Preventing XBRL mistakes is not just a compliance exercise. It is in fact a pragmatic necessity for telling your company's financial story to analysts, shareholders, potential investors, and others in the market. All publicly traded US companies now file their SEC financial disclosures in XBRL. With XBRL implementation complete, the focus has shifted to the quality of the XBRL tagging in corporate financial reporting. An XBRL instance document must exactly translate the corresponding EDGAR filing. Most filers face full liability for mistakes in their XBRL disclosures and therefore can no longer risk filing documents with tagging errors that skew their financial reporting.
Of course, the most immediate and obvious motivation for accuracy in XBRL filings is the need to be SEC-compliant. The SEC expressed concern about inaccuracies in XBRL tags in, for instance, a comment letter to Standard Drilling. Item 15 on Page 5 states:
We note that you have checked the Smaller Reporting Company box on the cover of your Form 10-Q. The XBRL Document and Entity Identification Information rendered as part of your filing appears to contain a number of data element errors, including but not limited to, your classification as a non-accelerated filer. Please revise to comply with the requirements of Section 405 of Regulation S-T and the EDGAR Filer Manual.
However, producing high-quality XBRL data is not just a matter of compliance. For every company that files with the SEC, ensuring accurate financial disclosures in XBRL is a matter of pragmatic, market-driven importance - as vital as the accuracy of traditional EDGAR filings. Precise use of XBRL is essential for your company's communications with the financial markets.
Ouch! Real-life XBRL mistakes that hurt
XBRL represents a great leap forward in corporate communications to the market, analysts, and the media. However, financial reporting in XBRL is only as good as the XBRL tagging that translates the EDGAR document into the XBRL instance document. Mistakes can be painful for filers. In addition to the grave SEC consequences they can provoke, XBRL errors distort - sometimes grossly - the financial story that a company is trying to tell shareholders, potential investors, analysts, and the media.
Here are examples of XBRL mistakes that Merrill experts have detected in actual corporate financials. The errors shown here are not outliers. These types of faults appear right now in the XBRL instance documents of many publicly traded companies, both large and small.
From the Form 10-K of a large filer, here is what the filer intended to convey, and what the erroneous XBRL tagging actually meant:
This example shows the potential severity of incorrect tag selection. There is a tag in the US GAAP Taxonomy for the $17.2 billion figure, but the company did not use it. Instead, it created an extension that told the story erroneously.
In addition, the same filer made a similar mistake elsewhere in the same 10-K:
Again, the selection of the wrong XBRL tag profoundly misreported the financial story.
From another company's 10-Q filing:
Although this statement is not necessarily incorrect, it is incomplete. What's missing? While the right tag was chosen for the 25 million shares, the filer forgot to add the Subsequent Event member. Moreover, the filer omitted another important piece of information: the fact that the stock was Class A common stock. A member for that exists in the US GAAP Taxonomy, and it should have been selected to tell the whole story.
Yet another mistake in a 10-Q filing:
How did this happen? Quite simply, the $39 million was erroneously entered into the XBRL as a positive instead of negative value.
“95% accuracy is not good enough - even 98% is not good enough”
For the October 2013 issue of Dimensions, we interviewed Professor Dhananjay (Dan) Gode, a Clinical Associate Professor of Accounting at New York University's Stern School of Business. He teaches corporate financial accounting and pursues research interests in financial analysis. In frank terms, Prof. Gode outlined the importance of XBRL accuracy for a company's financial communications with the market: “[E]ven if the data is 95% accurate, maybe you can use it in a classroom; but if you are doing any professional work, 95% accuracy is not good enough - even 98% is not good enough.” Prof. Gode told us about some of his bewildering experiences with inaccurate XBRL filings. “Now, I do not analyze the whole [XBRL] database; I'm just a user. If I see [a mistake], I get annoyed. I do not know how many more mistakes there are in the whole database. Even if you try to do four things and you find mistakes in each of those four queries, it leaves you quite a bit unsettled. That's [a] problem.”
XBRL quality is about telling your company's financial story to the market
The SEC's newly developed Accounting Quality Model demands high-quality XBRL. This quantitative analysis tool uses XBRL data to scrutinize financial disclosures and to select filers for closer review and potential enforcement action. (See Inside The AQM, in the December 2013 issue of Dimensions.)
Yet the SEC is not the only set of eyes scrutinizing XBRL instance documents. Financial data from XBRL filings is being gathered and made available to investors, analysts, and sundry participants in the market. In other words, people want to use XBRL data to do research on corporate performance and make investment decisions.
