Present Shock: When Everything Happens Now (24 page)

Behavioral finance is the study of the way people consistently act against their own best financial interests, as well as how to exploit these psychological weaknesses when peddling questionable securities and products. These are proven behaviors with industry-accepted names like “money illusion bias,” “loss aversion theory,” “irrationality bias,” and “time discounting.” For instance, people do not borrow opportunistically, but irrationally. As if looking at objects in the distance, they see future payments as smaller than ones in the present—even if they are actually larger. They are more reluctant to lose a small amount of money than gain a larger one, no matter the probability of either event in a particular transaction. They do not consider the possibility of any unexpected negative development arising between the day they purchase something and the day they will ultimately have to pay for it. Present shock.

Banks craft credit card and mortgage promotions that take advantage of these inaccurate perceptions and irrational behaviors. Zero-percent introductory fees effectively camouflage regular interest rates up to 30 percent. Lowering minimum-payment requirements from the standard 5 percent to 2 or 3 percent of the outstanding balance looks attractive to borrowers.
²⁵
The monthly payment is lower! However, the additional amount of time required to pay off the debt at this rate will end up costing them more than
triple
the original balance. It is irrational for them to make purchases and borrow money under these terms, or to prefer them to the original ones.

It proved just as irrational for American banks to depend on this domestic and international sleight of hand for so long. Eventually debtor nations began either defaulting on their loans (Argentina) or insisting on being included in the value chain (China). Lender nations soon found themselves in debt and nursing negative trade balances. Right around the same time, American consumers began to realize they could not support their own rate of consumption, even when leveraged by credit. Expansion seemed to be threatened, as the available surface area for new transactions had proved limited. Where to grow?

Instead of innovating in the real world, banks turned to the financial instruments themselves. Interest had worked just fine for centuries, squeezing wealth out of money itself. Could the formula be repeated? Could more time be packed into interest?

Interest had given money the ability to generate wealth through lending. Vastly simplified, interest is just a way of expressing money over time. Lend some money and over time that money multiples. Whether it is a bank lending money to a business, an investor buying some shares of stock, or a retiree buying a bond, everyone expects a return on his investment over time.

As these first-order investments began to slow down—at least compared with the rate at which the economy needed to grow—everyone began looking for a way to goose them up. That’s where derivatives came in. Instead of buying shares of stock or whole bonds or mortgage notes, derivatives let investors bet directly on the changing value of those instruments over time. Instead of buying actual stocks and bonds, investors buy the right to buy or sell these instruments at some point in the future.

Most of us understand how derivatives let investors make bigger bets with less money up front. Investors can purchase the options on a stock for a tiny fraction of the cost of the stock itself. They only pay for the actual stock once they exercise the option. Hopefully, the shares are worth a lot more than the strike price on the option, in which case the investor has made money.

But on a subtler level, the investor hasn’t merely leveraged his money; he has leveraged time. The option is a financial spring-loading mechanism, packing the future’s price fluctuations into today’s transactions. The investor is no longer investing in a company or even its debt, but rather in the changing value of that debt. In other words, instead of riding up the ascending curve of value, derivatives open the door to betting on the rate of change. Value over time, over time.

In the effort to compress time ever further, this process can be repeated almost ad infinitum. Traders can bet on the future price of derivatives—derivatives of derivatives—or the future price of those, or even just the volatility of price swings. At each step along the way, the thing being invested in gets more abstract, more leveraged, and more time compressed. In the real world, value can’t be created fast enough to keep up with the rate of expansion required by interest-driven currency, so it instead gets compressed into financial instruments that pack future value into present transactions.

Regular people end up compressing time into money the same way. Back when I still believed I could afford to purchase a three-bedroom apartment in New York City, the real estate agent showed me residences far out of my price range. She explained to me that I could qualify for an ARM, or an adjustable-rate mortgage, with a special introductory interest-only rate for the first five years. This way, I would have a monthly payment I could afford, even though I wouldn’t be working toward reducing the principal. After the first five years, the loan would adjust to a more normal rate—one much greater than I could afford. She told me that wouldn’t be a problem, since at that point I could simply refinance with another five-year ARM.

The idea was that that apartment would be worth more by the time I needed to get a new mortgage, so I would be able to refinance at a higher assessed value. As the market kept going up, the meager portion of the apartment I owned would be going up in value as well, giving me the collateral I would need to refinance. Each time, the cash amount I need to finance remains the same, but the percent of the apartment that debt represents goes down. So, in theory, the more the price of the apartment goes up, the more I own, and the more easily I can refinance.

Of course, that didn’t happen. Luckily for me I didn’t buy the apartment (or, rather, I didn’t buy the mortgage for that apartment). But the hundreds of thousands of Americans who did accept similar bargains ended up in big trouble. Instead of increasing, the value of the homes sunk below the amount that was owed on them. As of this writing, 31 percent of all residential properties with mortgages are under water, or what industry analysts call negative equity.
²⁶
Owing more on a thirty-year mortgage than one’s house is currently worth is just another way of saying present shock.

A few traders did see the writing on the wall and understood that the housing market had become too dependent on these temporally compressed lending instruments. Famously, even though they were selling packaged loans to investors and pension funds, Goldman Sachs determined that the financing craze was unsustainable and began betting against the mortgages through even more derivative derivatives called credit default swaps. When it came time for the company on the other side of those default swaps, AIG, to pay up, only the US government could print enough money to bail them out.
²⁷

By today’s standards, however, Goldman’s successively derivative bets seem almost quaint. Their investing decisions were still based in what they saw as the likely future of an unsustainable system. Their contribution to the tragedy from which they hoped to benefit notwithstanding, they were making a prediction and making a bet based on their analysis of where the future was heading. Temporally compressed though it may be, it is still based on making conclusions. Value is created over time. It is a product of the cause-and-effect, temporal universe—however much it may be abstracted.

