THE US Securities and Exchange Commission (SEC) is proposing a rule for all companies to file their annual and quarterly reports in eXtensible Business Reporting Language (XBRL) on a staged basis over the next three years.
Under the proposal, the 500 largest companies — including foreign firms that use US GAAP — will begin filing XBRL attachments with their annual reports early next year. They will be followed a year later by approximately 1,700 additional companies in 2010, while all remaining smaller companies and foreign issuers that use IFRS will submit XBRL in 2011.
In their first year of filing, companies will be required to file annual and quarterly financial statements tagged in XBRL along with footnotes and schedules that are “block tagged,” which means that each note has a single tag. However, in their second year companies will be required to tag their footnotes and schedules in detail, which will sharply increase the burden on companies.
In apparent recognition of the challenges involved, companies will be given a 30-day grace period when they prepare their first XBRL filing and when they file their first deep-tagged footnotes and schedules. The XBRL filings will be additional to companies’ traditional SEC filings and will not carry the same liability as traditional filings.
“Incremental” is the keyword for public companies
Ironically, for a technology touted for its efficiency, preparing XBRL financial statements will mostly be a labor-intensive, manual process for companies. This is because most firms will be tagging their financials and footnotes at the end of their normal financial reporting process, either by doing so internally using one of several software tools or by outsourcing the tagging to a financial printer.
The SEC pegs the average cost to prepare an initial XBRL filing with block-tagged notes and schedules at about $30,000, which includes software and, yes, labor. However, that cost drops by about 85% for subsequent filings as companies can reuse the templates they create for their first filings.
The cost and burden of preparing deep-tagged footnotes and schedules in the second year appears to be unknown. I think it could easily be double or triple that of tagging the face of financial statements.
Under the proposed rule, companies will also be required to post their XBRL information on their websites, but there does not appear to be any requirement for that information to be provided in a human-readable format. If companies are only required to post raw XBRL files on their sites, then this requirement should have no cost implications for companies.
Importantly, it is expected that enterprise resource planning software vendors will incorporate XBRL tools into their products now that it is clear the SEC and other regulators worldwide will require the technology. Integration with internal company systems will eliminate the incremental labor burden and actually result in cost savings, faster reporting and more accurate information for companies.
Will investors care?
Although almost everyone gushes about XBRL being a revolutionary technology for investors, I’m not so sure that the SEC’s proposed requirements go far enough to make this true for US companies.
This is simply because the information that will be available in XBRL is not timely. The fact is that the single most important disclosures are companies’ earnings news releases, which typically are published a week or more before companies file their quarterly or annual reports with the SEC.
Since there is no requirement for earnings releases to be tagged in XBRL, and since the SEC is unlikely to require simultaneous publication of releases and 10-Qs and 10-Ks anytime soon, investors mostly won’t be getting the most widely followed company disclosures in XBRL, at least not from US companies.
The biggest beneficiaries of XBRL are likely to be corporate data vendors like Thomson Reuters, Standard & Poor’s, Morningstar and EDGAR Online. These firms currently employ people to reformat data from SEC filings for delivery to investors. With XBRL, issuers will be doing that work for them.
Of course, the US XBRL files may be useful for screening companies and performing comparisons, but only if historical data is also provided, which does not appear to be the case at this time.
The bottom line then is that XBRL data for US companies will be available over the next three years. It will be a labor-intensive chore for companies to produce that data, to the point where it may prompt a good proportion of them to seek better ways to prepare XBRL rather than bolting it on at the end by “sending it to the printer.” And while it’s not clear how much investors will benefit from the data that is provided, there’s enough there to prompt innovation by vendors who serve investors. In that sense, the rule proposal seems well balanced and is an important step in the right direction.
For more information on the SEC’s proposal, see:
The Division Statement before the Commission Open Meeting by James Lopez
The SEC press release on the proposal
The webcast replay of the meeting
XBRL US Preparers Guide (PDF 1.6MB, 129 pages) What you need to know to prepare XBRL financials.
CFOs Lack Experience with Upcoming Standards — WebCPA, NY
It’s official: XBRL is mandatory — IR Magazine, UK (Ed – Ignore the headline.)
[...] Jones at the IR Web Report generally agrees with this appraisal and attempts to quantify the cost: The cost and burden of [...]
[...] SEC posts proposed XBRL ruleTop tech blog exposes scam at IPO firmNIRI’s earnings guidance survey makes no senseHistory made in JapanIROs more short-term focused than analysts — UpdatedIR Magazine Award winners earn superior returnsWeb 2.0: Think it throughXBRL in plain EnglishXBRL group assesses SEC rule impactSEC proposes rule to make XBRL mandatory [...]
[...] SEC proposes rule to make XBRL mandatory (May 15, 2008) [...]