Speech by SEC Staff:
Drilling for Disclosure: The Powerful Tool of Interactive Data


John W. White

Director, Division of Corporation Finance
U.S. Securities and Exchange Commission

AAPG/SPE International Multidisciplinary Reserves Conference
Washington, D.C.
June 25, 2007

Thank you, Pete [Rose]. I am very pleased to be here. My topic today is the SEC's interactive data initiatives. The power of interactive data is one of Chairman Cox's true passions at the SEC. He will be joining you later today. Ask him about it - you will see what I mean! Trying to follow Pete's direction to "make it light," I thought I'd kick things off by approaching this topic using concepts that you are all familiar with: drilling, extraction, production, and reserves. Please bear with me if I use a little poetic license.

I am going to discuss how companies, investors, and other market participants can use technology to drill into financial disclosures, extract precisely what they desire, produce faster, better, and less expensive information, and emerge with a critical resource, reserves of investor confidence, transparency, and understanding. I know it's a bit of a stretch - but I'm trying - its lunch - and Pete said make it "light." And, I will say a word or so about real reserves as I close.

Nearly three years ago, the SEC announced that it would study and test the feasibility of, and potential for, using tagged data in company filings.1 Since then, the SEC and private industry have made great strides in encouraging public companies, mutual funds, and private entities to explore and promote the use of interactive data in filings with the SEC.2 Just last week the Commission expanded its voluntary mutual fund interactive data initiative to include risk/return summary information. Interestingly, an industry participant, the Investment Company Institute, was essential to that effort - it developed the taxonomy that is the lynchpin of that project.3

Before I explain what tagged and interactive data is (and what a taxonomy is), I need to make sure you understand that the Commission, as a policy matter, disclaims any private remarks of any of its employees. I am speaking solely for myself today, and my comments do not necessarily reflect the views of the Commission, of any individual Commissioner, or of any members of the SEC staff, other than myself.

So let me tell you more about what interactive data has to do with drilling, extraction, production, and reserves in the information world. Interactive data, or XBRL, is a very powerful tool - a revolutionary drill bit, if you will - that individual investors, financial intermediaries, analysts, the financial press, and others, can use to access and transform the wealth of publicly available company data. Using interactive data, you can transform static, text-based information into dynamic reports that allow for personalized, rich interaction. It allows you to compare information within filings and across different companies and different reporting periods.

Let me explain interactive data benefits using concepts you are all very familiar with.

The Drilling Process

Let's start with the Drilling Process.

As you all know, public companies file a great deal of information with the SEC. While that information is available electronically, at no charge, to the public through the SEC's EDGAR system, it is available only in 'electronic paper' formats, which are static and often dense documents. When companies tag information in their filings, XBRL allows you, for the first time, to use software to drill down into the information in these filed documents. XBRL allows you to search for particular items of disclosure that have been tagged in a company's filing, retrieve that information, and analyze it. I will explain the critical tagging process and what's happening on that front in a few minutes. But first, let's talk about what I'll call the software drilling tools.

The SEC has supported private sector efforts to develop drilling tools for XBRL.4 To stimulate private, market-driven software development for viewing and analyzing tagged data, the SEC released its first XBRL prototype tool last December. The tool, called the Interactive Financial Report Viewer, is available, free of charge, on the SEC's website. Using this tool, not only can you view tagged financial statement data, you can compare tagged data from different companies and create your own graphics. I'll show you that in a few minutes.

The SEC plans to release a second prototype tool, with enhanced capabilities, later this summer. The source code for both of these prototypes will be available, free of charge, on the SEC website. In addition to these drilling tools for financial statement data, the SEC intends to launch an exciting new feature on the SEC website this summer to provide yet another example of the power of interactive data.5 This feature will highlight how XBRL can make it easier for investors to get information on public company executive compensation disclosures and will include such data for several hundred of the largest companies in America.

The SEC continues to encourage private entities to expand upon these initial efforts and create additional software with more extensive features. These software tools - whether created by the SEC or private parties - will be able to use filed tagged company data directly from the SEC EDGAR filing system.

The Extraction Process

With continually improving tools, companies, investors, and the market will be able to take the next step - the process of extracting the data.

Instead of requiring a reader to wade through the dense public company documents and manually identify information important to that individual, XBRL allows extraction of the information that an individual wants and stores it in a format his or her computer can quickly retrieve. Using the software tools of the reader's choice, he or she can choose the final format of that data.

XBRL isn't magic. It doesn't know what information is important to an individual nor does it know what format that individual wants to see it in. In order for tagged data to be useful, there must be a standard protocol for companies to follow when they decide what to tag and what tag to assign to each specific type of data. Individual users must understand what data is associated with each tag, so they know what information to extract from a company filing.

So that gets me to step one of the XBRL process, and perhaps the most significant part of the SEC's initiative, at least to date. This involves developing XBRL taxonomies for U.S. GAAP.6 For those of who have not been following this initiative, you're likely asking yourself right now "What is a taxonomy?" Think of it as a dictionary of computer labels - each label has its own definition. In a taxonomy, each data tag, or label, is defined as being associated with specific information resulting in a collection of definitions - just like a dictionary.

Once we have a full set of taxonomies for all items in U.S. GAAP, every item in a company's financial statements, such as net income, net sales, or advertising expense will be assigned a unique data tag, or a computer-readable label, that will represent the same information in each company's financial statements.

Let me read a few words from the SEC's website to you:

"Think of every fact in an annual report, every number in a company's financial statements, as having a unique barcode that tells standard software what the item represents and how it relates to other items in the report. Interactive data 'tags' all of the key facts in these large documents, so that software can instantaneously recognize them and serve them up to the investor."7

XBRL-US is leading the U.S. GAAP taxonomy project. XBRL-US is working closely with a wide array of industry groups and the SEC so that the finished taxonomies meet the needs of all constituencies. Taxonomies have been written for many industries already and when the work is completed this fall, any company in any industry will have the basis to report their financial statements, including the notes to the financial statements, to the SEC in XBRL format. In order to provide consistent application of these taxonomies, they will be accompanied by robust user guidance.

As I just mentioned, using these taxonomies, public companies will be able to present their financial statements, including the notes to the financial statements, in XBRL format. Taxonomies for the oil and gas industry are in development and, like the taxonomies for all other industries, will include data tags for all U.S. GAAP financial statement and footnote disclosure. XBRL-US also is considering whether it should develop data tags for the Supplemental Information on Oil and Gas Exploration and Production Activities required by FAS 69.8 The taxonomies being developed by XBRL-US are intended to be made available by them for public comment before they are finalized, and I'd encourage everyone here to get involved in that process, so the taxonomies will adequately address your industry's reporting needs.

The XBRL developments I have been discussing are not unique to the U.S., but are occurring worldwide. There is a global movement to use interactive data to share and analyze many different kinds of financial information. To name a few, the Tokyo Stock Exchange, the United Kingdom, and Singapore are developing XBRL requirements for certain securities, banking, or tax filings.9

The growing trend toward truly global, high-quality accounting methods has heightened the need to converge interactive data taxonomies and software tools to allow for cross-border analysis of companies. Last week, the Commission voted to propose rules to eliminate the requirement that foreign private issuers who prepare financial statements using International Financial Reporting Standards (IFRS) (as issued by the International Accounting Standards Board (IASB)) reconcile their financial statements to U.S. GAAP. Those rules, if adopted, could go into effect for as early as 2009, with the result that those foreign private issuers will simply file their financial statements in the U.S. using IFRS (without reconciliation to U.S. GAAP). The Commission also has announced plans to issue a concept release to solicit comments on whether, eventually, U.S. issuers should be permitted the choice of filing their financial statements using U.S. GAAP or IFRS as well.10

As IFRS continues to be adopted internationally, with the ongoing efforts of the U.S. Financial Accounting Standards Board and the IASB to converge the standards and principles of U.S. GAAP and IFRS, the need to develop a robust, comprehensive XBRL taxonomy for IFRS has taken on increased importance. The U.S. GAAP and IFRS taxonomy teams are working to align their taxonomies so that investors and other market participants will be able to use the same tools to evaluate companies, whether they report in U.S. GAAP or IFRS.11 This work is helping the accounting standards convergence project by highlighting where standards are converged and where they are not.

In addition, the International Accounting Standards Committee Foundation has been a leader in encouraging the use of interactive data, and has committed significant resources to creating XBRL taxonomies and other tools to make financial information easier to use. This April, the IFRS XBRL team released a Spanish translation, and they've followed that last month with a French version.12 These rapidly moving developments in the use of interactive data internationally will help global markets in insuring that IFRS is interpreted and applied consistently across borders.

