Mar 14

eXtensible Business Reporting Language. Part 2: What is XBRL?

by David Harper, CFA, FRM, CIPM

CFA |

We are highlighting lessons from a screencast called An Introduction to XBRL. The CFA Institute prequalified this tutorial for 0.5 credit hours under their Professional Development (PD) program.

Financial data skips human, goes directly to spreadsheet

Previously, we noted three potential benefits of XBRL: quicker "time to analysis," fewer transcription errors, and greater intelligence enabled by structured data. By tradition, an analyst copied data from a filing into his spreadsheet:

But the the most immediate benefit of XBRL is that transcription can be eliminated. Corporate filing data is served directly to a spreadsheet like Excel (or any application that can process XML): 

XBRL is Structured Data

For an analyst, most of the important information is unstructured; e.g., management's discussion & analysis (MD&A), footnotes, interview notes. The idea of XBRL is to give structure to data. Once data is structured, it becomes a database and can be processed, calculated, and mashed-up. XBRL converts unstructured financial reports into richly structured documents:

XBRL is XML

This is achieved with eXtensible Markup Language (XML). As we say in the tutorial, "XBRL is XML." Sometimes we also say, XBRL is a "flavor of XML" or an emerging XML-based standard. That's because there are many "flavors" of XML, or put another way, many different industry- or application-specific XML-based standards. XML itself is generic and universal. It is a markup language that consists of tags; the tags wrap (or contain) values, and in doing so, they give structure to the data:

As above, the value for pi (3.1415927 rounded) is contained in a start <Element> tag on the one side and an end </Element> tag on the other side. Further, an attribute of type is assigned to this particular tag (the others above have no attributes). The tags above are quite generic; XML is generic.

If an industry wants to define a set of tags that are specific to their purpose, they will band together and create a shared dictionary (a taxonomy). It's important to agree on a shared taxonomy, because in order to share data, participants need to use the same language (at least partially). So, there are currently hundreds if not thousands of industry-specific consortia who are creating their own application- or industry-specific sets of tags (i.e., taxonomies as part of an XML-based language).

In the case of financial reporting, XBRL International is "a not-for-profit consortium of approximately 450 companies and agencies worldwide working together to build the XBRL language and promote and support its adoption." Largely because of their work, instead of generic tags, we know that <OperatingRevenue> is the tag that contains Operating Revenue, for example. Every item on the five primary facing financial statements (e.g., Income Statement) has a corresponding XBRL tag:

Comments

  1. Be the first to leave a comment!

Leave a Comment