High Frequency Trading: A world that lives 0.03 seconds ahead of you

A non-descript building

Processing financial data

In Jersey City, just across the Hudson River, stands an unmarked building hidden in the urban jungle. Although inconspicuous from the outside, it boosts strong security on the inside with many checkpoints, mantraps and high-tech identity verification.

If you would get past all the security, you could enter the main section of the building: a dark room, full of giant unmarked cages, vigorously buzzing away. Looking down one corridor, it stretches for a hundred yards. The entire complex is the size of 3 football fields. It is cold in here, powerful ventilators are droning away, sucking up all the heat.

If you owned one of these cages, you could enter it, and find yourself surrounded by a vast array of computers. Here, you can identify yourself using the biometric hand scanner. You would then be allowed to access the activity of these machines: 10’s of thousands of shares are traded by these machines every second. Sold, bought, and resold. But why? And why here?

Rise of the Machines

The purpose of this immense facility is so that computers can be close to each other, minimizing the distance that information has to travel. As a result, shares can be sold ever faster. (This is called co-location.)

Each computer can process a transaction in a micro-second or less, a millionth of a second. In other words, each computer can process 400,000 transactions per second.
People aren’t needed here.

These computers trade on the stock markets, and it is estimated that 13% of all shares in America are traded here. That makes it the third largest stock market in the world, bigger than the London stock exchange. 21 Billion dollars were traded here last year alone.

And none of this was here 2 years ago.

Electronic spies that beat time and space

These computers are set up to detect all activity in the market. As soon as a regular share trader tries to make a transaction, the computers detect it, using something that in the jargon of the trade is called a ‘high-frequency algorithm’. Because the computers are so close to the center of trading, they can cut ahead of the sale. They issue and cancel the sale almost simultaneously, bullying slower investors to giving up profits. The computer will do this all day long, in a systematic way, making money out of the original trader’s intention.

“It’s become a technological arms race, and what separates winners and losers is how fast they can move,” said Joseph M. Mecane of NYSE Euronext, which operates the New York Stock Exchange. “Markets need liquidity, and high-frequency traders provide opportunities for other investors to buy and sell.”

Even if they make a modest loss in the process, they still profit. They benefit from the competition between stock exchanges that pay small fees (called liquidity rebates, usually a quarter of a percent of a share) collected by the biggest and most active traders. Spread over millions of trades each day, these small fees amount to huge sums of money.

The problem is that these trades simply happen for the sake of the trade (because of the payout made for the act of the trade). That means that the trader doesn’t take a position, or rather is on either side of the trade, effectively taking opposing positions regardless of market intelligence. As a result, it skews the markets and makes them opaque. The investors argue that they provide liquidity to the market by being on either side, but the controversy is that the way they do this creates murky prices, undermines the fundamental principles of the exchange and under extreme circumstances (which happen often) leads to volatility or at the very least badly informed decisions that destabilize the markets.

In the last 2 years, this type of non-human trading has risen to almost 70% of all share transactions in America.

The billions of dollars these ‘predatory traders’ earn are directly at the expense of such traders as pension fund managers. It spies on what they do, reads their movements on the stock market and instantly runs ahead of their activity, undercutting their selling behavior.

Is knowing information 0.03 milliseconds in advance of everyone else still insider knowledge?

Making Money? Check! But adding value?

Legally speaking, this is not insider trading. Insider trading is the practice of trading based on non-public information. It has been made illegal, because it gives the trader an ‘unfair advantage’ and the power to ‘disrupt the markets’ and ‘game the system’.

The big firms who run these computer programs claim that they are not engaged in insider trading, that they simply trade faster. In truth, they can read your actions, and cut ahead of you by a millisecond, and make the trade in your place, then sell it back to you at inflated prices.

These companies are making huge sums of money based on innovative technology that costs billions to put in place. But isn’t that the American way?

To understand how this changes society as a whole, it is important to remember how the stock exchange came into being.

