XBRL: The Language of Finance and Accounting

March 10, 2004

Dale Waldt

What Is It?

The eXtensible Business Reporting Language (XBRL) is a language for capturing financial information throughout a business' information processes that will eventually be reported to shareholders, banks, regulators, and other parties. The goal of XBRL is to make the analysis and exchange of corporate information more reliable and easier to facilitate.

In order to deliver or report information in a consistent form, the creators of XBRL expect the resulting vocabulary to affect the format of the business information throughout the entire life cycle of that information, from the initial creation of invoices, orders, and other documents and actions, through the collection, aggregation, and reconciliation processing done in the financial departments, and eventually, to the reporting formats such as regulatory filings, statements, and corporate reports.

• What Is It? • Where Does It Come From? • What Are They Working On? • What Else Is There? • How Does It Work? • Who's Using It? • Issues & Challenges • Resources

Organizations store much of the same information in their own internal systems, but in widely varying formats and granularity. This is common, even between businesses that share the same financial software products. This inconsistency makes it difficult for organizations to share information reliably or cost effectively when performing many aspects of their business, including applying for credit, reconciling accounts, or reporting to investors, or many other common activities.

To further complicate the problem, the way a single field or element of information is defined can be affected by jurisdictional regulations. The value of an asset may be defined quite differently in the U.S. than in other parts of the world that observe different accounting principles and practices. Hence, there may be many different actual definitions of a single element type such as "asset value", or "price", or other seemingly simple bits of data. These distinctions affect a wide range of organizations since there aren't many large companies that operate in a single regulatory jurisdiction anymore.

XBRL creates a vocabulary for describing exactly which bits of information are being included in a report, even to the point of taking regulatory jurisdiction and other variances into consideration.

Where Does It Come From?

XBRL originally began as work in the American Institute of Certified Public Accountants (AICPA), the U.S.-based national professional organization serving the accounting profession for more than 100 years. One aspect of the AICPA mission is to establish professional standards, in how accountants behave, perform, and communicate information. In other words, they define the principles and terminology used in that jurisdiction and all developed countries have an equivalent organization setting standards for their constituency.

Since a goal of XBRL is to facilitate data interchange globally, and AICPA is focused on the United States, the XBRL folks quickly resolved to form an international organization, XBRL International (http://XBRL.org), which would coordinate similar activities in each jurisdiction, where a local XBRL "chapter" would represent its specific needs. The structure of XBRL now recognizes both regional jurisdictions as well as various industry domains involved in business reporting.

There are currently 13 active jurisdictions listed on the XBRL web page as well as instructions for establishing new ones. Ten jurisdictions are active within XBRL. These are Australia, Canada, Denmark, Japan, Netherlands, New Zealand, Singapore, U.K., U.S., and the International Accounting Standards Board (IASB), with four additional provisional members in Spain, Hong Kong, Ireland, and Korea.

What Are They Working On?

In the world of vertical industry groups, Accounting & Tax Preparation is one of the biggest categories, and should be thought of as a mega-category of sorts. Every business has to account for and report information, whether private or public, profit or nonprofit, and regardless of what types of products or customers or location you work in. That is why several nearly distinct "industries" fit together in this larger grouping. There are currently six officially recognized, related industry domains called Supply Chain Communities (SCCs) that are actively involved in business reporting as well as several others not yet formally recognized by XBRL. Those officially recognized include Accountants, Analysts, Intermediaries, Software & Service Providers, Regulators, and Investors/Creditors. Each of these types of organizations perform a role that are either suppliers, recipients, or both, of business reporting information.

As you may have realized from this background information, the job of creating the taxonomies and processes to be used for reporting business information globally is a large undertaking. Currently, the work is divided into eight general Working Groups (WG) with domain-specific working groups being created as needed. The eight general WGs are Specification, Domain Basel II (Taxonomy), Marketing & Communication, Education, Research, Grants, and General Ledger. Over 170 organizations have joined XBRL from around the world and interest is high.

What Else Is There?

XBRL is envisioned as a standard that can be used in just about any business, and the hope is that many vertical industries will avoid reinventing the wheel when they try to define common components not unique to their particular area. But some areas already have invested a lot of time and effort in creating vocabularies and applications that use them that fit easily into the scope of XBRL. One of the best examples is the Financial Services Industry (actually another one of those mega-industries that is made up of several sub-industries such as Banking, Bonds, Securities, Risk Management, Software & Services, Regulation, and so on). There are many standards and specifications that have come out of Financial Services that overlap or relate to XBRL. These generally focus on executing specific financial transactions using a single vocabulary (for example, RIXML, FIXML, MDDL, IFX, and OFX). Some view XBRL as the logical opportunity to serve as a unifying common language to either enhance these specific standards or serve as an interchange format between them.

Similarly, other standards from organizations such as OASIS and other bodies have overlaps that could benefit from a common language for business information. For instance, there is some interest in rationalizing, if not actually coordinating, work being done in the OASIS UBL (Universal Business Language) TCs and ebXML (Electronic Business eXtensible Markup Language) TCs. UBL and XBRL have established formal liaison relationships to coordinate and communicate activity. In fact, XBRL has many official liaison relationships and is interested in working with other groups in the hope of leveraging and building upon standards from such groups as, for example, the UN GREFIS (part of the UN electronic commerce standards organization that focuses on insurance risk standards and processes).

