Finance and Management newsletter, Issue 136, Sept 2006

published by the Finance and Management Faculty

On Budget day (22 March) this year, Lord Carter of Coles released his long-awaited 'Review of HMRC online services' . Judging by the press coverage it generated, one might be forgiven for drawing the conclusion that the report was all about the filing date for personal tax returns and that it was therefore primarily of interest to accountants in practice who prepare returns on behalf of clients.

Such a conclusion would be erroneous. The report is far more wide-ranging, and a whole section of its recommendations is entitled 'For business and employers'. The report sets out which returns to government agencies must in future be filed online instead of on paper, and by what dates these filings should become mandatory. The government announced that it was accepting Lord Carter's recommendations.

But the report went further than that. One interesting detail was that it said not only that companies must, from 31 March 2010, file their company tax returns online but also that they should file the returns using XBRL. So what exactly is XBRL?

The background to XBRL

Admittedly, the name is a bit off-putting. It's yet another acronym, and one whose meaning is not immediately obvious, nor even guessable. And it looks just the sort of term that must have been invented by a boring IT-nerd. So can't we just ignore it, leave all this stuff to the 'techies' and get on with our real work?

No, we can't. Even if we don't know, or don't care, what it's called or how it works, XBRL is likely to become increasingly important to business over the next few years. This article focuses primarily on the practical impact of XBRL but it also seeks first to give a simple explanation of the underlying technology.

Let's proceed with an explanation of the technology by stages, starting with technology that will already be familiar to all. Hypertext markup language (HTML) is the coding language that is used for web pages. Your web browser reads that language and knows how to display it on your screen. It is a 'tagging language', which means that a block of text can be surrounded by 'tags' that tell the browser how that text should appear. For example, elephant denotes that the word 'elephant' must appear in bold type. There are other tags that indicate where on the screen the text should appear; and then, of course, there are the tags that turn a piece of text into a hyperlink to another site.

Extensible markup language (XML) does all of that, but is a bit more sophisticated. Besides telling a web browser (by means of a stylesheet) how to display a piece of text, it describes what that text is. For example, a list of names and addresses might use tags something like this: Mr John Smith, 23 Acacia Avenue, Birmingham... and so on. A programme that can read this kind of code could be told by a user to extract just the names, or just the names and phone numbers; to display or process those elements in some way; or to export them to another programme.

For XML to be useable, there has to be an agreed vocabulary (known as a schema): the computer that is reading the XML document has to know, for example, that the first line of a physical address is tagged 'addr1' and not 'addrfirstline'. There are XML-based languages already in use in many different fields including, for example, chemical markup language (CML) and even theological (ThML). In the business field, one of the best known is ebXML, which was developed as a joint initiative of the United Nations and OASIS to facilitate global e-business.

XBRL goes another stage further

Extensible business reporting language (XBRL) is one such use of XML. But it goes yet another stage further. Besides applying identifying tags to items of data, XBRL can show how items are related to each other, and therefore how they are calculated. 

For financial reporting purposes, a number of taxonomies have been developed and several more are currently under development. These taxonomies are dictionaries and categorisation schemes, which define the tags for particular items of data, such as 'net profit', 'assets' or 'equity'. Different taxonomies are needed for different countries' generally accepted accounting principles (GAAPs), for international financial reporting standards (IFRS), and for any specific business reporting requirements of particular industries or regulators.

'Extensible' is a key word. What it means in practice is that the language can be extended as required. If a need arises to use a new business data concept, it can be accommodated in a new taxonomy or in an extension to an existing taxonomy. This gives XBRL great power and flexibility.

Business benefits

So what practical benefits does XBRL bring? It's all about efficiency in the collection, transmission, storage and processing of business data. Potentially, any organisation that prepares, transmits or uses business data will benefit. Within a company, information that has come from a number of sources (for example, different divisions, using different accounting systems) can be assembled quickly and easily if those sources are outputting XBRL. Internal management reports can be efficiently and accurately generated using software that selects the required pieces of data by looking for the appropriate tags. Similarly, reports for external use - for example those required by lenders or regulatory authorities - can be automated; and software can be used to check and validate it.

Users of business data - investors, analysts and regulators - stand to benefit greatly from XBRL. For example, when a large number of companies in a particular industry sector all publish financial reports using XBRL, investors and analysts will be able to assemble genuinely comparable data from those companies and analyse it. Lenders will be provided with speedier and more accurate tools for tracking financial performance and assessing credit risks.

Mandation by government

Government and regulatory bodies are already reaping the benefits of XBRL, and are now turning out to be the strongest drivers for its adoption. It is easy to see why online filing gets the required information into the hands of the government agency much more quickly and easily. But that is just the beginning of the process, not an end in itself, since the information is (one hopes!) being collected for a purpose - or indeed for several purposes.

Once collected, some information may have to be made available for public access; and in many cases decisions have to be based on analysis of the information. If the information is collected in XBRL in the first place, many of these processes can be automated. Much costly and error-prone re-keying can be eliminated; the processes can be speeded up; and human beings can concentrate on 'higher level' decision-making rather than routine data manipulation.

Tax authorities can exploit the enhanced analysis capability in order to identify and target errors and fraud, thus using their investigative resources more efficiently. In the long run, the taxes we pay to fund these services are used more efficiently - so we all win.

XBRL-enabled e-filing has already been in place in Companies House for several months, and some 30,000 sets of accounts have been submitted. The launch has deliberately been low-key, so that any teething troubles can be ironed out while volumes are relatively low. Implementation is proceeding by stages, starting with the simplest kinds of accounts: those of dormant companies, audit-exempt abbreviated accounts and audit-exempt full accounts.

Because the vast bulk of accounts submitted to Companies House are the simplest, the XBRL-enabled forms already in place can accommodate well over half of them; and once the facilities currently under development are completed, that will rise to 85% - all accounts except those requiring audit.

Next in line is Her Majesty's Revenue and Customs (HMRC), which requires rather more information than Companies House; and work on the required taxonomies is not yet complete. Currently a company tax return consists of three components: a CT600 form, statutory accounts, and computations. The computations are paper-based and there is no standard format; similarly, the accounts are currently all on paper. Furthermore, companies also have to submit accounts elsewhere, and it is wasteful for them to have to submit slightly different versions of the same information to different places at different times.

Longer term, it will make sense to provide a single electronic submission point for all the information that different government departments and agencies require. XBRL will enable each to take those pieces of information that it needs. The justification for Lord Carter's recommendation that the use of XBRL for company tax returns should be mandatory within four years is therefore clear. This will be a powerful driver for its adoption by UK businesses.

XBRL has been seven years under development, and the worldwide accountancy profession has been playing a leading role throughout that time. In the UK, the ICAEW has been facilitator of the XBRL UK consortium since 2000. Taxonomy development - mapping accounting standards and other business reporting concepts onto a complex hierarchical structure - is a long process that requires significant technical input and considerable consultation with many interested parties. In the UK it is now coming to fruition in practical applications. Undoubtedly more will come on stream in the next couple of years.

Conclusion

One significant implication of the increasing use of XBRL is that businesses contemplating acquiring or replacing software applications would be well advised to consider whether to seek systems that have XBRL functionality - or, at least, to enquire what plans the software suppliers have to incorporate XBRL functionality into their products.

Author: Paul Booth is technical manager of the ICAEW's IT Faculty