The Auto-Pilot Trading Desk

In the future the car in your oncoming lane might be driven by software working over WiFi. Is there any difference with today's algorithmic trading firms?

What’s scarier: Google’s driverless car or a financial market full of robo-traders?

At the beginning of the year, the research firm IHS Automotive published a study entitled “Emerging Technologies: Autonomous Cars-Not If, But When.” In this report, they forecast “total worldwide sales of self-driving cars will grow from nearly 230,000 in 2025 to 11.8 million in 2035.” In a May 28 article entitled “Look Ma, no hands: Google to test 200 self-driving cars”, the Los Angeles Times reports Google goal in developing self-driving car technology is to “improve road safety and help people who can’t drive.” Google will most likely not get into the car manufacturing business, but instead build up a treasure trove of patents to license to the truly automobile industry.

Just as in the early days of the automobile when cars competed on the road with horse-drawn carriages, there will be a period of time when SDCs will compete for road space with conventional — hands on? –automobiles. Also, just as there were accidents involving spooked horses, runaway carriages and newbie drivers, there will probably be accidents involving cars driven by distracted drivers and those under the control of software. No matter how good the technology, as long as the variable of unpredictable animal or human behavior is involved, there will always be a chance of something occurring that the technology designers didn’t take into account.

Back in 1936, a third year law student at Duke named Richard Milhous Nixon wrote a paper entitled “Changing Rules of Liability in Automobile Accident Litigation.” The future President of the United States observed that rather than just extending “horse and buggy law,” some judges were developing new rights and remedies to adjust to the new mode of transportation. I expect to see a similar evolution in legal precedent as self-driving cars take to the roads. History has shown that there is a lag between new technologies and the creation of laws and regulations to deal with those new technologies.

In the brief history of automated trading systems we’ve seen “accidents” involving the interactions of automated and manual human-based trading. The “Flash Crash” was triggered by an algorithm that didn’t understand a thinly traded market. The “Hash Crash” was the result of HFTs reacting to humans trading on false news on Twitter of an explosion at the White House. In the evolution from horse and buggy to automobile and eventually to the self-driving car, infrastructure and the policing of that infrastructure plays a key role. Dirt roads gave way to paved roads, autobahns and superhighways are constantly being improved to handle ever faster speeds. Laws and regulations surrounding road and vehicle safety continue to evolve as well.

The evolution of trading has seen paper and voice give way to electronic deal-making, and copper give way to fiber and now microwave and lasers. Globally, regulators continue to wrestle with how to address the use of increasingly faster technology the failures of which can have consequences for both automated and manual participants alike. Books like “Flash Boys” make the proposition that certain strategies that can only be accomplished through the use of high-speed computers and communications should be declared illegal because of the advantage they give their users over market participants that don’t have the same ability.

There are primary differences in the evolution of transportation and the evolution of trading, of course. The first is that lives can be lost instantly in a car wreck. Lives have been ruined and even lost to acts of violence due to market downturns, but the number of market fatalities is minute when compared with traffic fatalities.

A much more subtle difference exists. Regardless of whether the vehicle is horse-drawn, motor-driven, manually controlled or controlled by software, the purpose of transportation remains the same: to get from point A to point B. The mode of transportation does not change that purpose. The primary purpose of the capital markets is to provide incentive and reward for investing in innovation outside of the markets themselves, in the products and services of the companies whose shares are being traded.

Another purpose is to reward speculators and market makers for the risks they take and the liquidity they provide. A market made up of competing algorithms primarily gives incentives and rewards to those participants who can best outmaneuver their adversaries in trading. Share prices on such a market may be the outcomes of so many dodges and parries between computer programs disconnected from reality. The fall or rise of a company’s share price may have nothing to do with the company’s actual value.

That is scary indeed.
[End of column]

Robert Stowsky is partner and co-founder of the consultancy Brookpath Partners.