Flash Boys: A Wall Street Revolt, Michael Lewis' new book, has placed a rather large cat amongst the Wall Street pigeons, most notably in the secretive world of high-frequency trading. From the eagerness of policymakers to jump onto a trending bandwagon to the chatter on online industry boards proposing a bonfire of well-thumbed Liar's Poker copies, there is a lot of high-frequency chatter flying around right now between detractors and apologists.
For better or for worse, high-frequency trading is a part of our financial landscape today. Ever since the geeks emerged from their garages and academic burrows in the last three decades (confession: I am one of them), mathematics and technology have fused to become integral to the financial markets. Today, the precision of numbers underlies the models that investors use to read the financial tea leaves. At the same time, raw processing power -- growing exponentially with every technological advance -- allows the crunching of more and more data. Together, they feed off each other, trying to suck in as much information as possible, analyze it all and create models that mimic the complexity of the real world.
This is an economic arms race. High-frequency trading is the latest evolution, where the soldiers -- once human traders -- have been replaced by drones. Whether this is predatory, beneficial or just an agnostic consequence of progress is an important question. But if we are to have an informed debate, let us at least then remove some of the myths from both sides in today's high-frequency hyperbole.
1. The hunt for speed is a new phenomenon
High-frequency trading and its obsession with speed may have burst into the limelight only in the last few years but the first computer traders began trading in the early 1970s. As soon as someone invented the computer, it was inevitable that someone else would see if it could be used to make money.
However, the search for speed goes back far further. In the 18th century, Japanese rice farmers were selling future deliveries of rice in return for money today -- a futures contract in today's derivatives parlance. This gave rise quickly to the first derivative speculators. One of these, Munehisa Homma, was perhaps the most successful financial trader in history, making as much as $10 billion in some years.
His edge was two-fold. First, he identified clear patterns in the rice markets that he codified into a primitive form of technical trading and used to make money from his peers. Second, he understood the value of getting information fast. Legend has it that he established a network of couriers, stationed every four miles between the rice market and his home -- a distance of nearly 500 miles -- to make sure he got his information as fast as possible. It was the human analogue of fibre-optic broadband.
In making money, information is an advantage. The more information, the greater the probability of deducing future outcomes. And that comes down to reach and speed. Reach wider and if others are already there, then get there faster. Dart in and out, taking little bits of the cherry and minimising losses. And do all this as often as possible. That goes back a long way indeed.
2. High-frequency trading is bad for the financial markets and investors
Technology is a double-edged sword. For every benefit, it also brings new risks. The first car accident took place in 1896, when Bridget Driscoll was hit by a car going at 4 mph -- little over a decade after one Karl Benz had invented the commercial automobile. The judge at the inquest apparently remarked that he hoped this sort of nonsense would never happen again. It did -- many times.
High-frequency trading has risks. I worry, for example, about the myopic horizon it accentuates and the ability to create large herds of automatons that could engineer rapid booms and busts. But to demonise it as wholly bad is also unfair.
Investors, in fact, have a lot to thank the field for. The drive to make money out of getting information faster and faster led to the creation of a whole infrastructure that every investor -- big or small -- benefits from today. Faster orders mean less waiting around for all manners of financial transactions. The competitive rat race between providers means that they have sacrificed margin for volume -- a financial Walmart strategy. Thus, costs have come down tremendously for both individual investors and large institutions when they buy or sell stocks, bonds, commodities and other flights of fancy. And all that volume of orders means that the market is flooded with plenty of phantoms who provide the liquidity we all crave, even if we gripe that it could always be better.
That means investors get to transact more easily whenever they want, keep more of their hoped for returns, have smaller losses when it goes wrong and better performance overall (if they're good). If you're an individual investor, your trading account rests a little easier and if you're an institution, your pensioners and clients get a better deal.
While some commentators may yearn for a return to some fabled "golden age" when the system was trustworthy, genuinely helpful and honest, they would be wise to examine the evidence. It was even more opaque, riddled with expensive fees and limited choices, and far more exclusionary to large swathes of society.