The market's use of XBRL was summarized lucidly by Campbell Pryde, the president and CEO of XBRL US, in a recent Merrill webinar on XBRL quality. Citing information gathered by his company, Mr. Pryde listed the many types of commercial distributors - from data aggregators to media companies - that are collecting and parsing the XBRL data in corporate financial disclosures. The SEC is also working on making XBRL-tagged financial data available to the public in various ways, such as .csv and Microsoft Excel formats. According to Mr. Pryde, in 2013 the number of governmental and commercial entities distributing XBRL financial data more than doubled over the number in 2012. Some are established data aggregators, such as Thomson Reuters, while others are new vendors. XBRL US is working on integrating XBRL data into Google search results, Mr. Pryde indicated during the webinar, to “get a much broader audience of people looking at financial-reporting information and being interested in what companies are doing and in how they are performing.”
Analytical tools require error-free data
Anticipation of growing market demand for XBRL-tagged financial data has led to the development of many innovative analytical XBRL tools, but these applications all need errorfree data to produce viable results. The Financial Executives Research Foundation (FERF) recently published a report about analytical tools that pull data from XBRL instance documents (Data Mining With XBRL And Other Sources, by Leena Roselli). “Even though XBRL's benefits may not seem evident at the present time,” concludes the author, “data mining tools are making the analytical process easier. Whether for internal use or external use, tagging is powerful.” Conversations with personnel at the tool-makers profiled in the FERF report reveal the importance of error-free XBRL filings for their products. For instance, 9W Search makes an application that uses a normalized dataset to provide corporate financials and ratios. Now available at popular investment websites such as TheStreet.com, the 9W Search platform also pulls the information in financial statement footnotes directly from XBRL instance documents. “The footnote data is untouched in our system, and the filing company is responsible for its accuracy,” emphasizes Susan Strausberg, the firm's CEO.
Another recently created product, developed by Calcbench, enables comparative analyses across corporate sectors or peer groups - for example, how Intel's balance sheet compares to those of its peers, or how annual revenues fluctuate in the semiconductor industry. “When an individual firm makes a mistake in its data, then the entire [data] universe is affected,” Alex Rapp, the cofounder, president, and CTO of Calcbench, tells us. “We are starting to see our information used more and more by investors, auditors, academics, and lawyers. Filers that cannot get it right are going to develop negative reputations in all of those circles.”
FinDynamics, a developer of applications for financial data, has an add-in tool for use in Microsoft Excel that pulls XBRL data from 10-Ks, 10-Qs, and other SEC filings. According to Ilya Vadeiko, the firm's executive director, the tool imports XBRL data to Excel in undistorted form. “Therefore the accuracy of our data is identical to that of the original source,” he asserts. “Obviously, the data quality of the original source is an irritating issue for us and for our users.” Mr. Vadeiko quickly points out another key reason to get XBRL tagging right: Financial data that is inaccurate because of XBRL errors makes it hard for analysts to produce accurate reports about the company - potentially leading analysts to avoid covering the company altogether.
The damage that bad XBRL data can inflict on a company's relationship with analysts is a recurring theme in our conversations with expert commentators. “One of the benefits of XBRL is for smaller companies to be able to make accessible their financials in a timely fashion, as smaller companies are last on the list of most large data aggregators,” notes Ganesh Rajappan, the founder and CEO of LogixData, an application developer. “Unfortunately,” he informed us, “having errors in XBRL makes this more of a dream than reality.” According to Fuad Rahman, CEO of software developer Apurba Technologies, often the most problematic XBRL data comes from small companies that outsource their XBRL-document preparation to unreliable service-providers.
It takes more than software: human experience and expertise are vital
It takes more than software to craft the kind of XBRL excellence that Congress, the SEC, and the market are demanding. Review by a human expert is crucial for any company seeking assurances about the accuracy of its XBRL tagging. XBRL submissions in which Merrill is involved are all prepared or reviewed by consultants who are experienced CPAs particularly skilled in XBRL usage. These consultants know the intricacies of XBRL and the SEC rules thoroughly, and they apply that understanding to XBRL documents. Merrill's unique expertise allows corporate executives to move forward with confidence, knowing that their XBRL filings do not contain errors.
“Errors that can't be checked by software. really require the involvement of XBRL-knowledgeable individuals,” emphasizes Lou Rohman, Merrill's VP of XBRL Services. “That's a key point of getting the XBRL right. By knowledgeable, I mean individuals who understand the SEC rules, the accounting disclosures, the intricacies of the XBRL itself, and then are able to apply that experience to the XBRL document."