Opening Cross: FIX That ITCH, But Don’t Fix the Latency Race

max-bowie

After discussing the need for a common data standard in this column last week—specifically how Nasdaq OMX’s ITCH protocol has become a de facto standard for data dissemination—I got an interesting response from the folks at FIX Trading Community, which is incorporating ITCH semantics into the FIX Protocol.

FIX recognizes that (a) ITCH has features that make it more suitable for high-speed, high-volume marketplaces than heavier FIX messages—for example, why represent numbers in text, as FIX does, when you can save bandwidth and latency by representing them in binary values?—and (b) ITCH has achieved widespread adoption in equities markets, and is now pushing into others, and (c) messaging protocols are not an area where market participants can gain competitive advantage, and should be standardized.

“There’s no competitive advantage in how bits and bytes are sent around the world, so we can all work together to make that more efficient,” says Sassan Danesh, partner at Etrading Software, which is conducting the gap analysis between ITCH and FIX. “ITCH is pretty well established, especially in the equities and listed world—and that’s beginning to merge with the over-the-counter worlds as regulators drive more OTC instruments onto electronic platforms, while other cash markets such as fixed income are now becoming high-frequency enough to demand low-latency data.”

Indeed, despite FIX’s efforts, ECNs preferred ITCH because FIX’s market data messages were too heavy, which ultimately led to the bandwidth-reducing FAST Protocol, says Jim Northey, co-chair of FIX’s Global Technical Committee and partner at LaSalle Technology Group.

However, FAST adoption was disrupted by a patent infringement lawsuit that led exchanges to implement alternative protocols. So, “there is still room for a compact binary format… so we have asked FIX’s Market Data Working Group to close off the FAST spec and look at a compact binary coding that would be easy for the industry to implement,” Northey says.

While the patent lawsuit stifled the innovation of FAST, there’s another factor that could also potentially stifle innovation in our industry: regulation. New York Attorney General Eric Schneiderman is continuing his crusade against unfair paid-for advantages being gained by high-frequency traders at the expense of ordinary investors. To date, Schneiderman has ended Thomson Reuters’ early release of University of Michigan consumer sentiment data and a BlackRock analyst survey program ostensibly to “front-run” analyst revisions, and has secured agreements from company press release providers Business Wire and Marketwired to stop selling direct feeds to HFTs. It’s unclear from the AG’s statements whether these services were offered illicitly for a premium only to HFTs, or more widely-broadcast feeds, but if the latter, outlawing more efficient distribution methods seems like an attempt to hold back the tides of technology advancement. And where does it stop: outlawing microwave data over fiber, such as Equinix expanding its microwave data distribution, or perhaps services like Estimize that pre-empt analyst estimates by crowd-sourcing data? Innovation always comes with a price tag, and trying to level the playing field by finding the lowest common denominator seems to me like the Olympics telling sprinters to run only as fast as the slowest competitor.

In addition, a statement from the AG last week appears to mis-state the purpose of co-location, suggesting it allows only high-frequency traders to see incoming orders before they hit the exchange order book, which of course is not the case: what co-location does allow HFTs and many others to do is to react more quickly to those orders after they hit the exchange.

While some are glad Schneiderman is probing trading venues that haven’t been subject to the same scrutiny as exchanges, others are concerned that any latency-reducing innovation could be censured by his well-meaning efforts, and that these moves may not protect investors, but create inequalities between institutional participants.

  • LinkedIn  
  • Save this article
  • Print this page  

You need to sign in to use this feature. If you don’t have a WatersTechnology account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: