based on the assumption that you’ve determined a pattern from what is essentially noise – market fluctuations. Graham noted a number of “trading strategies” which were essentially arbitrary algorithms trying to pull causation out of correlation (which is really the only conclusion you can come to when no consideration for the value of your investments is being made in your trading decisions). They all failed, and took a number of people out with them. To me, HFT seems very similar to such examples, the difference being here that the magnitude and volatility of trades both are much greater, as I expect the fall out will be.

I was talking about this with some friends at work, and today this article make it’s way onto my reading queue. Pretty wild to see how much craziness HFT has introduced into the stock market. Makes me wonder if the recent explosion at Knight Capital represents the floor or the ceiling to the problem. Also, I wonder how much the risk of HFT is correlated to the risk of the market in general?
All of that being said, these are the first thoughts that pop into my mind given an extremely naive understanding. I’d love for someone to point out the glaring hole in my argument. Regardless, first things that popped into my head, given the little I know.