LATENCY SPECIAL REPORT 2014: Lasers or lawmakers?

30 July 2014 | Richard Irving


The next big thing in ultra-low latency is lasers. But will lawmakers stop trading firms from using them to play the financial markets?



What is more likely to halt the esoteric market for ultra-low-latency connectivity in its tracks: an explosive new book by award-winning author Michael Lewis, who claims that a secretive band of trading firms are using the very latest in superfast optical technology to rig the $22 trillion US stock market, or a clever new system that delivers data at speeds so close to the speed of light that it may no longer be worthwhile trying to compete in the so-called “race to zero”?

Regulators have been circling around trading firms who use ultra-low-latency networks and specialist “black box” computer programs to play the stock markets ever since the “Flash Crash” of 2010, when the benchmark Dow Jones stock market index fell 1,000 points in under 30 seconds and the share price of a number of major S&P 500 companies fell to just one cent per share, while others increased in value to over £100,000.

The secretive band of firms, known as high-frequency traders (HFTs), use highly complex computer algorithms to take advantage of tiny aberrations in stock and bond prices across the plethora of electronic stock exchanges that have sprung up in the US in recent years. Tens of thousands of buy and sell orders can pulse through the markets quicker than the blink of an eye and some exchanges estimate that HFT strategies account for anything up to 50% of total trading volumes on any given day in the financial markets.

But Michael Lewis – who gave the world the phrase “Masters of the Universe” when describing high-roller traders at Salomon Brothers in his 1980’s book Liar’s Poker – alleges that the HFT community is rigging the market and cheating ordinary investors out of hundreds of millions of dollars every day.

In his new book, Flash Boys: Cracking the Money Code, the author claims that some firms are using ultra-fast networks to flush out bona fide orders on one exchange and then using what is in effect insider information to speed to another exchange and trade in those markets before the original order goes through.

You might think that there are just two types of order: “buy” and “sell”. But Lewis has identified no fewer than 27, many of which rely on superfast connectivity, including a particularly pernicious order that is designed to “disappear” the minute an investor tries to act on it.

In the book, Lewis lifts the lid on how Spread Networks was able to build and market its now-infamous $300 million ultra-fast link between Chicago and New York, as well as the incredible measures some HFT firms will go to in order to hide their identities at the data centres of the exchanges where they are based.

The debate heats up
Lewis’s allegations have hit a raw nerve: since the book’s publication, both the US Justice Department and the FBI have confirmed that they are investigating the industry for possible insider trading allegations, while securities regulators in Europe have introduced rapid-fire rules aimed at curbing some of the more maverick practices.

Meanwhile Virtu Financial, one of the world’s biggest HFT firms, subsequently shelved an initial public offering which would have valued the company at $3 billion, amid reports that details of its regulatory filing – in which it confirmed that it had made money on every day but one over the course of the last five years – had prompted the New York Attorney General, Eric Schneiderman, to write to the company seeking further information.

Schneiderman has subpoenaed six other HFT firms for information as part of a war on what he has called “insider trading 2.0”. Lewis’s take on the use of ultra-fast networks has many vocal detractors on Wall Street. Critics argue that HFT has made markets cheaper and easier to trade and they applaud the network providers for “democratising” the financial markets. However, regulators, including the highly influential Securities and Exchanges Commission (SEC), are starting to seek curbs on HFT such as a mooted “anti-disruptive” rule, which would kick in when HFT activity breaches certain parameters.

It is not the first time that policymakers have suggested putting the brakes on ultra-low-latency connectivity between various markets and with each new proposal comes a slightly better-honed argument in its support.

Ultra-fast network providers are reluctant to talk on the record, given the uncertain outlook for the low-latency market itself and the intense regulatory scrutiny that many of their customers currently find themselves under. The head of low-latency services at one leading network provider in Europe, speaking only condition of anonymity, is more bearish.

“The low-latency market has always been small – and limited in terms of growth potential,” he says. “I think you would currently have to be very cautious indeed in terms of the investments in new infrastructure that you would be willing to make, versus the addressable market going forward. This was pretty much an ex-growth market for us before Flash Boys came out.”

Fast money
But some smaller niche providers are still cautiously optimistic.

