Merrill Corporation - Dimensions eNewsletter - October 2012
Dimensions spoke by telephone in September 2012 with Susan Yount, Associate Chief Accountant with the SEC Office of Interactive Data, which is part of the Division of Risk, Strategy, and Financial Innovation. A CPA with over 20 years of experience, Ms. Yount joined the SEC in June 2009 to support the Commission's commitment to making financial disclosure accessible through the application of interactive data.
Has the phase-in of companies as XBRL filers been going as the SEC expected?
Yes, overall we've been really very pleased. As we expected, we do hear from filers that their initial tagging was time-consuming and the initial learning curve was steep. But the subsequent filings are taking less time as the process becomes integrated into their normal reporting cycle. We've also seen that the costs of these filings are generally coming down. As we expected, we also see some filers are starting to bring the process in-house as tools become more readily available and as they become more comfortable with the process. Now, at the start of the phase-in, I wondered if a full three years was really necessary. But smaller filers really seem to benefit from maturing software and the experience of early filers.
What is your overall assessment of the quality of the XBRL filings? Are you satisfied or frustrated with the quality, and why?
These filings are generally of very high quality, and we are generally satisfied with them. It's clear to us that filers and service providers have, for the most part, taken compliance with this rule seriously. One thing I've been surprised about: there still seems to be quite a few rumors making the rounds about how to comply with our rules. Some things we hear are just odd.
What are some of those?
Here are two rumors that I think are odd: that calculation linkbases aren't required; or that passing EDGAR validation is all you need to do to have a compliant filing. Now, of course, linkbases are clearly required, and passing EDGAR validation does not mean that the filing is compliant. But there's no need to guess. You can read the rules. You can read our guidance. If you're still unsure, you can send us an email at firstname.lastname@example.org, and we can provide guidance to you.
How quickly is guidance provided after someone asks you a question?
It depends. If the question relates to a filing that is imminent, we can respond to those quickly. If it is a request for a rule interpretation, those can sometimes take a couple of days to turn around. So if you have a question, it's best to come in to us as soon as you have that question.
The SEC staff has published helpful observations on common errors. Which errors by filers bother you the most? Which recurring mistakes are you surprised to see?
Well, I'm surprised that filers haven't done more to comply with this guidance. But I am particularly concerned that [some] service providers are positioning themselves as experts and selling outsourcing services that continue to contain these very basic errors. I don't know if people are not sure how to find this guidance, but you can find it on our website, at http://xbrl.sec.gov. There are still some pervasive errors that we've been publishing observations about since 2009, and I find it confusing that these errors still exist since we're more than three years into this process. Now that the phase-in is complete and limited liability is ending, and regulators and investors are starting to use the data, this would be a really good time for filers and their service providers to review that guidance and make sure they're not continuing to repeat those errors.
What are the top errors that you are most shocked to still be seeing?
The ones that I spoke about most recently at the XBRL US conference had to do with inappropriate use of negative values, inappropriate use of extensions, and missing or incomplete calculation relationships. But I'm highlighting just a few of the things that we've said in our staff observations. [Editor's Note: See also the lead articles in the May and June 2012 issues of Dimensions devoted to these observations and common errors.] With those three things, I'm highlighting things that should be fairly easy to identify and fix, but that doesn't mean we aren't still concerned about the other things we've talked about in those observations.
Is the SEC sending out comment letters directly to SEC registrants on XBRL problems in filings? If so, what types of problems would generate written SEC comments?
Comment letters have already been sent to filers about their interactive data filing requirements and accounting issues that we have identified during a review of their XBRL filings. When we started the phase-in, we were careful to give filers and service providers plenty of time to figure this all out-even going so far as to give them a few years of limited liability on their filings. But now the phase-in has ended, limited liability is ending, and our expectation is that filers have the tools and the knowledge they need to create fully compliant filings. Especially now that we at the SEC are using the data more, and investors are using the data, we're very focused on the quality of these filings.
What specific situations have prompted comment letters?
The majority of these comments have focused on missing XBRL files and missing website postings.
What is your message to noncompliant filers?
We have heard from filers that they do not believe that the SEC is looking at these filings. I want to make it clear that we are looking at these filings, and in fact we are looking at every filing that comes in the door. We are very concerned with the quality of those filings.
Will the SEC take any action if a filer is no longer in the limited liability phase and a file is submitted with significant errors?
The Commission could bring an action against a filer even if limited liability applies, depending on the circumstances. Filers should be aware that limited liability applies for a limited time and, generally, only to filers that make a good-faith effort to comply with our rules and that promptly correct any failures when they become aware of them.
What's your response to a CFO who wonders whether the benefits of implementing XBRL justify the time and expense?
A significant purpose of this program is to help investors get filers' financial information, because we believe that complete, accurate, reliable financial information is essential to the protection of investors and to the proper functioning of securities markets. Specifically to a CFO, I would say this: Investors are using XBRL data today to help them make decisions about investing in your stock. From a CFO's perspective, investors are not only getting more accurate, more timely information: now they are also getting your message. Before XBRL, investors for the most part got their information from third-party data-aggregators. This means that your financial information was retyped and mapped to perhaps generic classifications by someone you don't know, who is often not familiar with your company. All that time spent developing your messaging to investors can be lost in that process. XBRL is a way to get information to investors using the tags you select, so your communication about your financial results is preserved.
