What does the market look like in the era of electronic trading? Do we carry in our heads an out-dated picture of the market?
I’ve been reading Flash Boys: Cracking the Money Code, the latest chart-topping book by Michael Lewis, who has specialised in writing gripping non-fiction accounts of the excesses of Wall Street that read like financial thrillers. In this book Lewis tells the story of how high-frequency traders have managed to rig the market by improving on connection speeds to the stock exchange’s computers by tiny fractions of a second. When an common-or-garden stock broker places an order to buy shares, the computer of the HFT firm manages to detect it and speeds ahead to buy up the shares being offered and sells them to the clueless broker at a slight hike, making a small profit from the difference. In this way the market that mom-and-pop brokers see on their screen is not the one that they actually end up buying into. The hero of Lewis’s book is Brad Katsuyama, a trader for Royal Bank of Canada, a backwater firm in Wall Street’s eyes, who begins to investigate the murky world of HFT because he can’t understand why the market seems to be able to anticipate whatever orders he tries to execute. The market that he interacts with on his trading screen develops a seemingly uncanny, godlike ability to predict his every move.
The story that Katsuyama uncovers (and that Lewis dishes up in gripping detail) is one in which the HFT firms go to extreme lengths to gain a miniscule time advantage over their rivals. Not content with renting office space for their under-the-radar firms as close as possible to the main lines coming out of the exchanges, they pay to locate their computers as close as possible to the matching-engine computers run by the exchanges themselves, and help create new electronic exchanges with built-in kickbacks designed to attract particular kinds of orders that they can then exploit. The book opens with the extraordinary tale of the construction work on a secret $3 billion, 827-mile fibre optic cable that runs in a near straight line all the way from Chicago to New Jersey, involving some insane tunnelling through mountains, all in order to exploit price differences between the two exchanges and beat their rivals by reducing the time it takes for information to travel from 17 to 13 milliseconds (a millisecond is one thousandth of a second).
Much of what Lewis describes is jaw-droppingly outrageous, but not, it turns out, completely illegal: brokers are duty bound to get the best prices for their customers, and so are obliged to fall into the traps set by HFTs. The various investigations and prosecutions that are currently under way are therefore not likely to have much effect. Even if the regulators manage to clamp down on some of the specific ruses employed by these firms, it has always been the case that speed gains you an advantage in the market. The classic case cited by historians of the stock market is that of Nathan Rothschild used his family’s network of agents to receive news of the outcome of the battle of Waterloo in advance of everyone else (including the British government), enabling him to make a killing on the London markets before the story became public. Despite strong versions of the Efficient Market Hypothesis that insist that the market already knows any relevant information in advance and factors it into the current price, it is arguable that insiders have always rigged the market in their favour. It has never been the free, transparent and perfectly functioning market that its champions have claimed. In the exhibition there are cartoons from the Gilded Age in America, such as this one showing Wall Street as a merry dance in which the bulls chase the bears, and the lambs—the clueless amateur outsiders—are ritually fleeced. Insiders have also always tried to get privileged access through being connected to the exchanges more closely or more quickly. Prior to the invention of the stock ticker in the late 1860s, messenger boys would run from the stock exchange to the neighboring brokerages in the Wall Street district to deliver the latest prices, and the quickest ones soon gained a reputation. From the late 1880s onwards the New York Stock Exchange and the Chicago Board of Trade tried to claim that they owned the prices produced on the floor of the exchange, and mounted legal actions to deny ticker access to the bucketshops (low-rent versions of brokerages that were not much much than betting shops). The bucketshops launched their own legal challenges to what they saw as an unfair monopoly on the part of the NYSE and CBoT. At the same time, however, they also resorted to illegally tapping the wires coming out of the exchanges, and they paid messenger boys to peer in through the keyhole and climb up drain pipes to peer in the windows that had been soaped over to try and hide the action on the floor from prying eyes.
