ForeCastle

High frequency trading has been in the news more, thanks in part to Michael Lewis’ new book, Flash Boys. This article presents a simple explanation of how and why high frequency trading works, and why it is good for small investors.

We will begin by imagining a market with lots of small individual traders. Then we will look at how large institutional investors change the market. Next we will look at high frequency trading. Finally, we’ll explain how small investors are impacted.

Start by imagining a stock with no particular news about it. The price is stable, but there are lot of small trades. Some investors have enjoyed gains but now think the stock is overpriced. Other investors have seen gains and have decided to jump on the bandwagon. Some investors have been watching it, and now have money to invest. Others have owned it and are happy with the stock but need some cash. So lots of orders are coming in, pretty evenly mixed between buy and sell orders. The price trend for the stock looks perfectly steady.



Now consider that the traders are not all small investors. Large institutional traders are doing the same thing—some buying and some selling—but there’s a difference between them and individual investors. When a large mutual fund or pension fund places a buy order, it could be for a million shares, not a hundred shares. Similarly, sell orders from institutions come in very large quantities.

Over the course of the day, these large institutional orders cause a lumpy pattern. The chart shows what such a price line looks like. There is no noticeable trend up or down, but each institutional order moves the market up or down, and it takes a while for the price to return to the underlying trend line. That’s illustrated with the red line in the accompanying chart.

High frequency traders try to profit from the price movements caused by large institutional trades. When a mutual fund sells a million shares of a stock, the price dips—and HFTs buy on the dip, hoping to be able to sell the shares a few minutes later at the normal price. When a pension fund buys two million shares, the HFTs short-sell the stock, hoping to close their position at a profit. (Short selling is selling stock you don’t own; you borrow the shares from a stockbroker, sell them, and then later buy the stock to return the borrowed shares.)

HFTs are buying when the price is below trend and selling when the price is above trend. This tends to reduce the price fluctuations. When they are successful, prices look like the blue line on the chart. The blips are smaller and shorter-lived.

HFT is not as easy as this simple explanation sounds. First, there are many HFTs. If one is slow, the profit opportunity may have been captured by other HFTs. Second, not every blip is just a blip. If the stock is impacted by an downward trend in the overall stock market, the HFT would buy lots of different stocks—and then watch them all go down further. A good HFT has to be fast, but not so fast as to get caught be a surprise. In practice, the HFTs are no longer just looking at just one stock in isolation. They are looking at all the prices coming in, including stocks, bonds, commodities, futures and options. This massive data crunching helps them identify what are likely to be short-term blips but not long-lasting trends.

In the early days, it was fairly easy. As more companies got into the business, the easy trades were quickly taken by others. HFTs needed to move faster and faster, while crunching ever more data to avoid losing trades. Much of the attention they have received lately is due to their extreme efforts to reduce their reaction time, which is measured in milliseconds. This effort is not made to be faster than individual investors or institutional investors; HFTs are already faster than them. Instead, the effort is made to be faster than competing HFTs.

Now, how does high frequency trading impact those of us who are small investors? Look at that chart. If I place a simple buy or sell order, I may get lucky or unlucky. My buy order may be at a downward blip, but it may also be at an upward blip. I don’t want to get lucky if it means a chance of being unlucky; I’d rather trade at that underlying trend price.

Further, investors face a spread between the price at which they buy (the “ask” price) and the price at which they sell (the “bid”). This bid-ask spread compensates the market makers for executing trades at exactly the time that I want to trade. The more volatile the stock price usually is, the wider the bid-ask spread. HFTs tend to narrow the bid-ask spread by protecting the market makers from bad news while they hold their positions. Thus, my trading costs get lower.

High frequency trading is secretive and mysterious, but not at all evil. It make the stock market more efficient and helps small investors who trade at random times over the day. I could almost feel sorry for them being misunderstood—until considering that they have made far more money than I have.

More...