Market Data Gone Wild
Have you ever seen those Girls Gone Wild videos? Well of course you haven’t, but I’m sure you’ve seen the commercials. I always think to myself, what do those poor parents of those girls think? “What happened to my sweet little daughter? How could this happen?” Well, when I look at my market data bill each month, I feel pretty much the same way about the market data system in the US markets. Really!
When I first started in this business, the gold standard electronic venue was the Island ECN. It was fast, cheap, reliable, and on the cutting edge. And best of all, Island’s quotes were free. It was a smart business model that incentivized you to take their data and, in turn, trade with them.
Market data these days is an entirely different animal. Every exchange now charges you handsomely to see their market data. So don’t take these feeds, you might say? Well, then you can’t compete. Every quality broker or platform provider must take all these feeds to provide their clients with the highest caliber market intelligence for execution. It you don’t take pay for the feeds, you won’t have any clients. So these feeds really aren’t elective.
Everyone knows the US Equity Markets’ trading volumes are pretty low these days, with volumes now tracking about two-thirds of their 2009 highs. But somehow from 2009 until now, my corporate market data bills have tripled. And the number of servers I have needed to process this data has almost doubled. I also now have to deal with regular exchange mandated market data audits. So why are the trading volumes and market data volumes seemingly moving in opposite directions?
Market data madness
The root of all this madness is a perfect storm of many variables. First, the for-profit exchanges bought up the cheap/free market data venues, such as Island. Free is hard to compete with when you are for-profit. Second, as we know, the trading volume reduced markedly in 2009 after the credit crisis/housing bubble/flash crash, so trading revenue everywhere was on the decline. Third, the rise in computerized trading, coupled with exchange/dark pool pricing models, created a perverse incentives for high frequency traders to flicker quotes that aren’t truly executable to “test” the market within a stock, thus driving up associated processing and bandwidth charges.
As a result of these factors the exchanges were forced to get defensive, and understandably so. They were getting run over by quote volume. New market data charges not only became a way to control their own rising market data costs, but also a way to backfill the lost trading revenue. These exchanges all now have multiple à la carte feeds with complicated billing structures, teams of market data reps to administer them, and market data auditors waiting in the wings in case you mess it up. The problem is all their clients are already captive. They have no choice but to pay the rising charges, lest they cannot compete.
Many of the counter arguments in support of the high frequency debate spurned on by Michael Lewis’s book point to lower trading costs and tighter spreads. I wholeheartedly agree that this is a positive benefit of the rise in computerized trading. It really is undeniable. But the question we must ask is at what cost? We cannot ignore the market data beast that has been created as a result of HFT, and the other factors I have mentioned. It truly is a hidden tax to all brokers, platform providers, and customers. The ultimate concern is these costs will get passed on to the customer, either directly or indirectly, if they haven’t already, in the form of new fees and rising commissions.
As the market structure debate continues, the changing dynamics of the US market data system needs to be recognized and analyzed with as much vigor as the trading side. Without that, we are only looking at half of the problem.
Scott Ignall is the Chief Information Officer of Professional Trading Solutions, an industry-leading solutions provider for active and professional traders as well as hedge funds and prop trading groups. Previously, he was co-founder and CTO of Lightspeed ... View Full Bio
Tweet This