Taming the Market Data Beast
Market Data Migraines
So have IT vendors been able to cure market data management headaches for the buy side? No, says Richard Chmiel, VP of global sales and marketing for OneMarketData, a tick data solutions provider, with a laugh. "It is the single most important thing that firms bring in: data," he insists. "They bring in news, data, reference data from different sources, and the most important is market data -- even when the volumes keep going up and down."
The market data sector, however, still is plagued by multiple data sources and a lack of standards, Chmiel says. "There are inconsistencies across exchanges, across data providers -- that's what consolidators do," he says. According to Chmiel, the only way for a golden ticket -- a perfectly clean piece of aggregated market data -- to emerge would be for all of the exchanges to come together and create a common data output. "But why would they do that?" he wonders.
Instead, hedge fund managers have to scrub the data. "This is very time-consuming and very ugly work," admits Arun Kaul, Levas's partner at Olympian Capital Management. "It's difficult to scale but very necessary to ensure the integrity of the data. There is nothing worse than analyzing bad data or having missing fields."
There is a challenge in trying to integrate all of the information, says Kevin Samborn, a consultant for Sapient Global Markets. "Let's say you're a firm that trades swaps -- if you get a quote on the five-year swap directly from the broker, they send you some quotes on the yield curve. And when you close your books, if you're really small, you'll probably use the same quotes," he relates.
But, "If the swap is of any given size, you really should be downloading the yield curves from an objective, neutral source -- not your broker," Samborn continues. "Then they'll try to use Bloomberg or Thomson-Reuters [to get the data]. People are inclined to download data off a terminal and put the data in a spreadsheet, but you're technically not supposed to do that."
For many hedge funds, especially smaller shops, this is where third-party solution providers come into play. "We perform the scrubbing and cleaning and mobilizing that our end users don't have to do," says Darren Tedesco, managing principal, innovation and strategy, at Commonwealth.com, which provides market data services to wealth managers and financial advisers. "We centralize that in our back office."
According to Joel York, chief marketing officer of market data service provider Xignite, the challenges of buy-side firms when dealing with market data are threefold. "First, the cost and complexity of managing increasing amounts of market data as the amount of data has grown exponentially due to electronic trading, venue fragmentation and instrument complexity; second, getting that data into apps and algorithms instead of terminals; and third, adhering to and reporting on complex and varied exchange licensing requirements," he says.
Most buy-side firms face a Catch-22 when it comes to managing multiple data sources, according to York. "On the one hand they want to reduce costs through consolidation; but on the other hand, they are constantly trying to develop competitive advantage through better market access and new trading strategies, both of which imply the need for new and unique data sources," he says.
The key is to know the difference between commodity market data and unique market data. "Commodity data can be consolidated and even pushed to the cloud to lower costs and save precious IT cycles, whereas unique, proprietary data must be constantly sourced and protected," York says.
According to Mark Pesonen, head of the enterprise products and solutions group for Bloomberg, one of the leading market data providers (Bloomberg is the Coca-Cola of market data to Thomson-Reuter's Pepsi), every Wall Street firm deals with market data differently. "There are a lot of different factors that determine a firm's strategy for managing data, such as size, geography, asset types, trading strategies, latency requirements, security policies, etc.," he says. "Some firms install massive data management platforms at their sites so they can manage the data themselves, while others take advantage of managed services. As the volume and complexity of data increases, however, we are seeing many firms making the shift toward managed services, rather than dealing with the deluge of data themselves."
The Market Data Race
Meanwhile, even as capital markets IT enters an age of open source code and free operating systems such as Linux, dealing with market data remains a high-ticket item, and tighter profit margins inside investment firms have had an impact on the amount of money invested in market data management. "Everybody wants to keep investing in market data infrastructure, but banks are challenged by tight budgets," says OneMarketData's Chmiel. "The need and the desire to update is there, but the money isn't always there."
Chmiel adds that the quant funds that make up a large portion of the trading volume seem to be more dedicated to data management because they are making more money and thus have more to spend. "I put the long-only guys between the banks and the high-frequency guys," he says of the industry investment in data management. "They're doing better, but they're not knocking the cover off the ball."
When it comes to addressing market data, according to Asset Control's Lynch, the buy side's biggest challenge is cultural. The provision of financial data typically has been seen as a backseat function and as IT's problem, he explains. But, Lynch insists, "Whatever business you're in today, including capital markets, data and information is the foundation. It is not separate from the business." Phil Albinus is the former editor-in-chief of Advanced Trading. He has nearly two decades of journalism experience and has been covering financial technology and regulation for nine years. Before joining Advanced Trading, he served as editor of Waters, a monthly trade journal ... View Full Bio
Tweet This