What you are looking at it is not a video game from the 1980s or a completed assignment from an intro to programming class. You are looking at the stock market in action, and bear with me for a minute, because it’s important to understand just what’s going on here and how outrageous it is.
The twelve rectangular boxes represent stock exchanges. The turquoise one on the left, NYSE, is the New York Stock Exchange, the pink one on the right is the Boston Stock Exchange, etc. The color-coordinated shapes flying between these exchanges are trades. All of these trades are for the Mastercard stock. So far so good? This is just a visual representation of commerce among various markets on one particular, unremarkable stock.
Now the scary part: this video shows a slowed down 5 seconds—seconds!—of the stock market, and all of these trades are being made by computers.
There was nothing special about the Mastercard stock or 9:48 and 40 seconds on May 16 2012. This is happening all the time with every stock.
What’s the problem? To answer that question, we need to understand exactly what is going on. The price of a given stock is the same across all stock indices, or at least it should be, but there are occasionally differences between prices, which last a brief period of time. If you happened to notice that Mastercard was selling at $823.00 on the NYSE and $824.00 on the BOST, you could buy a stock on the NYSE and sell it on the BOST, and pocket a one dollar profit.
Your scheme wouldn’t work, however, because by the time you got around to executing those trades, it’s likely that the prices would have automatically adjusted toward each other. These price discrepancies often only exist for seconds at a time. That sort of free money rarely exists for long, and isn’t much to work with in the first place.
Unless, you are not a human, but a high frequency trading bot. That’s what you’re looking at in the video above: computers that spot differences in the price on a stock, and execute trades to take advantage of that difference in milliseconds.
Here is 10 milliseconds, less time than it takes for you to blink, of trading on Merck:
Markets are traditionally supposed to allocate capital efficiently, based on the beliefs of investors on what will succeed. Investing is supposed to be profitable if you place good bets on which stocks will perform well in the future. High frequency trading is an entirely different game. This has nothing to do with a stock’s future performance or the allocation of capital. This is all about having a computer so fast it can spot a price discrepancy before the exchanges do.
Put another way: high frequency traders are parasites, sucking money out of the economy, while providing no social benefit.
Any number of simple laws could effectively end high frequency trading. A tiny transaction fee would make the bulk of these trades not worth it. Insisting that a trade exist for half a second before it is completed, something that would affect exactly zero human traders, would wipe out a lot of trades that rely on millisecond differences. The only reason that we don’t have some kind of law against this is because the banks would hate it and 99 percent of the population wouldn’t notice, and the banks have huge political clout.
The solution is exposure. The more people are aware of this outrage, the harder it will be for it to continue.
And one more video. The rise of high frequency trading from 2007-2012: