Related Topics:
			Business Research: ERP, Finance Vendor Research: A3, AcornSystems, Adaptive Planning, Applix, Business Objects, Business Objects – ALG Software, Cartesis, Clarity Systems, Coda, Cognos, FRx Software, Hitachi America, Hyperion, Infor – Extensity/Systems Union, KCI Computing, Lawson, Longview Solutions, Microsoft, Oracle, OutlookSoft, PROPHIX, SAP, Edgar Online, UBMatrix
			Printer friendly version Email this article Send feedback to editor

Summary
The United States Securities and Exchange Commission (SEC) has been considering a requirement that public companies submit their periodic financial statement filings with eXtensible Business Reporting Language (XBRL) tags that its Electronic Data Gathering, Analysis and Retrieval (EDGAR) database can read and store. Tagging such data would make it easier for individual and professional investors to analyze a company’s results, particularly to compare companies on an “apples to apples” basis. Yet in the wake of implementation of the Sarbanes-Oxley Act and shortened filing deadlines, the SEC was reluctant to place another regulatory burden on public companies. Moreover, until recently there were some serious obstacles to adopting XBRL, including an immature accounting statement taxonomy and a lack of easy-to-use tools.

But now Ventana Research thinks it is time for the SEC to make XBRL tagging mandatory for basic financial statements (income, balance sheet and cash flow). After testing XBRL tagging products to learn what the process would require, we conclude that people will find it straightforward, inexpensive and not unduly time-consuming. However, public companies initially will need some guidance from auditors and securities analysts on how their figures should be tagged. We believe a “dress rehearsal” involving a coordinated effort and dialog between these parties would make the introduction of XBRL a smoother, less painful process. The result certainly will be worth it.

View
How burdensome would it be if the SEC required all public companies to include XBRL tags in their filing documents for the basic financial tables? The answer, based on an assessment we recently completed, is, not very. We wanted to time how long it would take to create the basic tags for the income statement, balance sheet and statement of cash flows. To conduct the test, we used seven public companies’ financial statements, representing a range of industries and from large and complex corporations to smaller and simpler ones. They were Administaff, Boeing, Brush Engineered Materials, General Electric, Gerber Scientific, Gray Television and Kroger. One reason for choosing a range of industries was to see how often we had to create our own tags because there was no appropriate one in the standard taxonomy. XBRL was designed from the start to work with nonstandard tags (that’s how it is “extensible”) in order to achieve maximum flexibility. However, in our judgment, using custom tags for public company financial statements negates an important feature of XBRL tagging: enabling easy comparisons across companies’ financial statements.

To test how long it might take a finance department to apply XBRL tags to its data for electronic filing, we used tagging tools provided by two software companies, Hitachi and Rivet Software. Rivet has a generally available version of its Dragon Drop tool, but we used beta versions of both it and Hitachi’s. (Another vendor, UB Matrix, is also in beta with its Reporting Manager.) We received minimal training in the use of their tools, mostly in the form of short demonstrations. Realistically, though, we expect it will take anywhere from one to five hours of initial training to prepare people who understand accounting and financial statements to use these tools. In our judgment, the amount of time will depend on their degree of familiarity with the basic concepts of how XBRL works. To do the tagging, there is no need to have a deep understanding of taxonomic structures, of XBRL or XML coding. Knowing how financial statements work is far more important.

A great deal of work has gone into building out the U.S. Generally Accepted Accounting Principles (GAAP) taxonomy that organizes the accounting concepts and drives the “magic” of XBRL. For example, the taxonomy “knows” that the allowance for doubtful accounts is a component of net trade receivables, which is a component of all receivables. It knows that “receivables” is a “current asset” that is part of the larger set of assets; that therefore it is a debit value as of a given date, expressed in millions (or, if you prefer, thousands) of U.S. dollars, and so on. It also “knows” how to roll up and calculate subtotals. The taxonomy contains most of the financial intelligence needed to make XBRL tagging work, and is scheduled to be updated over the next couple of months. While the current income statement and balance sheet definitions are relatively comprehensive, the cash flow statement portion of the U.S. taxonomy is incomplete. For that reason, we decided against including this part of the financial statements in our test.

