THE US Securities and Exchange Commission (SEC) will meet on Monday (April 21) to consider making XBRL mandatory, probably starting with large companies next year. It’s not a surprising development by any stretch, but it is a momentous one.
The big question is who will be first in line to be required to use XBRL. The SEC has a non-unanimous recommendation from the Advisory Committee on Improvements to Financial Reporting for a staged roll-out of “furnished” rather than “filed” XBRL reports starting with the 500 biggest US companies next year, then another 1,500 companies a year after that, and finally making all companies use it in the “long-term.”
But Gary Purnhagen, writing on the Hitachi XBRL blog yesterday sees the SEC being more aggressive in the roll-out:
I believe we’ll see the SEC propose that the 1,200 largest domestic companies in terms of market capitalization be required to furnish XBRL exhibits beginning in 2009. A company will not have to make its first submission of XBRL exhibits with its 10-K; rather, it will be able to wait until its first 10-Q. These filings will not be official, hence the word “furnished” — just as the current companies participating in the VFP are “furnishing” their financials tagged with XBRL in exhibits.
Having the financials furnished will also alleviate the need to have them audited. Clearly, the accounting industry is not ready for that, and at this time having to do so would make this new ruling very expensive. The main difference between the current volunteers and the companies being mandated to submit XBRL exhibits will be that, under mandate, companies will be required to tag footnotes to financials on a block level.
The proposed rules will also offer a phase-in for the rest of the SEC registrants, probably giving this first group a 12-month period of time before the next wave of phase-ins. The SEC will break up the phase-in groups by:
• The rest of the large accelerated filers (market cap over $700 million)
• Accelerated filers (market cap over $75 million)
• Non-accelerated filers (market cap less than $75 million)
It doesn’t matter really because whether it’s 500 or 1,700, it’s probably going to be more than that because companies are going to be under pressure to use XBRL if their peers are providing it and analysts start asking for it.
XBRL is definitely NOT one of those things you can “worry about closer to the time.” It’s not something you should just fob off to outside vendors without first knowing what that decision means in the long term.
So seriously, start panicking. Get a committee together inside your company to start discussing the topic. Learn about it as much as you can so that you can ask intelligent questions and make good decisions.
XBRL is incredibly complicated (15,000 concepts and 20,000 relation in US GAAP XBRL). If you want to learn about it, really understand it, I suggest you try Charlie Hoffman’s Mastering XBRL course. He is the father of XBRL and his course is designed for “CPAs, chartered accountants, and other business users who have some knowlege of accounting.” As prerequisites he lists the following characteristics:
- You are not scared of technology. In fact, you are likely curious and like to figure out how things work.
- You probably can program or want to learn how to program, probably using Microsoft Excel or Microsoft Access.
- You understand and use SQL.
- You may be a consultant or an “internal consultant” (the person everyone else in your organization comes to to solve problems).
- Etc.
Finally, this is going to be expensive, especially if you leave it to the last minute. I don’t care what the SEC says about the costs of XBRL or the relative ease of implementation. They don’t have accurate pricing to go on because the costs and experiences of the voluntary filers are irrelevant.
There is a scarcity of XBRL expertise. No one can accurately predict the impact that a sudden surge in demand for very scarce resources is going to have on pricing.
Finally, IR website managers and their vendors should start thinking now about how they will provide and present the new XBRL data on their websites. Filing, sorry furnishing with the SEC is just one part of the job.
Excellent post, and the reference to my post isn’t bad either. I’d like to comment on some of your points. First, you’re right in that the initial number of companies being phased-in does not matter. Once you have a compelling data base of 500 of the top companies, analyst and small investors will begin to use it. This will create a need for better consumption tools than the SEC’s funded on-line tools, which will further demonstrate the power of XBRL. This in turn will drive companies that want better analyst coverage to begin providing their financials tagged in XBRL. As I’ve been saying, if Cox builds it they will come.
You also make a good point on the cost. Many of the volunteers have gotten a deep discount if not a free service to date just so their vendors could say they provide XBRL. 70% of the Fortune 1,000 use one of the big three financial printers for some SEC EDGAR filing services. We can expect that a similar percentage of the Fortune 1,000 will indeed turn to their financial printers for XBRL. It’s interesting to note that at this time the three have very different approaches to assisting companies with XBRL tagging. Who’s to say their selected approach is the right one for the company looking to furnish their financials with XBRL. And you are right on with your concern regarding the sudden surge in demand overwhelming these vendors. Not only could it affect price but it could also affect quality.
My last point is regarding your past commentary regarding the SEC’s Notice and Access ruling. When the SEC made the argument that the electronic version of the proxy and other annual report material was inherently better than the printed version do you think they were talking about HTML or PDF? I think not, it was always XBRL they were referencing.
Hi Gary,
Great piece by you on the Hitachi XBRL blog. When the SEC referred to the e-proxy formats, I really thought they were asking for HTML and PDF. Actually, they were explicit about that. What we got was anything but HTML, which has been incredibly frustrating to watch. Frankly, the way e-proxy has been botched is why I’m urging people to start paying attention to XBRL with a sense of urgency.
So many bad decisions have been made around e-proxy precisely because people did not plan ahead and because they weren’t capable of telling the difference between a .gif and text, or a cookie that improves usability and one that infringes on a user’s anonymity.