The Arms Race in High Frequency Trading
There are two areas that were spared in the 2008 debacle: macro and high frequency trading. Macro funds on average were up ten percent or so last year because most of them skirted the edge of the major dislocations; their strategies focus on liquid instruments and are not oriented toward credit. High frequency trading did well because it thrives in an environment of high volatility and demand for liquidity, and 2008 was a hot house for both. Every year, people pile on to whatever strategy did well the previous year – this tendency is worth a book or two on its own – and so this year high frequency is destined to be the darling of the fund of funds.
But I think the days for high frequency trading are numbered. For one thing, high frequency trading is capacity constrained like few other strategies. The high frequency trader is basically a stand-alone market maker; he is sitting there to provide liquidity to others. And one way he provides it is to pull in the positions that others will shortly be demanding – thus the need for speed. If the footprint for high frequency traders gets too large, they become liquidity demanders themselves, and the gig is up. The Renaissances of the strategy will make their way through, but generally we will see a lot of shooting stars.
A second reason is that high frequency trading is embroiled in an arms race. And arms races are negative sum games. The arms in this case are not tanks and jets, but computer chips and throughput. But like any arms race, the result is a cycle of spending which leaves everyone in the same relative position, only poorer. Put another way, like any arms race, what is happening with high frequency trading is a net drain on social welfare.
In terms of chips, I gave a talk at an Intel conference a few years ago, when they were launching their newest chip, dubbed the Tigerton. The various financial firms who had to be as fast as everyone else then shelled out an aggregate of hundreds of millions of dollar to upgrade, so that they could now execute trades in thirty milliseconds rather than forty milliseconds – or whatever, I really can’t remember, except that it is too fast for anyone to care were it not that other people were also doing it. And now there is a new chip, code named Nehalem. So another hundred million dollars all around, and latency will be dropped a few milliseconds more.
In terms of throughput and latency, the standard tricks are to get your servers as close to the data source as possible, use really big lines, and break data into little bite-sized packets. I was speaking at Reuters last week, and they mentioned to me that they were breaking their news flows into optimized sixty byte packets for their arms race-oriented clients, because that was the fastest way through network. (Anything smaller gets queued by some network algorithms, so sixty bytes seems to be the magic number).
If we get out of the forest and look at what is going on, some questions come to mind. Does anyone really get a benefit in having the latency of their trade cut by milliseconds – except for the fact that their competitor is also spending the money to cut his latency? Should anyone care if a news event hits market prices in twenty-nine milliseconds rather than thirty milliseconds? Does it do anything to make the markets more efficient? Does it add any value to society?
We usually do not think about trading in terms of social value, but trading often does have social value, and it should. The objective of trading is to provide liquidity to the market, and to make sure that prices best reflect all available information – the usual efficient market argument we all grew up with. The solution? How about having everyone agree to standards in terms of hardware and related configurations. A high-frequency arms limitation treaty. We could call it HALT.