The SEC is likely to propose a new rule next year that publicly traded companies file their financial reports using XBRL, the interactive data language SEC Chairman Christopher Cox has long hailed as a powerful tool for investors to study financial results.

Speaking at a Sept. 25 press conference to unveil a new “taxonomy” of XBRL terms used to classify financial data, Cox said he has asked SEC staff to “finalize a recommendation to the Commission that we could act on next spring, about how to deliver the maximum benefit of tagged data to investors, including converting all public company disclosure into interactive data format.”

Cox

When asked about a timeline, Cox said he expects the SEC to consider recommendations in proposed form in the spring of next year, “and if it follows the course of an ordinary rulemaking, we’d probably complete that in the fall of next year and would lay out a schedule for implementation.” That would also neatly coincide with Cox’s departure from the SEC, which presumably will happen when a new White House administration takes office in 2009.

Cox also said the new XBRL taxonomy should be ready for public review by Dec. 5 at the annual XBRL International Conference in Vancouver, British Columbia. Various auditing firms, accounting rulemakers, and technical experts are examining the taxonomy now.

Cangemi

Many in the business community have suspected that Cox might try to mandate XBRL. Michael Cangemi, CEO of Financial Executives International, says his group “is pleased that this stage of the process is complete,” but diplomatically adds: “We believe setting target dates for technology implementation is risky and should be very carefully studied, with the involvement of the major accounting software developers.”

In preparation for the upcoming public review, Michelle Savage, spokeswoman for the XBRL US consortium, admits “there has been some hesitation and a little bit of inertia,” by the business community. “With the new taxonomy, it will be drastically easier to implement,” she says.

Cox has been pushing XBRL as the next big thing in financial reporting since 2004. Essentially, XBRL “tags” individual pieces of financial data so they can be more easily presented in an online financial report. For example, Microsoft—a strong proponent of the language—would tag the number 51,122,000,000 as revenue, so people reading the report could quickly identify it as Microsoft’s 2007 sales.

With the proper taxonomy, XBRL enthusiasts say, the language could encompass all manner of financial data, from inventories to contingency accounts to payroll. And with the right software on users’ computers, XBRL could give rise to powerful new data analysis tools, allowing investors to pore over financial reports and quickly compare them with other companies.

That said, only a few dozen companies are participating in the SEC’s pilot program for filing statements in XBRL. By comparison, more than 10,000 public companies file quarterly, annual, and other statements at a rate of some 700,000 documents each year on the text-based EDGAR system that the SEC uses today.

Taxonomy 2.0

The old taxonomy for voluntary filers was the product of a volunteer effort led by XBRL US and had only about 2,500 tags. Earlier this year, however, the SEC stepped up its commitment with a $5.5 million contract to develop the new, more comprehensive system. A principal complaint about the old one was that its definitions were not diverse enough to match the wide range of financial data that companies might include in their reports; the new taxonomy—with 15,000 tags— is intended to solve that problem.

Microsoft, for example, found 600 items in its financial statements that had no identity in the old XBRL taxonomy, says Mark Bolgiano, president and CEO of XBRL US. “I don’t know if we’ll get that number down to zero, because they’re the only company in the world with Xbox sales in their revenue statements,” he quips. “But we’re taking a huge step forward in compatibility.”

Liberty Media has joined Microsoft among the companies incorporating XBRL sooner rather than later. (Filers submit their XBRL statements in addition to, rather than in place of, regular filings.) Mark Burton, Liberty Media’s vice president of financial reporting, said in an interview just before Cox’s Sept. 25 announcement: “I guess our expectation here is it’s going to be mandatory within a year or two, and rather than wait and have to compete with thousands of other companies, we’d rather get ahead of the curve.”

Burton said he’s doing most of the work himself. “I haven’t found that it’s that difficult conceptually,” he said. “I think it’s been relatively straightforward and easy to implement.”

Exelon Corp. became the first utility to join the SEC’s voluntary program in June. Sabrina Miller, a spokeswoman for the company, says Exelon “recognized that this most likely would be an industry standard, and we wanted to participate and be involved in the development of tags, tools, and standards.” Initial filings that initially took weeks to prepare are now done in a matter of days, she says.

Burton occasionally calls for tech support from Rivet Software, which in September 2006 won a $250,000 contract from the SEC to create a “reader” program that can present XBRL data. Rivet has worked with roughly half the companies filing voluntarily, says Mike Rohan, CEO of the company.

The XBRL reader, now posted on the SEC’s Web site, receives about 40,000 visits per week. The Commission published the source code to the reader in September; Jeffrey Naumann, an SEC technologist working on the XBRL program, says the goal is to get other software companies and enthusiasts to use the code as a foundation to create more XBRL tools.

The ‘Pain in the Butt’ Part

Despite apparent progress, achieving Cox’s vision of transforming the text-based EDGAR filing system with XBRL filings remains a challenge. The stubborn obstacle? According to Rohan, it’s the taxonomies.

Kugel

Current taxonomies remain limited to basic income statement, balance sheet, and cash flow and shareholder equity data, which professional analysts (and skilled amateurs) can lift from tables in text filings without too much effort. According to Robert Kugel, senior vice president at Ventana Research, professional investors would find it much more valuable to tag Management Discussion and Analysis, footnotes, and other more unstructured portions of SEC filings.

But those sections of company reports are “more of a pain in the butt,” Kugel says, “because, quite frankly, they don’t come in a predigested table like an income statement, where all you have to do is copy, paste, parse, and extract.”

Bolgiano says footnotes and disclosures are tagged in the latest XBRL US taxonomy under development.

Kugel also praises the CFA Institute, which is an industry voice for market analysts and financial experts, for its work in building XBRL’s reputation in stock analysts’ minds. “It's extremely important for analysts to be involved in the process, because they’re the ones who are going to be consuming this stuff,” he says.

Truly widespread adoption of XBRL will require mass conversion to the language, which Naumann says may require an SEC mandate. “You really need everyone to move at once to get critical mass,” he says. “No analyst firm is going to invest in a whole XBRL process to assess 10 percent of the companies they cover.”

Bolgiano says the SEC would open a public comment period once it is satisfied with the results of the technical review.

Ric Marshall, chief analyst at The Corporate Library, says the move from EDGAR to XBRL will be no less significant than the switch from paper reports to the EDGAR online database 14 years ago.

Even the incremental costs incurred by companies in the voluntary filing program will go away once XBRL is the norm, Marshall says, since XBRL is laden with automated features that should cut filing costs—and the number of hands touching financial data before the statements are filed.

“It's something that’s going to happen,” Marshall says. It’s just a question of when.”

Topics