The private sector must take credit for developing interactive data. Through the efforts of XBRL International and its global affiliates, as well as other private sector leaders in this area, there is a "revolution" coming in the way companies and market participants exchange financial information. As I just mentioned, XBRL-US is developing the U.S. GAAP taxonomies with significant industry input. Once the U.S. GAAP taxonomies are completed, I expect a next logical step will be to develop taxonomies for narrative disclosures in SEC filings. The Investment Company Institute is taking the lead in developing taxonomies for certain narrative and other non-financial statement disclosures in mutual fund prospectuses. And a consortium of accounting, media, and business experts with leadership from former Harvard Business School Professor Robert Eccles, is developing interactive data taxonomies that will cover key performance indicators that are not required by the SEC, but that will be enormously useful to market participants.13 Mr. Eccles is Chairing a breakout session this afternoon on "Legal, Ethical, and Reputational Aspects of Reserves Estimation and Management Use." I encourage other industry groups, such as yours, to work collaboratively in developing narrative disclosure taxonomies for corporate filers.

So that's it on taxonomy development. Let me shift back to the use of interactive data.

The Production Process

When talking about interactive data and XBRL, the SEC staff, quite appropriately, tends to start with, and focuses on, the benefits for investors - and those are many, as I have suggested. But there is another angle in looking at interactive data's benefits. Interactive data also can be a powerful tool for companies in producing their disclosures.14

In disclosure production, preparers of disclosure will be able to use XBRL to produce and quickly analyze their disclosures, topic-by-topic, in at least three significant ways:

  1. Companies can use XBRL to quickly compare related disclosures within the document they are preparing. (As you know, a company may disclose the same information in different places and in different contexts within one public report.)

    XBRL will allow a company to quickly and confidently achieve consistency and accuracy prior to filing a disclosure document with the SEC.
  2. Companies can use XBRL to quickly and confidently compare current disclosure to past disclosure. Presumably, in the preparation process, companies today are, in many cases, already comparing their current disclosures to their past disclosures (for example, comparing particular items in MD&A in a current 10-Q to what was said in prior 10-Qs). XBRL will make that comparison faster and easier.
  3. Companies can use XBRL to see what their peers and competitors have said about similar items. A side-by-side, company-to-company disclosure comparison is a very powerful XBRL tool.

In all three cases, investors and analysts will be able, and are likely, to do the same thing, almost instantly. Using XBRL is a good way to anticipate their reactions to your company's disclosures (and to enhance your company's disclosures in advance), as well as be prepared for their inquiries.

If you'd like to hear more on this angle on using XBRL, the SEC has held a number of XBRL roundtables, which are available on our website. I direct you, in particular, to a segment near the beginning of the March 19th roundtable, where I gave a demo on the benefits to companies of using XBRL in producing their SEC filings.15


Finally, I know how important reserves are in the oil and gas industry. Equally important are the reserves of investor confidence, transparency, and understanding that I believe will flow from the use of XBRL.

By providing a fast and easy way to extract, analyze, and compare financial information, XBRL has the potential to create a more informed investor base and foster increased transparency for market participants by making it easier to find and analyze these public company disclosures. Increased transparency of disclosed information using interactive data will further enhance the ability of market participants to evaluate and value public companies and make informed investment decisions. Investors will have access to a wealth of financial information through tagged data in a manner that is simply not possible today. The information is there - it's just not easy to get to.

Finally, let me say that there is no reason for companies to delay tagging their data. The interactive data initiative is one of Chairman Cox's key priorities. There is an ever-expanding group of volunteers that submit their annual and quarterly reports to the SEC using tagged data.16 Companies continue to join the more than 40 XBRL pioneers (the group includes Microsoft, General Electric, 3M, United Technologies, Ford, Pfizer, Xerox, PepsiCo, Bristol Myers, Lockheed, EMC, Dow Chemical and ADP ). The volunteer group currently includes two oil and gas companies - Anadarko Petroleum Corporation and Brazilian Petroleum Corporation - PETROBRAS.

While these filers are using XBRL on a voluntary basis, the future obviously could hold a requirement that tagged data be used in public company filings with the SEC.

For my part, the Division of Corporation Finance will continue to be a key player in the SEC's work with XBRL. If the time becomes appropriate, we stand ready to draft the needed rules requiring tagged filings. We also look forward to being users of XBRL in carrying out our review responsibilities.

Now to the demo.

XBRL In Operation

To illustrate how XBRL works, I am now going to walk you through some examples, using the Interactive Financial Report Viewer accessible from the SEC website and the XBRL submissions of the two oil and gas companies who are participants in the voluntary program -- Anadarko Petroleum Corporation and Brazilian Petroleum Corporation - PETROBRAS. This is not an actual live demonstration, but these slides are recent screen captures, using the Interactive Financial Report Viewer and XBRL tagged data of these two companies -- all on the SEC website today. (Of course, this also is not an endorsement of these companies or their filings that appear in the EDGAR database.)17

[Slide 1 - Logo]

[Slide 2 ] This first slide is the Interactive Financial Report Viewer that was launched last December and is available for free on the SEC website. This viewer provides a first glimpse into how users will be able to access and work with interactive data. The second prototype viewer to be launched later this summer will further showcase the possibilities of interactive data.

