At seven minutes past one on the afternoon of Tuesday April 23 this year, a tweet from the AP news agency in Washington was published. It read "Breaking: Two Explosions in the White House and Barack Obama is injured." This was not true - the AP account had been hacked by a shady group calling itself The Syrian Electronic Army - but within milliseconds the tweet had been noticed and flagged by trading computers on Wall St.
Programmed to scan the net for words or phrases that might affect stock markets, the machines had seized on the tweet, noted the proximity of the words "Obama", "explosions" and "White House" and unleashed a torrent of trades. In seconds the Dow Jones had plunged 140 points and more than US$200 billion of capital had been wiped out.
A few minutes later the report was exposed as a hoax and the markets quickly returned to their pre-tweet levels. But, to many, the idea that one fake tweet could have such a huge impact on the financial markets was incredible. Who was running Wall St? Humans or machines?
If you thought "humans", you were woefully out of date. Over the past decade or so there has been a technological coup d'etat on the trading floor. The old "Masters of the Universe" - the Gordon Gekko types with their slicked-back hair and US$5000 ($6256) suits - have been superseded by unbelievably powerful computers capable of analysing vast amounts of data and buying or selling shares in the blink of an eye.
Today, if you visit a trading floor, instead of men screaming down the phone, you are more likely to see rows of studious-looking people (mostly still men) sitting at screens, quietly monitoring trades being carried out on their behalf by computers.
About 70 per cent of the orders to buy or sell on Wall St are now placed by software programs, and the studious-looking people are the mathematical geniuses responsible for writing these programs. It is the age of the algorithm.MATHEMATICIANS made their first forays into the financial world in the late '60s. University of California maths professor Edward Thorp published a book in 1967 called Beat the Market, in which he laid out what he claimed was a foolproof way of making money on the stock market, all based on a system he had previously devised to beat casinos at blackjack. The blackjack system had been so successful it had forced casinos to change their rules and Beat the Market proved to be even more groundbreaking. In 1974 Thorp founded a hedge fund called Princeton/Newport Partners and proceeded to make a killing.
At the same time, scientists' job prospects had nosedived. Since the 1969 moon landing, the US Government had diverted funding for science programmes to the war in Vietnam.
"A generation of physicists who had gone to graduate school left with their PhDs and entered a severely depressed job market," explains James Owen Weatherall, author of The Physics of Finance. Seeing how much money there was to be made on Wall St, many decided to move into finance.
In Britain, the fall of the Soviet Union led to an influx of Warsaw Pact scientists. In both cases, these scientists brought with them a new methodology based on analysing data and also a faith that, using sufficient computing firepower, it was possible to predict the market. It was the start of a new discipline, quantitative analysis, and the most famous "quant" of all was a shambling donnish maths genius called Jim Simons.
For those who know their physics, Simons is a living legend. A piece of mathematics he co-created, the Chern-Simons 3-form, is one of the most important elements of string theory, the so-called "theory of everything".
In 1982 he founded a hedge fund management company, Renaissance Technologies, whose signature fund, Medallion, went on to earn an incredible 2478.6 per cent return in its first 10 years, way above every other hedge fund on the planet, including George Soros' Quantum Fund.
Its success, based on a highly complex and secretive algorithm, continued in the noughties and over the lifetime of the fund, Medallion's returns have averaged 40 per cent a year, making Simons one of the richest men in the world, with a net worth in excess of US$10 billion.
Of his 200 employees, a third have PhDs, not in finance, but in fields such as mathematics, physics and statistics. Renaissance has been called "the best physics and mathematics department in the world" and, according to Weatherall, "avoids hiring anyone with even the slightest whiff of Wall St bona fides. PhDs in finance need not apply; nor should traders who got their start at traditional investment banks or even other hedge funds. The secret to Simons' success has been steering clear of the financial experts."NOT SURPRISINGLY, old-style traders hate the quants. Not only have they pushed them off the top of the trading tree, there is also a basic clash of cultures. They are not flash and, invariably, are rather awkward socially.
So what exactly do quantitative analysts do?
Patrick Boyle and Jesse McDougall run a hedge fund out of a town house in Islington, north London. When I meet them they are seated in a small room dominated by three computer screens. They start work at 7am and end about 11 at night. "We have computer screens in our kitchen and living room," says Boyle, 37. "So we can monitor the markets while having dinner and we can log in remotely if we are out in the evening." He shows me a chart tracking his fund's performance. The line doesn't dip when the rest of the market dips and rises faster than the FTSE index, which measures London Stock Exchange performance.