3. The human is irrelevant in financial markets today
You could be forgiven for thinking that computers have taken over the financial markets and that people are irrelevant. In fact, nothing could be further from the truth.
Financial markets are not static entities. The vast majority of money in the markets is still controlled by all too human participants. The managed futures industry -- a collection of hedge funds that often use high-frequency strategies -- is about $330 billion in size. You could increase that by a factor of 10 if you wanted, but it still pales into insignificance compared to the $80 trillion managed by fund managers globally.
Financial markets are thus little more than collective nouns for the actions borne of the hope, greed and fear of countless human participants. They portray emotion as much as any underlying economic reality and the volatility we observe is driven by the competition between these emotions. Most high-frequency traders look for regular patterns or trends to exploit. They are in effect, trying to mimic this human behavior as well. They do make up a disproportionate amount of the volume in the financial markets, about 50 percent of equity trades according to recent estimates, for example. But even this only means they accentuate human emotions, not replace them.
4. High-Frequency traders know how to manipulate financial markets
All models are broken. And high-frequency traders are no exception to the rule.
Their supposed manipulation of financial markets is based on their intricate models, that approximate patterns in human behavior. Their success reflects that we are often predictable, but human behavior is also dynamic. Changes in the environment can soon render models useless. Additionally, the competition between peers means that any additional informational edge can quickly erode as others pile in to take advantage. This is the reason why the most successful in the business are constantly spending vast fortunes on better technology and more research.
Additionally, as humans, we have a propensity to fall in love with precision, no matter how illusory it may be. I recall a well-known fund manager, whose currency model worked wonderfully and made them a fortune for a decade. As its performance began to erode, they decided to update the model. The original designer had left long ago and so, a new team of Ph.D. experts was set to the task. When they took the model apart, they realised -- likely with horror -- that they had been misinterpreting the results all this while. When it said buy, they mistook it as a signal to sell and when it said sell, they mistook it as a signal to buy. Idiocy and luck had masqueraded as one of their most successful models all this while.
We forget -- models are also built by people and subject to all our errors. Garbage in will inevitably mean garbage out.
5. High-frequency traders make lots of money
Given the headlines, it's easy to believe this. The myth of the powerful geek-god is compelling. Like any field of success, there are legendary managers with enviable track records and vast fortunes to buttress. Virtu Financial, whose IPO was recently pulled, claimed to have made money almost every day for five years running.
However, that does not mean they always get it right, for the reasons we discussed above. The last few years have been proof of their fallibility. The Barclay CTA Index, for example, which tracks high-frequency funds, shows they have collectively lost money in five out of the last six years. The reason is simple: since 2008, policymakers have become all important to financial markets, who now dance to the tune of central bankers. The result is new patterns of behavior and an unpredictability of outcomes, which their models have been unable to capture. It's not a great result for these masters of the financial universe.
High Income Percentage: 11.7% GDP Growth: 6% Patents: 6,468 Source: Credit Donkey
High Income Percentage: 11.3% GDP Growth: 10.2% Patents: 1,197 Source: Credit Donkey
High Income Percentage: 20% GDP Growth: 4% Patents: 6,468 Source: Credit Donkey
High Income Percentage: 12.7% GDP Growth: 5.6% Patents: 3,597 Source: Credit Donkey
High Income Percentage: 16.1% GDP Growth: 5.1% Patents: 6,252 Source: Credit Donkey
High Income Percentage: 11.1% GDP Growth: 7.45% Patents: 3,113 Source: Credit Donkey
High Income Percentage: 10.9% GDP Growth: 11.4% Patents: 2,460 Source: Credit Donkey
High Income Percentage: 23.6% GDP Growth: 9% Patents: 1,790 Source: Credit Donkey
High Income Percentage: 16.8% GDP Growth: 9.1% Patents: 4,537 Source: Credit Donkey
High Income Percentage: 25.1% GDP Growth: 18% Patents: 10,256 Source: Credit Donkey
Follow Bob Swarup on Twitter: www.twitter.com/BobSwarup