How Does It Work?

XBRL represents a broad set of taxonomies. These taxonomies consist of a W3C XML Schema modeling descriptions and a classifications system for information found in financial statements and other business reporting documents, plus supporting documentation and XLink links to other supporting files. Taxonomies may represent hundreds of business reporting concepts with mathematical and definitional relationships, along with text labels in multiple languages, references, and display suggestions.

Taxonomies may receive recognition from XBRL in three different levels. Recommended taxonomies are considered to have the same official status of the XBRL specification itself. Approved taxonomies are compliant with XBRL and were either developed under XBRL processes or similar ones. Acknowledged taxonomies are compliant but were developed by other organizations under their processes. All recognized taxonomies are available for use royalty free. The current state of an XBRL taxonomy of Public Working Draft means further work may continue, but Final means it is complete and will be permanently available at the XBRL web site.

To follow is a sample of a raw XBRL data instance. It precisely describes key financial information, including that Rock Gravel Corporation is reporting a numerical content for the calendar year 2000 in Hong Kong dollars (HKD) with a precision of up to four digits.

<numericContext id="rg.cy00.hkd" cwa="false" precision="4">

 <entity>

  <identifier scheme='http://www.gov.hk'>rg</identifier>

 </entity>

 <period>

  <startDate>2000-01-01</startDate>

  <endDate>2000-12-31</endDate>

 </period>

 <unit>

  <measure>iso4217:hkd</measure>

 </unit>

</numericContext>

The following example shows an XBRL instance indicating that, related to the previous numerical content via the id= attribute, and according to the rules of the Generally Accepted Accounting Principles (GAAP), Rock Gravel Corporation had operating costs of HKD $358,300,000 in that year:

<?xml version="1.0" encoding="UTF-8"?>

<gaap:opc numericContext="rg.cy01.hkd">-3583000000.</gaap:opc>

These snippets of XBRL instances only give the slightest indication of the power and complexity of financial reporting data and XBRL itself. To date XBRL taxonomies have an aggregate total of over 3,500 defined elements. The complexity of these interrelations are managed through a sophisticated set of XLinks.

Financial statements can be created in XBRL in several ways. Several XBRL-aware accounting software products support XBRL as a native import and export format. These tools allow users to map charts of accounts and other structures to XBRL tags. XBRL can be created manually or generated from a reporting system. XBRL can be generated from EDGAR-formatted information using transforms found on several web sites, allowing users to leverage their EDGAR generation applications. Or, similarly, other formats can be transformed as a back-end process. It is likely that a combination of these will be used by companies wishing to exchange information in XBRL.

Version 2.1 of the conformance suite, made up of hundreds of example taxonomy fragments (W3C XML Schema and XLink files) and instance documents, containing both valid and invalid usage, was completed December 31, 2003.

Who's Using It?

XBRL is enjoying many examples of adoption, and is currently in live use for some critical financial industry applications. Some highlights from an extensive list of development based on XBRL include:

• U.K. Inland Revenue is using XBRL in electronic filing based on XBRL-UK GAAP Taxonomy
• Australian lending (LIXI) information being collected from all Australian banks in XBRL
• Sumitomo Mitsui Bank of Japan integrated XBRL as interchange format for letters of credit for international trade as part of complex supply chain application
• The Tokyo Stock Exchange has been accepting corporate financial information in XBRL format since early 2003

Vendors have also shown a lot of support for XBRL. For example, 100 percent of all North American tax preparation software vendors have committed to supporting XBRL as an interchange format for importing and exporting data from their systems. More examples can be found at the XBRL web site in its XBRL Progress Reports (from the main menu at http://www.xbrl.org select What is XBRL > XBRL Progress Reports).

Issues & Challenges

The future for XBRL is not without its challenges. Some issues that face XBRL implementers are the daunting changes and development that will be needed when the EU consolidates versions of accounting standards currently under GAAP standards in 2005 under the International Financial Reporting Standards (also known as International Accounting Standards,) which will affect all 7,000 listed corporate entities in the EU. Estimates for systems-enhancement costs for global banks alone range from £20 to £100 million pounds, more than half of it in IT costs. The goal of this change is to provide consistency and transparency in financial reporting information. Sound familiar?

Even though XBRL seems well suited for the task, this is still a daunting amount of change to manage, and some will have to be convinced that a new format will make it easier. Some critics still claim that XBRL is too complex, so not everyone has been won over yet. Still others respond that it is only as complex as the underlying data models in use by the governing bodies.

According to Walter Hamscher, executive director of XBRL International, "Like any standard that has to do a point release, we have to manage expectations around stability--the fact is that XBRL 2.1 is a big step forward not so much in terms of incompatibilities and such, but it is so much more solid, well tested and worked through--we are working with vendors to get them onto 2.1 with all possible speed. I think there is still a deficiency of off-the-shelf vendor software that is properly enabled relative to the scope and sweep of XBRL's applicability. I think we need more open source. Many end user organizations will see that as an obstacle, although regulators, interestingly, don't seem concerned by that because their planning timelines stretch well beyond this or next quarter and they already see XBRL as inevitable."