"We do it with maths," he says. "We buy stock market data and we analyse it. It's like weather forecasting - we can say that there is a 65 per cent probability that the market will be up between open and close, so we are able to have better than 50 per cent odds on short-term movements and over time if you call short term well, you can make money."
Who wrote the computer program they use? "I did," says Boyle. How do you write something like that? "Slowly."
The writing of the computer program may be slow but the speed of transactions is super fast. Some quants specialise in what is called High Frequency Trading (HFT), which involves large numbers of trades over very short periods. "In one millisecond the price could go up by one penny," says McDougall. "You do it thousands of times on hundreds of shares and you make money."BOYLE and McDougall's hedge fund doesn't do high frequency trades, so I talk to Simon Jones, 36, who ran the quants desk at a major bank until a few months ago.
"The guys and women who worked with me were the best of the best," he says. "They came from all over the world: from India, Russia and China." The job was intense and highly competitive. "Let's say I have noticed that the moment the Dow goes up the FTSE goes up," says Jones. "The first person to notice that and make a trade can make money but to do that means getting the data from New York to London and then getting my trading decision across the Atlantic and me buying my FTSE before anyone else does."
Speed is critical. Firms have actually started moving their servers nearer to an exchange to speed up connection times.
In 2010 a company called Spread Networks laid a new direct cable between New York and Chicago, which shaved a little bit more than 1000th of a second off the transmission time between stock exchanges.
For the opportunity to use a similarly fast tube between New York and London, Jones' old bank was asked to pay US$50 million. "It would have given us an advantage over others of about a six thousandths of one second."
This focus on the shortest of short-term gains has vastly increased volatility. "Warren Buffett owns shares in Coca-Cola and when they go down he says 'I'm holding on to them because I think they will go back up'," says Jones. "But the HFT guy, all he cares about is the next millisecond. And when too many people start panicking about the next millisecond that's when you have a crash."
May 6, 2010, was the perfect example of such a crash. So many shares were traded that day the online trading section of the New York Stock Exchange froze and between 2.30pm and 3pm the Dow Jones lost and then regained nearly US$1 trillion. In what became known as the "Flash Crash", shares in management consultancy firm Accenture plunged to just above zero. Apple shares went up to US$100,000.
"None of us knew what to do or what would happen next," says Dave Lauer, a quant who was working on a HFT desk that day. "It was terrifying."FOR LAUER, the Flash Crash was a wake-up call. "I started to see how the race to be fastest had left things in a very fragile state," he tells me. "I remember thinking, 'How will I explain to my future child what I do for a living?"' Lauer quit his job and last year told the Senate Banking Committee that High Frequency Trading had brought the market to crisis point.
The Flash Crash was partly caused by the HFT strategy of "spoofing"; making bogus offers to buy or sell shares to flush out the intentions of rivals. On the day, an astonishing 19.4 billion shares were traded, more than were traded in the entirety of the 1960s, but hundreds of millions of them were never actually sold; they were merely held for a few thousandths of a second as traders tested the waters.
Isn't there something wrong with a system that promotes so much volatility to the benefit of no one except a handful of hedge funds? Can it be a meaningful investment of time and technology? Buffett's business partner, Charlie Munger, has described HFT as "basically evil". "I think it is very stupid to allow a system to evolve where half of the trading is a bunch of short-term people trying to get information one millionth of a nanosecond ahead of somebody else," he said this year. HFT is certainly of no clear benefit to everyday investors - savers in pension funds and life policies.
The quants I meet do voice some doubts. "Some of the guys who come from pure science and maths backgrounds are used to solving a problem and it works," Patrick Boyle says. "They think they can find a formula that will perfectly describe how the market moves. That is the philosopher's stone - it is utterly impossible." The danger is that in only seeing numbers and patterns the human dimension is forgotten.
After 16 years in the city, Simon Jones plans to travel. "A quant can earn up to seven figures," he tells me, "but sometimes I do wonder whether I contributed positively to society.
"I was working with the best of the best ... If there was a pay bonus structure similar to what we had in the City for curing cancer, we'd have found a cure for cancer."
I find that sad and a little bit frightening. So, I ask, quants: good or bad? Jones looks at me and says, "Humans just found a new way of being greedy."
