SEC buys into XBRL

Add the Securities and Exchange Commission to the growing list of financial regulatory agencies getting ready to adopt Extensible Business Reporting Language (XBRL).

Under a stock exchange pilot program likely to debut next March, company officials submitting annual reports to the regulatory agency will be able to voluntarily attach an unofficial, supplementary XBRL financial exhibit to their filings.

A derivative of Extensible Markup Language, XBRL is an open metadata standard for tagging business and financial data. It simplifies the analysis of financial data.

"It's actually not only a dictionary, but it takes your accounting ledger and

provides not only the link bases of where you got the data from, but how you derived that data," said Richard Heroux, program manager of the SEC's Electronic Data Gathering, Analysis and Retrieval (EDGAR) system.

"From a standards perspective on financials, it's the most advanced standard right now," Heroux said. "We're going into this pilot [test] with eyes wide open to see what comes out."

Agencies regulating the banking industry already are in the process of releasing a shared data repository of quarterly call reports submitted in XBRL. Bank officials will begin complying with the mandate starting next year, according to Federal Deposit Insurance Corp. officials.

"Change is really coming," said Liv Watson, an EDGAR Online Inc. senior vice president and one of the specification's original designers. "It's almost like having e-mail in the early days. You never got an e-mail because nobody had it, but could you live without it today?"

The SEC program will test the agency's ability to parse and render XBRL-tagged data, and it will provide investors and others the chance to do the same, Heroux added.

Officials at some companies, including EDGAR Online, Microsoft Corp. and Morgan Stanley, have indicated an interest in joining the pilot test, said Walter Hamscher, a member of the XBRL International Steering Committee and an XBRL designer.

XBRL eliminates the need to rekey data, said a Morgan Stanley analyst. "Right now, analysts get financial data, and they type it into a spreadsheet," the analyst said. "XBRL is going to get data directly and incorporate it directly into the data spreadsheets, which means few mistakes and less typing and more time analyzing."

Studies show that analysts can spend up to 80 percent of their time preparing data, Watson said. "How about if we switch those numbers and spend 20 percent on the mechanics of the data and then make more analysis or do something else?" she asked.

In addition, widespread adoption of the specification could foster greater understanding of company financial data. "I can look for trends over a longer period of time" once financial information is in a searchable XBRL database, Watson said.

"All of the facts in a financial report either refer to a period of time or else an incident in time," Hamscher said. XBRL "essentially gives you the rules of the language, how you say one or the other."

Why use XBRL?

Tagging financial data with Extensible Business Reporting Language (XBRL) offers several advantages over Extensible Markup Language metadata, said Richard Heroux, program manager of the Securities and Exchange Commission's Electronic Data Gathering, Analysis and Retrieval system.

XBRL is more than just a dictionary of data elements and tags. "With XBRL, you're talking about a set standard where you can actually dig down deeper into the tagging structure and see where the data was derived from," he said.

An XBRL taxonomy "adds additional language that says things that XML schema can't," said Walter Hamscher, a member of the XBRL International Steering Committee. Taxonomies define data elements such as the ways company officials report time periods or how they refer to business entities.

Using a language analogy, Hamscher compared the difference between taxonomies and schemas to "the difference between the rules of English grammar and the rules of iambic pentameter poetry."

The idea is that financial data comparison becomes automatic, he said. Individual companies can publish new taxonomies because the standard is open but in a manner that allows XBRL software encountering a new element to incorporate it by using embedded descriptions of the element. "The idea is to accelerate a set of terms that people can agree on," Hamscher said.

Making taxonomies easily extensible also ensures their long-term relevance, said Liv Watson, one of the specification's original designers. XBRL International, the nonprofit consortium fostering the language's development and adoption, is not interested in saying "here are the taxonomies, and that's the way it's going to be for the rest of our lives," she said. "Those things are living organs."

— David Perera

Reader comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above