How is XBRL data being used internally by SEC staff?
We've used it to identify issues in the underlying accounting and disclosures in individual filings. We're starting to use it to answer questions about the overall quality of accounting and disclosures. For example, we recently analyzed all pension discount rates used by filers. We're using the data to help make risk assessments for filings. The FASB is also using this data in its accounting standards post-implementation review process, to gather data about how accounting standards have been disclosed. One thing we can probably all agree on is that, without data that can be gathered and analyzed in a timely manner, it's harder for us to determine if our regulations are having the intended result. With XBRL, we do analyze each filing as it comes in the door. When we look at things like, for example, pension disclosures, collecting that data by hand is very time-consuming. To collect things like pension discount rates automatically-is an exercise that might have taken us months but now might take minutes.
Are there any plans to adopt XBRL for 1934 Act filings and other items currently not mandated to be tagged, such as 8-K earnings releases, proxy statements, MD&A, or financial information outside the financial statements?
I can say that the requirements for tagged data are expanding. For example, we saw new requirements for tagged data in Dodd-Frank-one of which, for example, was the data around certain payments made by extractive industries. We have a taxonomy open for comment right now related to that rule. [Editor's Note: See News Briefs & Updates in this issue of Dimensions for more on this and on new SEC Form SD.]. We're seeing interest in tagged data beyond just financial statements and beyond just the SEC. I think what might drive the expansion of tagged data in the information that we take in is that, as investors continue to use this data, they're going to start to understand the potential benefits more fully-what kind of analysis is possible with tagged data that they haven't had before. I would encourage investors to let us know if they would find value in having additional information tagged, and if that would make it easier for them to analyze the data that they get from us, compared to what they're doing now.
So you want to hear comments from the XBRL user community?
Absolutely. We love to hear from the user community.
Is there any progress on SEC approval of an IFRS taxonomy that would allow companies which prepare financial statements in accordance with IFRS to file interactive data with the SEC?
Today, different modeling in the IFRS taxonomy would lead to filings which can't be used by software developed for filings using the US GAAP taxonomy without significant program modifications. The whole notion of data transparency implies a common data modeling standard, and one solution to this problem would be to create an “adapter” that would allow each taxonomy-any taxonomy-to be mapped to a common modeling standard. The analogy that we use is that this is just like the power adapters which are necessary in different countries. When you take your laptop overseas, you take your little cord adapter with you. It's all electricity coming out of the wall, but it may need to be adapted before it can be used. The SEC and IFRS staffs are working now on an adapter. Ultimately, the timing is unclear. But we think it's important for investors to be able to access tagged data using a single software platform, regardless of what taxonomy the filer is using.
When will the SEC make inline XBRL (“iXBRL”) an optional filing format?
We're currently considering whether allowing filers to submit in inline format would be beneficial to both filers and investors. We're aware that the rendering of the interactive data still seems to be an issue. We've said that the rendering won't match your original filing and that what we're interested in is your data, not the rendering, but we see a lot of filers spending a lot of time changing their data to try and achieve a particular rendering result. This generally results in poor quality or noncompliant data. We're also aware that the current state of affairs requires filers often to maintain their financial statements and footnotes in two separate formats-identical data but two separate formats. Maintaining that is difficult, it's time-consuming, and it's prone to error. We think inline XBRL removes some of these issues, and we think that filers might find that helpful. We would like to hear from filers whether they think something like an inline format-going from an HTML version plus an interactive data document to one single document-would be helpful for them.
What other benefits might iXBRL have for companies, investors, and consumers of financial data?
We think that we really might get better data from filers using inline XBRL, because it takes that rendering question off the table. As an accountant, I understand the desire to make things match visually, because that's how we're trained-that's what we understand. In a data world, that's no longer valid. That's been very hard for filers to get their heads around. We'd like to take that piece off the table. We are considering this as a voluntary format, because we think at this point that filers might want to decide whether they think it's helpful or not. We think that it might particularly benefit smaller filers, who may have fewer resources available to them than larger companies.
Would the SEC's review of filings change with iXBRL?
What we're talking about is just a different document format. The data would still need to be tagged, but it would reside in a single document. We would still bring in the same information, but in a different format-like using a different version of a word-processing program.
Any final message on the current state of XBRL filing, and its future?
The Office of Interactive Data has spent a lot of time talking about the promise of XBRL-more, better, faster information, with tagged financial data available at the same time as the underlying filing. Until now, we've asked filers and investors to imagine a world with that kind of data available. Today, we don't have to imagine it. I'm happy to report that the phase-in period of Regulation ST, Rule 405 (which requires the inclusion of XBRL-formatted financial information with many operating company filings), is now complete. All operating companies are now fully tagging their financial statements and footnotes, and we've received over 50,000 filings, containing almost 30 million financial reporting data points. Of course, we're not done yet-more needs to be done to improve data quality, and more needs to be done to make the data readily available to investors. But today, it feels like we're standing on solid ground to take those next steps.
NOTE: The views expressed here are entirely Ms Yount's and do not necessarily reflect those of the SEC or any other organization.