XBRL – Part Deux

Much to my surprise, my recent post about XBRL has generated more emails than any other post I’ve written.  People from London and Washington D. C. have offered to help me see the error of my ways, so I’ve decided to continue to try and see the light.  (But, I mean, who would have thunk? This seems like a pretty dry topic to me.) 

I’m at an age where I’m not embarrassed to say that there are some things I just don’t get.  For example, in Game Theory class I never did figure out who was wearing that damned ret hat.  Be that as it may, this stuff is like an unscratched itch and it bothers me until I get some semblance of understanding, dim though it may be. 

So I went back and did some more digging on the web and I found an article about the 18^th XBRL International Conference held in Washington, D. C. last October.  What caught my eye in the article was a presentation where a Microsoft official stated that his company had already installed a XBRL enabled tool on its Investor Central website.  The specific statement was that the site allowed investors to drill down into segment results that sum into top line numbers on the financial statement.  I thought, “Ah ha, here’s a concrete example of XBRL in use that can really help me understand what’s going on.”  So off I went to the Microsoft website to look at the way they report segment results in their Investor Central.

I’ve inserted a screenshot of what I found.  The first page under segment reporting gives you the breakout of revenue and operating income.

If you click on one of the line item headings, it takes you to another screen (shown below) that drills down into the line item detail and the language from the Management Discussion and Analysis section of the 10-Q that discusses that segment.  So far, so good – it doesn’t give you any more information or insight than reading the 10-Q, but it assembles the data in one spot so that you’re not confined to the logic of a governmental filing form as you try and figure out what Microsoft’s various business segments are doing.  Nice, but hardly revolutionary.  Then I took a closer look at the way Microsoft had laid out the information and I lost my appetite for XBRL. 

The segment I had chosen by random to look at was Microsoft’s Online Services Business.  Microsoft starts the page with a headline stating “Online advertising revenue grew 7% in a weak ad spending market” and follows it up with three bullet points about search revenue up double-digits, continued display revenue growth and healthy engagement growth in page views and search queries.  It all sounds very positive until you look down into the actual detail pulled over by XBRL and you see that actual revenue growth for the segment was flat and oh, by the way, operating losses increased by 91% to $471 million.  Maybe that’s just a rounding error to Microsoft and not worthy of a mention, but it sure seems like a lot of money to me. What Microsoft did was to take the opportunity created by having an additional, non-SEC filing page to create some positive, non-XBRL commentary in order to try and spin the results on a horrible quarter in the segment.  If this is XBRL, you can have it; I’d rather try and figure things out without the commentary trying to divert my attention elsewhere.

Also, curiously enough, something appears to be awry with the transporting of data, as the last sentence on Microsoft’s web page refers to “headcourt-related [sic.] expenses” whereas their 10-Q filing gets it right.  If this data is just being moved around automatically, how could that happen?  And how confident does that make you feel that all the numbers have been transported correctly?

Maybe I’m just a skeptic (regular readers of this blog knew that already), but I don’t see what all the fuss is about with XBRL.  It’s just rearranging the data that we already had, which to me seems like a lot of work for not much benefit.  Better understanding and benefit come from working through the data.  Balanced against this we seem to have given companies yet another opportunity to spin the data.

Then again, maybe I’m just an old curmudgeon and this is another instance of not being able to figure out where that damned red hat is…