I was discussing the United States Securities and Exchange Commission’s (SEC) eXtensible Business Reporting Language (XBRL) mandate with a former head of investor relations at a Fortune 100 company. His take on it is much the same as that of everyone else involved with corporate reporting: it doesn’t produce much value and costs a bundle to comply. I related to him my thoughts on the lack of progress I saw in making the XBRL mandate more useful to corporations and investors alike. Making XBRL data readily available to the public – not just for SEC enforcement purposes – is consistent with the SEC’s three-fold mission to protect investors, maintain fair, orderly, and efficient markets, and facilitate capital formation. In addition to giving XBRL-tagged data greater practical value to investors, the trove of company data assembled by the SEC could be used by a wide range of people working within corporations.

Recently, in the context of our long-range planning benchmark research, a CFO asked me if I had capital spending metrics for her specific industry. To be sure, there are paid-for services that provide these sorts of metrics but (like me) CFOs seem to be preternaturally cheap. For them, another approach would be to collect relevant data from U.S. public companies in that industry from the SEC’s EDGAR database. Before XBRL, assembling the data was time-consuming, since it involves sorting through the list of companies in a specific SIC code grouping, selecting the existing relevant companies while ignoring the defunct ones, and retrieving their most recent 10-K (annual) filings. After the introduction of XBRL the process is… still time-consuming. Indeed, I checked and there’s exactly no difference in the time it takes to retrieve and analyze the data today and the time it took in the 1990s. The main difference is that back then, the filings were in text format; today they are rendered in HTML. The steps required to get the data, arrange it in a spreadsheet and analyze the results are almost exactly the same.

Contrast the difficulty of getting data from EDGAR with the relative ease of retrieving numbers from the Bureau of Economic Analysis (BEA) and the Bureau of Labor Statistics (BLS). These two government data sources make it relatively easy to download specific data sets into spreadsheets or flat file structures for further analysis. While the original idea was for the SEC to provide people with an interactive analysis tool that would enable them to make use of XBRL-tagged data, I’d settle for EDGAR providing a useful interactive data retrieval tool. In particular, I’d like one that would enable me to (for example) select an SIC code, select elements from the XBRL taxonomy, select time periods, and then download the result in some readily usable format such as a spreadsheet or as flat file data for further interactive analysis.

One reason EDGAR data remains a static, hard-to-use pile of numbers is that too many people involved in financial software companies and the SEC continue to act as if the data in EDGAR is readily consumable. It is not. To address the lack of useful tools, XBRL US began the XBRL Challenge. It’s a good first step, but much more must be done. Since the XBRL mandate was put into action, U.S. filers have collectively spent hundreds of millions of dollars complying with the requirement. Since fiscal 2008, the SEC’s budget has increased by almost $400 million, or 40 percent. A rounding error in this sum would have been sufficient to fund the development of a method of extracting XBRL tagged data and an interactive XBRL-enabled analysis tool.


Robert Kugel – SVP Research