In theory the invention of the stock ticker meant that trading could now take place instantaneously anywhere: there’s a marvellous Charles Dana Gibson cartoon from 1903 that shows the exasperated family of a market-obsessed patriarch who has set up a ticker under the tree of their holiday cottage. The idealised market that the stock ticker in principle created was a completely frictionless one, in which the location, identity and human foibles of the individual buyers and sellers were no longer an issue. In neoliberal economic dogma, the stock exchange was heralded as the ultimate embodiment of the market as a divinely efficient information processing machine. In practice, however, the market did not always run smoothly. Leaving aside the suspicion that the market was being rigged by insiders, there was often a small but crucial delay on the prices printed on the ticker, particularly at frenetic moments in the midst of financial crisis. The sheer volume of transactions caused a backlog of what could be reported on the tape, sometimes by as much as half an hour. By the time that traders at a distance had received the latest prices over the wire they were out of date, and their orders might likewise encounter delays before they finally reached the floor of the exchange. Those who could afford offices close to the exchanges, and who rented their own direct telegraph lines, found themselves at a considerable advantage over the less well sited or equipped second-rate brokers and bucketshops. Sometimes the bucketshops worked their scams by causing a deliberate delay in the wire service feeding the office, ensuring that they could be one step ahead of the punters; in other cases, the wires reached no further than the carpet at the edge of the room, and weren’t connected to the exchanges at all, with all the prices just being made up on the spot by the dodgy employees.
The problem with Lewis’s book (nailed by Andrew Ross in this review in the Guardian) is not merely that it focuses on a few bad apples, in its portrait of HFT as a temporary perversion of the otherwise smooth running of the stock market. Its real flaw is that it suggests that the only ones smart enough to do anything about the problem are insiders like Katsuyama, whose free market solution is not better and more democratic regulation but the creation of a new electronic exchange that levels the playing field and undermines the HFT firms’ advantage. For Lewis, the regulators are dinosaurs who just don’t get it, and he is no doubt correct that many of the developments that have been taking place have been unseen by those meant to watch over the market. For example, the official report into the Flash Crash of 2010 (when the Dow dropped without warning by 9% within minutes) had to rely on the records kept by the exchanges that were detailed down to the second. The problem, however, was that transactions in the market now happened at the speed of microseconds, requiring a completely different perception of time: “There are one million microseconds in a second. It was if, back in the 1920s, the only stock market data available was a crude aggregation of all trades made during the decade. You could see that at some point in that era there had been a stock market crash. You could see nothing about the events on and around October 29, 1929.” If it was impossible to unpick exactly what taken place, it makes it far harder to determine what checks and balances could be introduced into the system to do something about it.
What interested me most about Lewis’s book, however, is the way that the market is becoming increasingly invisible and unimaginable, and this makes to harder to work out how to control it. For example, Lewis decribes how Brad Katsuyama and his team “were building a menal picture of the financial markets after the crisis [of 2008]”:
The market was now a pure abstraction. It called to mind no obvious picture to replace the old one that people still carried around in their heads. The same old ticker tape ran across the bottom of the television screens—even though it represented only a tiny fraction of the actual trading. Market experts still reported from the floor of the New York Stock Exchange, even though trading no longer happened there.
The problem now was that “no mental picture existed of the new financial market. There was only this yellowing photograph of a market now dead that served as a stand-in for the living.” Many of the contemporary artits in the “Show Me the Money” exhibition have tried either to work out what a new mental picture of the market should look like, or to comment on the difficulties of representing an increasingly abstract and inpenetrable market that now seems to take place inside machines, in the absence of any visible human interaction. A photograph by the Beate Geissler and Oliver Sann, for example, shows a trading desk in the office of a high frequency trading firm in the Willis Tower in Chicago. Save for a few sheaves of paper, Geissler and Sann’s image is devoid of traces of human presence. In the book of the exhibition, the journalist Justin Fox comments on the people-less landscapes of contemporary high finance, while Andy Haldane, the incoming Chief Economist at the Bank of England (and recently included in Time magazine’s list of world’s 100 most influential people), reminds readers that human relationships and trust are still central to the workings of the financial machine.
Although Lewis conjures up by way of contrast a misty-eyed view of the golden age before HFT ruined things (a view that readers of Liar’s Poker, Lewis’s account of the go-go years of bond trading at Salomon Brothers in the late 1980s, are under no illusion ever really existed), he does suggest an interesting analysis of the problem created by HFT as one of unfair privilege: “The U.S. stock market now has a class system, rooted in speed, of haves and have-nots. The haves paid for nanoseconds; the have-nots had no idea that a nanosecond had value. The haves enjoyed a perfect view of the market; the have-nots never saw the market at all. What had once been the world’s most public, most democratic, financial marker had become, in spirit, something more like a private viewing of a stolen work of art.” Of course, Lewis only invokes class privilege as an analogy to explain the unfairness of some traders being able to game the system. But his comparison of HFT with the private access to treasured works of art is intriguing (albeit because they are stolen rather than just bought by the mega-rich), because it raises the question of how far the state should be able to intervene in the free market—whether of derivates or art treasures—for the good of the people.