XBRL (Extensible Business Reporting Language) is an open standard which supports information modeling and the expresion of semantic meaning frequently required in business reporting. XBRL is XML-based. It uses the XML syntax and related XML technologies such as XML Schema, XLink, XPath, Namespaces, etc. to articulate this semantic meaning. One use of XBRL is to define and exchange financial information, such as a financial statement.

XBRL is a standards-based way to communicate business and financial information. These communications are defined by metadata set out in taxonomies. Taxonomies capture the definition of individual reporting concepts as well as the relationships between concepts and other semantic meaning.

The XBRL format is governed and marketed by a international consortium (XBRL International Incorporated) of approximately 600 organizations, including, companies, regulators, government agencies, infomediaries and software vendors.

XBRL International is supported by its jurisdictions - independent bodies, typically organised on a country-specific basis - that work to promote the adoption of XBRL and the development of taxonomies that define the information requirements of their specific domains. XBRL is being adopted around the world in order to migrate business information process from paper-based and legacy electronic proprietary formats more fully onto Internet oriented processes (both for external and internal reporting processes).

The U.S. Federal Deposit Insurance Corporation (FDIC), in coordination with the Federal Reserve Board and the Office of the Comptroller of the Currency, launched an XBRL project in October 2005 involving the collection of quarterly bank financial statements (Call Reports) from over 8300 U.S. banks. Use of XBRL is mandatory and the data is posted on the Internet for public use and analysis. A case study prepared by the banking regulators can be found here.

The first European project for the Dutch Waterboards, and extra XBRL projects in Europe including Spain, Belgium and others, have given Europe a leading role in implementations. Currently a major project of the Dutch Government for XBRL reporting by all businesses as well as (semi-)government organizations like cities and health institutes will make it the first nation-wide implementation. The 'Dutch Taxonomy Project' is outlined here.

The wiki repository of more XBRL projects can be found at Worldwide XBRL Projects, to be freely explored and updated.

Document structure

XBRL consists of an instance document, containing primarily the business facts being reported, and a collection of taxonomies, which define metadata about these facts, such as what the facts mean and how they relate to one another. XBRL uses XML schema, XLink, and XPointer standards.

Instance document

The instance document holds the root. The document itself holds the following information:

Business Facts - facts can be divided into two categories

Items are facts holding a single value. They are represented by a single XML element with the value as its content.

Tuples are facts holding multiple values. They are represented by a single XML element containing nested Items.

Contexts define the entity (i.e. company or individual) to which the fact applies, the period of time the fact is relevant, and an optional scenario. Scenarios provide further contextual information about the facts, such as whether the business values reported are actual, projected, budgeted, etc.

Units define the units used by numeric or fractional facts within the document, such as USD, shares. XBRL allows more complex units to be defined if necessary.

Footnotes

References to taxonomies, typically though schema references.

This is an example of a fictive Dutch company's IFRS statement instance file : 38679000000 35996000000 870000000 ... 10430000000 ACME 2004-01-01 ACME 2004-12-31 ACME 2004-01-01 2004-12-31 iso4217:EUR

Taxonomies

Taxonomies are a collection of XML schema documents and XML documents called linkbases by virtue of their use of XLink. The schema must ultimately extend the XBRL instance schema document and typically extend other published XBRL schemas on the xbrl.org website.

Schemas define Item and Tuple "concepts" using elements. Concepts provide names for the fact and designate whether or not it's a tuple or item, what type of data it contains (monetary, numeric, fractional, textual, etc.) among some other metadata. Items and Tuples can be regarded as "implementations" of concepts, or specific instances of a concept. A good analogy for those familiar with object oriented programming would be that Concepts are the classes and Items and Tuples are Object instances of those classes. In addition to defining concepts, Schemas reference linkbase documents.

Linkbases are a collection of Links, which themselves are a collection of locators, arcs, and potentially resources. Locators are elements that essentially reference a concept and provide an arbitrary label for it. In turn, arcs are elements signifying that a concept links to another concept by referencing the labels defined by the locators. Some arcs link concepts to other concepts. Other arcs link concepts to resources, the most common of which are human-readable labels for the concepts.

The public search engine www.abra-search.com provides many public XBRL taxonomies for browsing and searching without the need to install dedicated XBRL software.

This is the taxonomy of the above shown instance file:

Critics

"XBRL is not an accountancy product . It is a consulting product. It is a product in search of a market, and they found it in government mandates. The consulting firms that promoted it are drooling at the fee potential, which is huge.

....R. Corey Booth, the SEC's director of information technology, painted a sombre picture, admitting that the SEC received just 22 XBRL filings from nine companies in 2005, most of which were involved in promoting XBRL, that there was no investor demand for XBRL information, and the difficulty of rendering XBRL documents in human readable format."

"If it was really that great, it wouldn't have to be mandated...It's being pushed by the people who have an interest in pushing it."

European adoption

Belgium, Cyprus, Denmark, France, Finland, Germany, Italy, Lithuania, Luxemburg, Netherlands, Norway, Poland and Spain decided optional or compulsory adoption of XBRL.

In November 2006 the FSA decided not to adopt this would be standard in the United Kingdom

Other European countries followed: Austria, Czech Republic, Denmark, Estonia, Latvia, Portugal, Slovak Republic and Sweden have also definitely decided not to adopt this format while Hungary, Estonia, Greece, Lithuania, Bulgaria and Italy remain undecided.