New data format may uncover company misdeeds
WASHINGTON (Reuters) - Professors have long been friends to securities regulators, uncovering suspicious trading and accounting patterns that lead to multimillion-dollar settlements.
These researchers are about to get a new tool in the form of XBRL-tagged data, which will provide a treasure trove of easily accessible information to comb through, and could potentially lead to more enforcement actions.
The U.S. Securities and Exchange Commission last week proposed a timetable requiring about 500 of the largest public companies to begin filing their financial data in XBRL, or extensible business reporting language, in early 2009. Smaller companies would get an extra year or two to get ready.
XBRL tags are like bar codes attached to each piece of financial data and eventually text. The data can be downloaded into spreadsheets or accessed with an online viewer that allows users to easily find and compare key company data.
Corey Booth, chief information officer for the SEC, said analysis of companies' XBRL data could uncover some suspicious activities.
"It stands to reason that when you have that much more data available, it will produce more insight and some very interesting findings pertinent to us," Booth said.
The SEC's recent high-profile cases in stock option backdating were spawned by academic researchers examining corporate data.
Professors at the University of Iowa and Indiana University spent months poring over data from SEC documents and found many options grants were timed to exploit marketwide price depressions that not even insiders could predict, meaning that at least some of the grants must have been backdated.
The result: Hundreds of companies were investigated by the SEC or conducted their own internal inquiries.
Marvell Technology Group Ltd (MRVL.O) and Broadcom Corp (BRCM.O) agreed earlier this year to pay $10 million and $12 million, respectively, to settle SEC backdating cases.
Randy Heron, a finance professor at Indiana University who co-authored the backdating research, said XBRL will help spot irregularities in companies' financial data.
"Anything that will allow an electronic compilation of data is going to help in situations like stock option backdating," Heron said. "We really needed large volumes of data to get that together."
XBRL-tagged data will help investigators uncover illegal activity, such as insider trading and manipulation of financial statements to massage earnings, Heron said.
BIGGER AND BETTER
Currently, academic researchers use services like Standard & Poor's Compustat, which can compile data derived from companies' SEC filings. But subscriptions are expensive and carry the risk that the data was manually entered incorrectly. Continued...
