Understanding XBRL

Here I am at the Hot Topic Discussions, in the group called Understanding XBRL. I’m a bit late as I accidentally went to the Governance talk instead. I felt a bit like that kid in Geology class on the first day who has to scuttle away saying “I thought this was Psych 101″.

Anyway, I don’t have the speakers names at my disposal, but there was a fellow in the know from the SEC who is involved in the “Interactive Disclosure” group, who was terrific. (Update: his name was: Jeff Naumann, Office of Interactive Disclosure, U.S. Securities and Exchange Commission). There was a group of us NIRI folks, all sitting in a group-therapy style circle of chairs, who peppered him and the other two discussion leaders (I’ll look for your names — sorry!), with what I thought were good questions.

So here they are, with their responses as well as I could transcribe them:

What’s the difference between XBRL and “Interactive data”?
Same thing, and the SEC prefers to call it the latter.

What’s this group XBRL.US?
This is a group of users and policy developers who are designing the XBRL system and creating the Taxonomy upon which XBRL will be based. Their site is a clearing house of information and resources, and so a terrific place to start in learning about this.

Why is EDGAR so pro XBRL?
They see it as a logical evolution of their purpose. They started with paper filings, moved to electronic filings in ASCI text, then HTML, then XML about 5 years ago. And now XBRL represents the next major expansion. (They also see it as a way to greatly reduce their workload.)

What documents will need to be in this format?
First will be financial statements, next other categories of reporting documents. For instance, in the first version Notes to the financials will each be tagged as a single blob of text; in the next rounds, each number and percentage will eventually be tagged individually.

What’s so good about this XBRL system?
Financial stements tagged using the XBRL taxonomy is easy to move quickly into XBRL readers and financial data aggregators. This replaces the interim step of say, moving all the quarterly results into financial models and spreadsheets by copying and pasting from a PDF.

This also allows retail investors to use a third party XBRL reader pre-configured with financial analysis routines, which will be instantly updated when the XBRL feed is received or data is downloaded. Voila, instant analysis!

Will XBRL be published in RSS format on company websites, and so be automatically syndicated to subscribers?
Yes, this is likely to be the approach used by many companies. As with current RSS feeds, this will automatically deliver the information to subscribers.

What do you mean by “taxonomy”? This is to do with “Tagging”, right?
This is simply the 10,000+ list of financial terms XBRL.us has brought together to represent all the line items in America’s current financial reporting library. It was based on US GAAP, the accounting disclosure checklist, and then was tested by applying the taxonomy to 100’s of real financials. This was done by some 100+ accountants assisting in the project who should probably be thanked at some point.

Why should we (companies) bother doing this (besides we’re going to have to)?
Your financial information is already being tagged, just not by you. Undoubtedly you could do a more accurate version then the current aggregators of financial data like Yahoo! or Google Finance . So, with the current, uncontrolled tagging of your financials, there’s many different versions of your financials out there.

Remember, many investors, especially retail investors, rely on these sources as their main sources of information, and likely think you’re the source of it.

The XBRL taxonomy, applied initially by you, will undoubtedly lead to greater accuracy in these aggregated versions. So this is clearly a good thing for IRO’s, no longer having to explain the discrepancy in their reporting!

Is there software available for reading XBRL?
Yes, prototype readers are available on EDGAR, XBRL.us, and SEC websites ( look for links to XBRL or “Interactive tools”). These can be downloaded for free.

Analytical tools are also already appearing. One can imagine the exotic readers that will likely soon appear.

What should companies do now?
Get the taxonomy in place now using current filings. The taxonomy can be put in place now, as you are un likely to change the way you report in the interim. The biggest slug of work is in applying the taxonomy (“tagging” your statements) the first time of course.

Is it an oversimplification to think this is a mapping table?
No that’s exactly what it is.

What’s the most difficult part?
The original mapping of terms; mapping your financial statements to the XBRL taxonomy.

