Thursday, December 1, 2011

Response to FEI's Open Letter to the SEC

By: Nathan Summers, Software Engineer
CompSci Resources, LLC

By now you've probably heard that the Committee on Corporate Reporting of Financial Executives International (FEI) has sent a high-profile letter to the SEC requesting that certain detailed tagging requirements be eliminated. It seems to me like the letter focuses on the pain of compliance as opposed to the benefits and increased transparency that can be gained through the use of eXtensible Business Reporting Language (XBRL). XBRL is still in its infancy and may not be as useful today as it will be when the body of XBRL data increases.  We should give the standard time to develop and grow before it is dismissed as simply a compliance burden.  

As I was reading through FEI's letter, I thought certain counterpoints ought to be considered in order to tell the full story of how detailed tagging can benefit investors (and therefore the companies they invest in).
  • FEI says that detailed tagging is a huge burden that will only increase in costliness. In my experience, detailed tagging is actually becoming easier and less costly as organizations become more familiar with the process and XBRL software improves. I also think that one of the most important measures of the quality of an XBRL software product is how quickly it can accurately be used to do detailed tagging.
  • The letter states that FEI estimates that 75% of detailed tagging data is not used by data consumers, according to research conducted by themselves and the SEC. In my experience, investors consider the uncommonly used data that they have found useful to make decisions about to be extremely proprietary, so it's not surprising that they only reported using the most basic of data.  
  • XBRL is useful for a lot more than just comparing a company's most recent quarterly report to that of its competitors.  Any time a number in a financial report is useful for any reason, there is a huge gain to productivity in being able to extract it automatically instead of having to locate it in a text document and manually rekeying it. If a number isn't important enough for users of the data to access it easily, it's probably not important enough to report at all. Since consuming XBRL data is still in its infancy, even if some numbers aren't currently used, if they are important enough to report, there is a good chance that someone will use them in the future.
  • While it is arguable that the least useful information in detailed tagging are the extension concepts, even company-specific extension concepts are comparable across multiple filings by the same company. An analyst following that company that is interested in that number can use XBRL-based analysis software to track that value over time, and perform calculations based on it. Another important purpose of extension concepts is that they allow taxonomy authors to see what values are commonly reported but don't yet have standard concepts.
  • FEI says that XBRL is hardly used, but it bases that on measurements of XBRL documents directly downloaded from company websites. Given that going to a company's website to download their XBRL is one of the most time-consuming and difficult methods of obtaining the XBRL, it's not surprising that those numbers are low. Downloads of XBRL off of company websites is probably not a good measurement of how much impact the XBRL has, since XBRL is more conveniently available directly from the SEC in several different methods.
Based on these critiques, the logical conclusion is that since XBRL is still in its infancy, it is not as useful yet as it is surely to become. It will be exciting to see how things develop and how this information will begin to be used.

Nathan Summers has been working with XBRL for five years and is a member of the XBRL International Specification Working Group. He is a developer for CompSci Resources, LLC. His opinions are his own.