Data standards are beginning to emerge for financial information, including news, market data and research. The new formats make use of Extensible Markup Language (XML), a general data standard developed for the internet that is now being adopted by many industry sectors. The new financial versions of XML promise to make it easier for firms to distribute, receive and collate information such as business news stories, price data and equities research. Some claim it will help improve investment decisions.
Although information suppliers have tended to take the initiative and draw up draft specifications, they are putting these into the public domain, either through industry associations or newly formed consortia. These groups are coordinating the further development of the standards, managing industry consultation and the formal ratification of specifications.
In the area of business news, Reuters initially created the News Markup Language (NewsML) as a means of combining elements of stories in different media, such as text, sound, graphics and video, into a single standardised package that could then be delivered via a range of different channels, including conventional news feeds, the web and mobile devices.
“Most trading starts with a news story of some kind,” says Mark Hunt, Reuters’ director of XML strategy. Interest rate announcements, company quarterly figures and other news items set off a train of events that lead to investment decisions and trades. In 1999, Reuters put forward a draft specification for multimedia news which it donated to the International Press Telecommunications Council. Last year, the IPTC ratified NewsML as a standard and a number of news providers, including Reuters, BusinessWire, Press Association and Dow Jones’, have pledged to adopt it.
NewsML has many advantages, not only for the information providers but also for investment firms and other financial institutions. First, it can lower costs because firms no longer need to cater for the multiple formats that providers use. It will also mean more flexibility because they can receive the information on a variety of platforms, including mobile devices. Firms will also find it easier to extract the information they need from stories, for example financial results from a company quarterly statement. NewsML can handle ‘nested’ standards, for instance the specific XML-based format that has been developed for business reports, called Extensible Business Reporting Language (XBRL). And because the information will be in formalised elements within standardised packages, firms will be able to automate the extraction and collation of data.
Many of these advantages apply to the other new standards that are emerging. Market Data Markup Language (MDML) enables firms to receive price and other market data from any vendor that conforms to the standard via a single pipe. Bridge Information Systems wrote the draft for MDML and has passed it over to the financial information services division of the Software and Information Industry Association (FISD), which has set up a working group to take the standard initiative forward.
In the area of research there are two new formats – Research Information Exchange Markup Language (RIXML) and Investment Research Markup Language (IRML). RIXML is the product of a consortium that includes 11 asset managers, including American Century Investments, Fidelity Investments and Putnam Investments, and 10 broker-dealers, including Deutsche Bank, JP Morgan, Lehman Brothers and Merrill Lynch. The consultancy arm of KPMG has been working with the consortium to produce the business case and formal specifications and Version 1.0 of RIXML is due for release in January.
IRML has the support of 30 firms, including Barings Asset Management, Charles Schwab, Deutsche Bank, JP Morgan Asset Management, Lehman Brothers and Newton Investment Management. The standard is at a similar stage to RIXML, with its draft specification due out this quarter. This will cover equities and company-related data, although the intention is to extend across interest rates, foreign exchange and so on in the future. The consortium has produced a prototype aimed at the buy side and wants to create a buy side panel to provide feedback, says Warren Sample, director of special projects at Multex, the New York-based financial e-marketplace specialist that initiated the IRML project.
There is some confusion surrounding the fact that there are two specifications purporting to cover the same area. It is not uncommon in new areas of technology for competing proposals to emerge simultaneously and then fight it out in the marketplace. However, the consortia claim that they are looking at different parts of the problem, with RIXML focused on classification and other high-level issues, also known as ‘metadata’, with IRML more focused on the components of research.
“RIXML.org_is complementary to what we’re doing and we want to work synergistically with them,” says Sample. The fact that a number of institutions are members of both consortia should mean that they make an effort not to duplicate each other’s work. There should also not be major issues of compatibility because both initiatives are based on XML.
Like NewsML and MDML, the new standards for research promise lower costs through a simplification of technology requirements to handle information from multiple sources, as well as greater flexibility in collating and analysing information and new possibilities for automation of research functions.
Furthermore, according to Joseph Sommer, director, US electronic trading and connectivity services at Credit Suisse First Boston and sell-side co-chair of the steering committee, “RIXML ultimately could yield better investment decisions as portfolio managers can focus on the content that is most relevant to their areas of interest.”
However, it will be some time before the standards really start to take effect. Dushyant Shahrawat, an analyst at market research organisation TowerGroup, points out that the large sell side firms are generally in good technological shape to adopt the new standards, “but it will be more of an issue for mid-tier broker-dealers because their front-end systems don’t tend to speak XML”. And the buy side is in an even worse position. “They’re behind the ball in terms of having the ability in their systems to receive and exchange XML-based information,” he says.
It may take one to two years before the industry can fully implement the standards, suggests Shahrawat, but when it does the payback could be significant.