[Slide 3] All of the filings made through our XBRL voluntary program are available via the viewer. As you can see here, I've selected Anadarko Petroleum Corporation's annual financial statements from 2006. The income statement data is displayed here as you would see it in Anadarko's annual report on Form 10-K.

[Slide 4] Using interactive data, it's effortless to navigate between various sections of the document that are defined in XBRL taxonomies and, using the section headers at the top, I've now navigated to the balance sheet section.

[Slide 5] Of course using interactive data means you'll have greater capabilities to use data beyond what is possible with today's static formats. I'll show you a few examples of these capabilities using our viewer. The private sector is expected to greatly expand upon those capabilities.

I'm now back on the income statement and, as you can see, you can use the viewer to graphically chart the data based on individual user preferences. In this case, let us build a chart showing the sales trends for oil and condensates, natural gas liquids, and gas for the past three years.

[Slide 6] …and instantly you have the results. This can be done for any of the tagged information provided in the company filings.

[Slide 7] Let's go back to the Anadarko income statement. In addition to working with a single company, the viewer has the capability to perform comparisons across companies. On the left side of the screen is an area to select the companies and reports to compare.

[Slide 8] And here we have the comparison between the income statement of Anadarko and the income statement of Brazilian Petroleum Corporation - PETROBRAS, a foreign private issuer that is the other voluntary oil and gas XBRL filer.

[Slide 9] If we scroll down, you can see that there are some "company specific" items that can't be compared. Part of the reason for the inability to compare some of these company specific items is because each voluntary XBRL company had to extend the base taxonomy to include company specific disclosures that were not included in the base taxonomy.

This is an issue today because the taxonomy is not yet complete for all oil and gas industry disclosures. As I mentioned, XBRL US is working on the completion of taxonomies for all industries and when complete later this year, the industry taxonomies should lessen the need for extensions and allow greater ability to perform company-to-company comparisons. There remains ample opportunity for your involvement either in the development of the taxonomies or for review during the public comment periods that XBRL US is planning - I encourage your industry involvement so that the developed taxonomies reflect your unique reporting needs - this is an opportunity for you to participate in that process.

[Slide 10] So far we've only been focusing on financial data, but interactive data and XBRL also can be used to report non-financial data as well. Anadarko provided their FAS 69 Supplemental Information on Oil and Gas Exploration and Production Activities schedules in XBRL also - and as you can see here, that information also is easily accessible via our viewer.

[Slide 11] And just like the financial statement data, we can chart this information as well. Using the charting tool in the viewer, you can select the periods and data you want to chart.

[Slide 12] So here's a chart for the trend in proved and unproved properties by location.

[Slide 13] Similarly, when other companies in the oil and gas industry begin reporting their supplemental information in XBRL, you also will be able to perform company-to-company comparisons of this data - the blank table on the screen would be filled in.

[Slide 14] Logo.

That's the end of my demonstration. We are still at the early stages, but I hope you can appreciate the potential of XBRL and interactive data.

Division of Corporation Finance Update

I didn't want to leave without saying at least one thing about actual oil and gas reserve disclosure. I know that is an area of significant interest to all of you and that many of you have expressed concern about the SEC's current disclosure requirements. I am aware of a need for a review of our rules with a view toward updating.

So, how do we get started? We'd like some more in-house expertise. As some of you may already know, we are in the process of interviewing applicants for an academic engineering fellow to join the Division staff for the upcoming academic year. As you can imagine, academic fellows bring a unique perspective to our disclosure program - they often help us look at things in different ways and offer us new approaches to considering complex matters. While I won't speculate what this ultimately means for our disclosure requirements, I can say it is a first step. Beyond that, all I'm comfortable saying at this point is stay tuned. And stay involved in these very important discussions about reserve disclosure.

Also, linking back to what I have been discussing today, one thing that does seem clear is that, because of the special complexities related to reserve disclosure, investors will particularly benefit when reserve information is available in a format that can be managed and analyzed as easily as possible.


I hope I have highlighted for you the powerful benefits that XBRL provides to public companies, investors, and the markets.

As the global support and development of XBRL increases, I expect that investors, analysts, and the markets will strongly encourage public companies to use interactive data or XBRL when presenting their disclosures, even those beyond the financial statements.

I encourage you to consider how you, as public company representatives, oil and gas specialists, and market participants, can help further the interactive data initiatives in your industry.

Thank you for inviting me and listening today -- Interactive data is real priority for the Commission.