Will making adjustments to ones taxonomy lead to disclosure issues?
The person asks in a hypothetical voice: “I’ve been mapping it this way, now I want to do it this way? Is this a disclosure issue: I need to remap my mapping”? (Not sure there was an answer to that one as everyone laughed a bit.)

Do auditors have to review the tagging?
Some are relying on auditors, some rely on service providers, some on their own accounting group. Under the voluntary program (1st stage of the SEC program), one half have done it themselves, half have used outside vendors. Even those using outside vendors will still need to review their tagging, of course.

How much judgment is involved?
In the supporting documentation, every term has a paragraph description and references to the authoritative source upon which the tag was created. There is a “Preparer’s Guide” (an exhaustive document apparently available on XBRL.us (I think) for assisting you in making your financial statements fit the taxonomy.

What gets tagged in the MD and A?
Nothing, not in the first phase.

How long does it take to prepare the tagging?
Anecdotally it sounds like about 80 hours staff time when working with an outside vendor, (sounded a bit low to the panel leaders). About 120 hours for companies who came up to speed on their own and did the tagging. Once the tagging is done the first time, it will be much, much faster exercise in future, of course. Most companies will probably re-use the same templates (as they do for their financials).

What else should we think about from the start?
Make sure others in your company (like the CFO!) are aware of this upcoming requirement for XBRL reporting. Understand where your company fits in SEC’s schedule for meeting this requirement. Start thinking about whether you’re a do-it-yourselfer or are requiring outside assistance.

What is the schedule?
(Sorry, I didn’t quite get the details on that down). Larger companies are required to meet the guidelines first, middle-sized bunch the second year, and everyone by third year (2010 — please check yourself!).

The concurrent schedule I believe was related to what had to be tagged, with Notes going in as “whole notes” in a text blob “Note 1″ = “one tag name”, etc, with every figure and percentage in the notes eventually getting their own tag, a bit later. It sounded like all the financial documents and reports would eventually be brought in, and early adopters may do that voluntarily.

How does the exercise of tagging unfold?
Companies will first put themselves into one of the 5 main taxonomies: 4 are for financial oriented companies (banking, insurance, etc.) and the final reserved for all the rest, called “Commercial and Industrial”. This is the “Which industry am I?” step.

The you will begin to go through each of you main financial statements applying the taxonomy, using the support documents, and probably your accounting team to slot items into the right term.

Do folks file XBRL statements separately?
They are not required to file at the same time currently, and can be added after the principle version of the documents are filed. They are a separate file format.

Can I read those documents myself?
These are “machine readable documents”, so you need a reader to make them “human readable”. Just like you need a browser to read HTML, eh?

What’s required for corporate websites?
Companies will be required to make the XBRL versions available on their websites.

Are other filings necessary, like Section 16, 144, Proxys, Executive Compensation (portion of proxy)?
No, but some demos have been done on other documents (on sites mentioned previously). And we should probably expect these all to be required within a few years.

I’m still freaked out by XBRL – won’t it help my competitors to understand our financial statements? (real question!)
This really doesn’t change anything that’s already available, it simply accelerates the broad strokes analysis; it’s just a tool for information extraction.

How will this affect IRO’s following a release?
It will likely lead to better questions from a larger number of shareholders, and these questions will come more rapidly following the release. You will essentially face better informed investors sooner, as more time can be put in by them into understanding the results of their financial models with less wasted on inputting the information.

What are the tagging tools like?
Some of the tagging tools are Word and Excel plug-ins.

Eventually tagging will be embedded in enterprise management, general ledger, SAP, etc. software. Microsoft already displays their financial information using XBRL data from their own SEC filings. This updates their site automatically.

What about the cost? Are companies, especially smaller filers, complaining about the additional costs?
Folks are estimating the initial cost of (I presume) the first round of tagging, to be between $30-70k. The biggest cost is staff time in getting up to speed, understanding the taxonomy and finally applying it.

The second time round, of course is far less onerous as the taxonomy is figured out, and the process should be limited to copying and pasting the tags in to the documents or vice versus.

Can’t wait to see the aggregation and analysis tools!