“I wonder to what extent regulators will ever be able to put the genie back into the bottle,” says a senior source at one US network operator. “The evolution of ultra-low-latency networks has created this psyche in the minds of traders that speed conquers all. That will be very hard to change, no matter what restrictions you end up putting in place.”

Moreover, while regulators might be able to foist speed limits around America’s eleven regulated exchanges, there are more than 50 so-called “dark pools”, or unofficial markets, run by various investment banks and brokers where such limits would be hard, if not impossible to enforce.

“I think it will be very interesting where the market for low-latency markets goes from here,” says the US network provider. “At the end of the day, lower latency makes for a better network. But if limits restrict providers from competing purely in terms of speed, then perhaps resilience and redundancy might once again come to the fore.”

On the last page of Flash Boys, Lewis introduces a cliffhanger. While cycling in the Appalachians of Pennsylvania he discovers a microwave transmission tower with a Federal Communications Commission licence number: 1215095. As Capacity reported last year, a number of network providers have established microwave links between New York and Chicago, cutting the latency on the route from 13 milliseconds to around 7 milliseconds – very close to the time it would take for light to travel between the two cities (around 6 milliseconds).

“It isn’t possible to keep any of this secret anymore,” Lewis writes. “A day’s journey in cyberspace would lead anyone who wished to know it into another incredible but true Wall Street story of hypocrisy and secrecy and the endless quest by human beings to gain a certain edge in an uncertain world. All that one needed to discover the truth about the tower was the desire to know it.”

Lewis does not go on to reveal the identity of the owners of FCC licence 1215095, but Capacity can confirm that it belongs to Epsilon Networks, a joint venture between BGC Partners, a securities broker and Thesys Technologies, a firm which specialises in developing networks for HFT firms. Thesys is owned by Tradeworx, the company that provided US regulators with a specialist high-speed analytical tool to help them determine whether HFT is good or bad for stock markets. Arguably, the point is not whether or not Tradeworx is conflicted in serving both regulators and the HFT community, but rather, whether the inexorable search for ever-faster connectivity – the so-called “race to zero” – can run for much longer.



As technology edges closer to the limits determined by the physical speed of light, it is perhaps the law of diminishing returns, rather than the threatened actions of regulators, that is likely to determine whether network providers continue to compete in the ultra-low-latency market.

That dynamic was thrown into sharp relief earlier this year, when Anova Technologies unveiled a new 35-mile laser link between the data centres of the New York Stock Exchange, in Mahwah, New Jersey and the Nasdaq stock market in Carteret, New Jersey, using technology first developed for use by jet fighters. The laser beam – which is no wider than a human hair – can transport data around 100,000 miles a second faster than fibre and, unlike fibre, can travel in a straight line, even accounting for the curvature of the earth.

The new route is less vulnerable to wind and rain than microwaves and it can carry more payload than millimetre waves. A flexible mirror system weeds out weather-related atmospheric distortions and a stabilising system helps to maintain contact between each relay point on towers that can twist and sway by up to 3 degrees.

The race ahead
In the race to zero, Michael Persico, founder and CEO of Anova Technologies, is undoubtedly sprinting down the home straight: an ultra-fast route that comes to within a whisker of the limits of physics and delivers “five-nines” reliability is going to be hard to beat. The question is whether, regulatory fears pending, the race is worth the prize. Persico is ambivalent.

“The drive for greater speeds is all part of the evolutionary process,” he says. “You don’t ask people to make cell phones slower and you don’t ask people to make computers slower. The whole world is accelerating. The point, surely, is that you have to accelerate the development of the ecosystem that surrounds this technology.” In other words, he says, it is the responsibility of the rest of the market to catch up. Moreover, Persico does not see his network as a pure play for the HFT market. “We have enough capacity to provide high-speed access for everyone. This is about democratising the process – we want to level the playing field for everyone,” he says.

To make this dream a reality, Persico – a serial financial technology entrepreneur with two successful deals already behind him – has snapped up more than 5,000 individual FCC spectrum registrations granting him the rights to roll out a laser network linking America’s entire US financial infrastructure.

“Pushing the boundaries of physics is in our DNA,” he explains. “We may look like a carrier, but we feel more like an engineering firm. That makes us appreciate our jobs a little more and ultimately gives our customers a better set of products.”