It took us slightly less than four hours to tag the first company’s (Boeing) quarterly income statement and balance sheet. It took an hour and a half to do the last (Brush Engineered Materials). Of the group, General Electric was the most difficult, taking five hours because of the larger number of items that required tagging. GE’s business includes both industrial and financial services components, so its periodic financial statements present these two elements and a consolidated view. Based on our test, we estimate that on average, it will take the person responsible for tagging about five hours (plus or minus an hour or two) to become familiar enough with the tagging tool and the process itself to achieve basic proficiency. It will then take the person another five to 10 hours to do the initial tagging of the company’s basic financial statements. Once that template is established, though, it should require no more than an hour or two each quarter (and for the proficient, probably less than that) to put the new numbers into the template and make whatever changes (if any) are needed to the financial statement’s structure. In addition, layers of review by internal resources and external auditors and initial setup time with the company’s financial publisher (typically Bowne, Donnelley or Merrill) will consume some additional hours. The cost of the tagging software will be trivial. In other words, we believe it would not be a burden if the SEC required all accelerated filers (companies that have US$70 million or more of public securities) to tag the basic financial statements in their periodic filings with XBRL.

At this point, we believe the time required to do the tagging is not the most important issue that needs resolution. Rather, it will be which specific tags companies use for their financial statements. Even though the income statement and balance sheet taxonomies are relatively robust, during testing we had to create nonstandard tags for all but one company (Brush), despite attempting wherever possible to use a standard tag. We also found that in all cases, a company might complicate the process by using even more nonstandard tags. The reason for choosing to use a nonstandard tag might be justified if the item was a rarity in financial statements. On the other hand, the difference might be trivial (if, say, the tag’s wording does not match the company’s accounting caption exactly) and therefore avoidable. Most of the time, though, deciding whether to use a nonstandard tag will be a judgment call, and that is a looming problem.

As we went though the tagging exercise, we tried to put ourselves in the position of a corporate vice president or director of public reporting. Those people will not look at making judgment calls about how to tag financial statement items as a career-building exercise worthy of much thought. And we expect auditors, if asked, to feel uncomfortable advising their clients. They may be inclined to choose a precise nonstandard tag to remove any chance that someone will consider the standard tag “wrong.” On the other hand, auditors would recognize that using too many nonstandard tags may create suspicion that the company is hiding something or that it will frustrate analysts trying to compare companies’ results side-by-side.

These concerns can be allayed with proper preparation. Ventana Research advocates having an XBRL “dress rehearsal” before companies are required to tags their financial statements. After the newly updated and expanded taxonomies are completed, there should be a period of discussion among public companies, auditors and securities analysts to work out the treatment of the financial statement line items. At a high level, a steering committee is needed to develop a process for identifying gaps between the U.S. taxonomy and specific items in each company’s reports. Industry-specific working groups consisting of public company representatives, securities analyst groups that focus on a specific industry and industry-focused practice leaders in audit firms should collaborate to apply the process. In the end, a set of recommendations from these working groups should be examined to see where additions to the U.S. GAAP taxonomy would be helpful, along with a set of specific guidelines for applying XBRL tags. This would complement some of the higher-level recommendations for applying XBRL that will be coming out over the next six to nine months.

Having the dress rehearsal before companies start tagging financial statements should eliminate most of the anxiety companies would experience if they had to make these decisions in a vacuum. It also will ensure the highest degree of accurate and useful standard tagging because all parties will have communicated their needs. Without the dialog, public companies are likely to be wary of the requirements and rightfully resist a mandatory requirement. Without the up-front discussion in a structured format, it will take several years to sort out the mess after mandatory tagging begins. Establishing a review process ahead of time will also be useful as XBRL tagging requirements expand to footnote items.

Assessment
By our reckoning, the cost to a public company of adopting the basic level of XBRL tagging for its financial statements is trivial, while the benefits to the public of tagging are huge. We recommend the SEC make tagging mandatory for annual and quarterly financial statements (Forms 10-K and 10-Q, respectively) beginning with periods ending after Dec. 15, 2008. The infrastructure needed for U.S. companies to be able to file their basic financial statements using XBRL tagging will be in place. Some tools for automating tagging already are available and will be mature by that point. We believe finance professionals will require minimal training to become proficient, and the tagging process does not take a significant amount of time. The financial publishers are ready to get this information into EDGAR. By the second half of 2007, the SEC will have online tools available so individuals can make use of the tagged information. After years of talking about XBRL, we find the technology has matured to the point where public companies should use it to communicate to investors.