Bob Schneider is a Partner in 
Wilson So is the Director of Hitachi Consulting Corporation XBRL Shows Promise for Managing Risk, but Hurdles Remain
Written by Matt Kelly Posted on March 17, 2009
Matt Kelly is editor-in-chief of Compliance Week, a magazine and online newsletter on corporate governance, risk, and compliance. Prior to his role at Compliance Week, Kelly was a reporter and contributor on corporate compliance and technology issues for magazines such as Time, Boston Business Journal, eWeek, and numerous other publications.
Now that the U.S. Securities and Exchange Commission has put a mandate in place for corporations to start filing financial statements in XBRL, Corporate America is dutifully churning away. We can expect the first instance documents to start appearing sometime in July, but the curiosity and drama surrounding XBRL and U.S. financial reporting has faded to a dull background noise.
Thankfully, other voices in the XBRL world are turning up their volume.
A new wave of XBRL enthusiasts has now begun a campaign to bring that same promise of XBRL — transparency and homogeneity — to business risks in the banking sector. Yes, many leaders of this movement are XBRL software vendors who have a vested interest in its success; and no, the path to achieve this new XBRL taxonomy isn’t yet entirely clear. But the idea has merit, especially with the banking system going down the tubes like it is these days.
Start with the IBM Data Governance Council. The council has begun holding seminars and webcasts on its vision of an XBRL risk taxonomy, which would identify risks in the banking sector the same way XBRL’s taxonomy for U.S. GAAP identifies data in financial reports. Such a system might, for example, allow a regulator to peer through a bad-debt line item and see the individual loans feeding it; that would take hours of diving through spreadsheets today.
Other parties throwing their support behind the XBRL risk project include the Enterprise Data Management Council and a variety of XBRL vendors.
Unto itself, the idea of applying XBRL to risk management makes plenty of sense: risk management is still much more art than science even at the best of companies, and the financial sector has proven to be a disaster at knowing what its risks are. Sometimes that has manifested as a company whose risks suddenly explode and wreck the balance sheet; just as insidious is the drag it places on truly healthy companies now too scared to dabble with markets they no longer trust. The end result is less liquidity, less clarity, and less prosperity. We are in dire need of a solution to that problem, and XBRL seems to be an attractive candidate for the job.
Still, this idea needs to travel a long road before it reaches the same conclusion as XBRL in financial reporting. In that instance, we had a reasonably well-understood realm of data (U.S. accounting terms) to match to an XBRL taxonomy. We had a single authoritative regulator (the SEC) that would lead the way for companies and investors to follow. We had clarity of goal and leadership.
That clarity is still severely lacking here. Financial institutions are overseen by many regulators — the Federal Reserve, the Federal Depositors Insurance Corp., state banking commissions, and even state insurance commissions — and Washington is poised to overhaul that whole system, dramatically, in the near future. And finding a taxonomy of risks in this realm strikes me as nearly impossible, considering how fast and furious new risks emerge in this sector.
How will this effort all end? I don’t know. But it’s a conversation worth listening to.
March 17th, 2009 at 9:32 am
This is an excellent summary. Just a few more thoughts for the table:
# Since risk is in the eye of the beholder while GAAP is in the eye of FASB, it will be interesting to compare how robust the first draft of a risk taxonomy is compared to the GAAP taxonomy. How many standard tags? How many extensions? What will be the process for determining whether a particular fact merits a standard tag?
# Is a crowdsourced development process possible in this space like it was in the GAAP space?
# The comment process for the GAAP taxonomy was fully transparent to anyone who wanted to watch it. Will the same be true for the risk taxonomy?
# How will this integrate with the work XBRL US has done to create taxonomies for RMBS, TARP, and particularly the mutual fund risk/return taxonomy? The credit rating agency taxonomy? Can a risk taxonomy be made to wrap-up into summary concepts that can be understood by people with undergraduate math skills, or will it take an advanced degree to understand even high-level concepts?
I don’t know the answers to these questions, but that last thought prompts another: is there a metric to measure complexity itself? Computers can analyze the complexity of language – according to Word, this post is written at a 9th grade level with a 59.8 reading ease score. Of course, that was before adding the past few sentences. Perhaps those are metrics that could help answer the question left by yesterday’s New York Times op-ed on credit ratings: what now? Perhaps prospectuses should be data tagged for the complexity of their prose disclosure!