A majority of equity trading today is designed to circumvent that universe of time-generated value altogether. Computer-driven or algorithmic trading, as it is now called, has its origins in the arms race. Mathematicians spent decades trying to figure out a way to evade radar. They finally developed stealth technology, which really just works by using electric fields to make a big thing—like a plane—appear to be many little things. Then, in 1999, an F-117 using stealth was shot down over Serbia. It seems some Hungarian mathematicians had figured out that instead of looking for objects in the sky, the antiaircraft detection systems needed to look only for the electrical fields.
²⁸

Those same mathematicians and their successors are now being employed by Wall Street firms to hide from and predict one another’s movements. When a bank wants to move a big quantity of shares, for example, it doesn’t want everyone to know what it is doing. If news of a big buy leaked out before the big buy could be completed, the price may go up. To hide their motions, they employ the same technique as stealth planes: they use algorithms to break their giant trade into thousands of little ones, and do so in such a way that they look random. Their sizes and timing are scattered.

In order to identify this stealthy algorithmic movement, competing banks hire other mathematicians to write other algorithms that monitor trading and look for clues of these bigger trades and trends. The algorithms actually shoot out little trades, much like radar, in order to measure the response of the market and then infer if there are any big movements going on. The original algorithms are, in turn, on the lookout for these little probes and attempt to run additional countermoves and fakes. This algorithmic dance—what is known as black box trading—accounts for over 70 percent of Wall Street trading activity today.

In high-frequency, algorithmic trading, speed is everything. Algorithms need to know what is happening and make their moves before their enemy algorithms can react and adjust. No matter how well they write their programs, and no matter how powerful the computers they use, the most important factor in bringing algorithms up to speed is a better physical location on the network. The physical distance of a brokerage house’s computers to the computers executing the trades makes a difference in how fast the algorithm can read and respond to market activity.

As former game designer Kevin Slavin has pointed out in his talks and articles,
²⁹
while we may think of the Internet as a distributed and nonlocal phenomenon, you can be closer or farther from it depending on how much cable there is between you and its biggest nodes. In New York, this mother node is fittingly located at the old Western Union Building on 60 Hudson Street. All the main Internet trunks come up through this building, known as a colocation center, or carrier hotel. Its fiber optic lines all come together in a single, 15,000-square-foot “meet me” room on the ninth floor, powered by a 10,000-amp DC power plant.

If a firm owns space anywhere in that building, its computers are sitting right on the node, so its algorithms are operating with a latency of effectively zero. Algorithms running on computers all the way down on Wall Street, on the other hand, are almost a mile away. That adds up to about a two-millisecond delay each way, which means everything. A fast computer sitting in the carrier hotel can see the bids of other algorithms and then act on them before they have even gone through. The purpose of being in close isn’t simply to front-run the trade but to have the ability to fake out and misdirect the other side. Algorithms don’t care what anything is really worth as an investment, remember. They care only about the trade in the present.

In a virtual world of black box trading where timing is everything, getting closer to the “meet me” room on the ninth floor of 60 Hudson Street is worth a lot of money. Firms are competing so hard to position their computers advantageously that the real estate market in the neighborhood has been spiking—quite unpredictably, and only explained by the needs of these algorithms for quicker access. Architects are busy replacing the floors of buildings with steel in order to accommodate rooms filled with heavy servers. Both the real estate market and the physical design of Lower Manhattan are being optimized for algorithms competing to compress money into milliseconds. It is becoming a giant microchip or, better, a digital stopwatch.

When the only value left is time, the world becomes a clock.

LIVING IN RAM

When an economy—or any system, for that matter—becomes so tightly compressed and abstracted that only a computer program can navigate it, we’re all in for some surprises. To be sure, algorithms are great at increasing our efficiency and even making the world a more convenient place—even if we don’t know quite how they work. Thanks to algorithms, many elevators today don’t have panels with buttons. Riders use consoles in the lobby to select their floors, and an algorithm directs them to the appropriate elevator, minimizing everyone’s trip time. Algorithms determine the songs playing on Clear Channel stations, the ideal partners on dating websites, the best driving routes, and even the plot twists for Hollywood screenplays—all by compressing the data of experience along with the permutations of possibility.

But the results aren’t always smooth and predictable. A stock market driven by algorithms is all fine and well until the market inexplicably loses 1,000 points in a minute thanks to what is now called a flash crash. The algorithms all feeding back to and off one another get caught in a loop, and all of a sudden Accenture is trading at $100,000 a share or Proctor & Gamble goes down to a penny.
³⁰
Ironically, and in a perfect expression of present shock, the leading high-frequency trading exchange had a high-profile flash crash on the same day it was attempting to conduct its own initial public offering—and on the same day I was finishing this section of the book. The company, BATS Global Markets, runs a stock exchange called Better Alternative Trading System, which was built specifically to accommodate high-frequency trading and handles over 11 percent of US equities transactions. Their IPO was highly anticipated and represented another step in technology’s colonization of the stock exchange.

BATS issued 6 million shares for about $17 each, and then something went terribly wrong: their system suddenly began executing trades of BATS stock at three and four cents per share.
³¹
Then shares of Apple trading on the BATS exchange suddenly dropped 10 percent, at which point the company halted trading of both ticker symbols. Embarrassed, and incapable of figuring out quite what happened, BATS took the extremely unusual step of canceling its own IPO and giving everyone back their money.