CFOs may not be able to avoid the topic of XBRL for much longer. A new
and more comprehensive version of the data tags that turn financial
statements into “interactive
data” has been completed and goes into market-testing this month.
XBRL-US, the
nonprofit organization hired by the Securities and Exchange Commission
to develop the
tags (small pieces of computer code), will work with at least 13 of the
48 companies currently
filing SEC documents
in XBRL to see how well the
new version works, according
to CEO Mark Bolgiano. Since
taking over the project last
year from the Financial
Accounting Standards Board
(which inherited it from the
American Institute of Certified
Public Accountants), the
group has been standardizing
tags for primary financial
statements and footnotes,
expanding the number from
less than 2,500 to 15,000.
That sixfold increase
means companies will have to do less customization, which
should make it easier to convert the typical items found on
financial statements into XBRL format, according to Bolgiano.
Critics, however, are concerned that it will make XBRL more
costly. “They’ve built a Taj Mahal when what we needed was a
nice four-bedroom house,” says one source close to the matter.
He predicts that conversion will require substantial — and
expensive — consulting help similar to what was required for Sarbanes-
Oxley compliance. The new language as currently drafted may
not satisfy one major constituency — Wall Street analysts —
since its structure is too complex, says Eric Linder, CEO of Savanet,
which makes a software tool that analysts would use to read
XBRL statements.
While the SEC has not yet mandated the use of XBRL, it
has invested $54 million to create the technology necessary to, as
it stated last year, “pave the way toward universal XBRL.” Chairman
Christopher Cox’s public enthusiasm for the project, as evidenced
in speeches, multiple roundtables, and congressional testimony,
has only heightened the expectation that XBRL will
replace the current HTML-based system. There’s no official
word on when or whether that might happen, but the eventual
public release of the final XBRL code (expected in 2008) will be a
major milestone, says Jeff Naumann in the SEC’s Office of the
Chief Accountant, since the current EDGAR system can already
handle XBRL documents.