A revolution for small investors

The next problem is that the current database providers have to adjust numbers to clean up the data and make it comparable. They are doing the right thing. But unless you study their adjustments closely, you might not be sure what you are getting.

Thanks to an XBRL dictionary turned over to the SEC this week by XBRL US, accounting concepts and the types of numbers linked to specific XBRL codes are tightly defined and uniform across companies.

Another shortcoming now is that it can take a day or two for the numbers behind a news release to hit today's fee-based databases. That's because numbers have to be adjusted and keyed in. With XBRL, shortly after someone pushes the button to send company data to the SEC, it will be available to everyone.

"For the first time, investors will be consuming the data as filed," says Mark Bolgiano, the chief of XBRL US.

Brave new world

If you consider the potent mix of full transparency, minimal costs and real-time access to data on the horizon, it's easy to imagine where this might go.

Envision a world where citizen analysts and bloggers -- the kind who now bring down media titans and politicians -- collaborate to easily create analytical tools that uncover accounting shenanigans that might otherwise go unnoticed for years. "I think it's revolutionary for individual investors," Willis says.

Though it might not prevent another Enron-style rise and fall built on funny numbers, Bolgiano says, "it makes it a heck of a lot harder to pull off. It shines a lot more light on the data."

(To get a rudimentary sense of how investors might exploit an XBRL database, check out the demos at Rivet Software Crossfire.)

The SEC also offers some basic XBRL readers. Unfortunately, you can't actually do much cross-company analysis because so few U.S. companies use XBRL. For a complete XBRL database stretching back years, you'll have to spend a few thousand dollars a year on Edgar Online's IMetrix.

Progress report

XBRL emerged in 1998 out of accountant Hoffman's frustration with how time-consuming it was to weave together numbers from different company reporting systems.

Even though XBRL has been around for years, so far in the U.S., companies only use it voluntarily. In fact, only 78 companies now report numbers using XBRL, including PepsiCo (PEP, news, msgs), General Electric (GE, news, msgs) and Microsoft (MSFT, news, msgs). (Microsoft is the publisher of MSN Money.)

Hoffman thinks all companies will have to use it by the end of 2009, beginning with 500 to 1,000 large companies in the fourth quarter of this year. Childs, of Edgar Online, and Willis don't expect full implementation until 2010.

Challenges remain

Even with XBRL fully in place, there will be challenges for investors who want to use it. Above all, you'll have to understand complex accounting concepts, says Eugene Imhoff, a corporate-finance professor at the University of Michigan's Stephen M. Ross School of Business.

• Video: Pushing accountability on Wall Street

Companies naturally resist being told how to report their numbers. They do this because they want to show they are different from all their competitors or because they want to spin numbers to put their best foot forward.

"Companies are still going to be pounding square pegs into round holes and coming up with things that are not what they appear to be," Imhoff says.

XBRL coding allows room for companies to create their own tags if they think they have a special situation. But too many of these may create puzzles for investors.

And the SEC isn't going to make companies perform audits to verify their tagging is accurate. This means mistakes are inevitable, says Lynn Turner, a former chief accountant at the SEC who now serves as a consultant to the forensic-accounting firm Kroll. So even though all the numbers in the SEC's XBRL database will be neatly tagged, "it is likely to be wormy," he says.

At the time of publication, Michael Brush did not own or control shares of any equities mentioned in this column.

< previous |  1 | 2 |