Making capital available to trade

Stock exchanges were set up to raise new shareholder capital for growing companies. Regulations were put in place to lower cost and offer a fair and equitable capital system which has the confidence of all participants because everybody has a fair chance of participating in the market. In order to keep this system viable, practices such as insider trading were made illegal, so that investors would remain in the stock exchanges and trade in a transparent ways. As a result, a large number of people poured their money in stock exchanges, investing in the future of business.  Then, in 1998, the Securities and Exchange Commission authorized electronic exchanges to compete with marketplaces like the New York Stock Exchange. Their intention was to open markets to all of us, so that we could participate armed with nothing more but a desktop computer and a new idea. But what happened was very different: 1% of the richest banks started building super computers, and they were perfected in the last few years. They positioned them close to marketplaces like the New York Stock Exchange and actually are aimed at driving prices up for those very people who the new regulations intended to involve.

How We Ate Your Lunch -  A Case Study

On July 15th 2009, Intel had reported high earnings the night before. Investors quickly realized that as a result of Intel’s success, Broadcom, a semiconductor company, would be in a strong position. Traders should be investing in Broadcom shares. But they faced a conundrum: buying large number of shares would alert the markets and drive up prices. So they divided their orders in smaller batches, ducking and diving as they went on their business. The markets opened, and shares started trading at $26.20.

The orders were being issued, in small and cloaked forms. This is where the high-frequency sellers kicked in: their computers saw a pattern. They managed to be faster than the traders by 30 milliseconds (0.03 seconds) and executed what is known as flash orders. Their computers started buying Broadcom shares and then resold them to the investors who placed the orders in the first place, but at the higher price of $26.39.

What happened was that the high-frequency computers saw the movement in the market, executed a sale faster than the traders could and then resold these trades to the traders at a higher value. This practice rakes in huge profits every day.

That world is changing

The former executive vice-President of the NASDAQ  is wary of the new system of ‘High-frequency trading’. He warns that it undermines the very purpose of a stock exchange: raising capital for growing companies. Instead, the large profits from High-frequency Trading have led big investment funds to focus on the shares of the biggest companies where shares are plentiful. As a result, the feeding capital for the future of the US, the new growth companies, is cut off. This means less growth, innovation and jobs.

Instead of speculating on future innovation, all money is currently caught up in a gaming system that simply analyzes movements in the shares of the largest companies, and artificially creates value and trade without these companies actually offering equal amounts of innovation or improvement of service to justify this.

The former executive Vice-President of the NASDAQ is warning that this practice is undermining the US economy and undercuts the integrity of the markets.

“You want to encourage innovation, and you want to reward companies that have invested in technology and ideas that make the markets more efficient,” said Andrew M. Brooks, head of United States equity trading at T. Rowe Price, a mutual fund and investment company that often competes with and uses high-frequency techniques. “But we’re moving toward a two-tiered marketplace of the high-frequency arbitrage guys, and everyone else. People want to know they have a legitimate shot at getting a fair deal. Otherwise, the markets lose their integrity.”

Fighting back: Dark Pools

There already has been a growing backlash. Companies now increasingly trade in ‘dark pools’, away from the stock markets and their High-frequency Trading Algorithms. Dark pools are currently estimated at 30% of the trading market.

A dark pool is essence a place where a willing buyer meets a willing seller in a closed room, and makes a non-public deal. This means that the actors are making decisions disconnected from the rest of the market, and that makes it very hard to understand the value of the buy. In effect, without assistance of the market, the buyer and seller are throwing their hat at the actual value of the trade and it could be argued that trades in the dark pools throws this part of the market back hundreds of years and all the instabilities that come with that. Fair enough, the buyer and seller are highly educated, but they are NOT an island.

The greatest innovation in financial systems was to have all actors participating in the price formation process, a price that reflects the business is being done and ensures transparency by treating all buyers and sellers equally. Today however, markets are less transparent than 40 years ago.

This has consequences. The more trade happens in dark pools (and it is constantly increasing) the more it undermines the system of pricing in general, because your prices don't reflect a sufficiently large enough part of the market. Many people are making their decisions on current prices; if these prices don’t actually reflect actual value as well as they can, this can lead to unexpected upheavals, booms and busts and volatile markets.

Can you spare some change?

All of this is now part of a larger debate on how to regulate markets: how do we reduce banks appetite in trading and increase its appetite for traditional banking: taking in deposits, giving out loans, raising share finance for growing businesses. But with such big profits available in what is essentially casino trading and the current system in America of legally buying the votes of Senators through lobbyists, it is unclear this growing profitable instability in the markets can be tamed before a new crash occurs.
And all the while, it is draining liquidity from the markets just as the economy needs it most, lavishly pouring money on the established businesses while cutting off funding from growing industries, and in effect, the future of America.