The use of XBRL in the European Financial Markets

Eddy Wymeersch

Chairman of the Committee of European Securities Regulators

The use of XML and its related languages especially XBRL in the financial markets is likely to raise much attention the next few years. Different projects are being studied: linkage of the business registers, remote access to the annual accounts, organising the distribution of the financial information by listed companies, use for reporting by banks and insurance companies, as some applications that are already in place or being developed. The impact of an efficient information distribution system could help protect creditors, make cross border relations more secure and support the competitiveness of our financial markets.

In today’s financial markets, financial information should be widely available and easily accessible. Regulations have been enacted in many states to ensure that the information to be published is defined, supervised and published. Several European directives are governing the subject: company law information in the directives on annual accounts, on consolidated annual accounts and on auditing should be mentioned here along with the transparency directive, the market abuse directive and the prospectus directives. The States have all dutifully implemented these European directives and the result is an overall impressive apparatus of financial disclosure that has significantly contributed to the efficiency of our financial markets.

Although efforts are still needed to make this information more reliable – see the debate about the use of IFRS in the present financial turmoil – one can state that the present regulatory apparatus is quite satisfactory.

What is lacking are instruments to make this information, preferably repackaged and cut to measure, available to the user. These instruments  have largely been left to the private players, the data vendors and other information providers. A highly competitive business has sprung up, accessible mainly to the big players and large investors. The cost of this information is consequently quite high.

It is here that the present debate about the use of XBRL comes into the picture. The use of this technology is the best - if presently the only key - to make this increasingly voluminous data available to all market participants. Therefore it is useful to present you a few elements of the development in Europe, especially in the financial markets.

What is being outlined here may look a brave new world to some, but we are not that far away from it.

All European states have a more or less automated business register, where numerous data about all persons or firms that are economically active are stored. Searching in these databases allows everyone to determine whether the proposed contract party is effectively in business, what his precise identity is, whether he is located at the address indicated, and so on. Between the most advanced of these business registers, linkages already exist allowing to query and retrieve data on firms located in other jurisdictions.[i] However, other business registers have not yet automated, or have not established links that allow cross- system searches.

Business registers at present only contain some basic information, such as name, address, type of business, activities, names of directors etc. However, as the First Company Law Directive only lays down minimum requirements to Member States regarding the data to be registered and the electronic accessibility, individual business registers still show large disparities regarding both their contents and their data structure.  In this regard, it is important to point to the BRITE project[ii], a EU funded research project the aim of which is to improve the access to and interoperability between business registers. A strongly integrated European network of national business registers could allow to identify the different directorships or other mandates exercised by an individual, or even the shares he holds in different public or private companies across Europe. This tool would be very powerful in policing business, by tracing disqualified directors, identifying corporate control positions, combating money laundering and other criminal practices, or finding crooks attempting to defraud investors. Concerns relating to privacy considerations have to be overcome. To make a system like this work, not only will it be necessary to move to a form of unique identifier for companies at EU scale: this forms part of the BRITE project as well[iii], and it is expected that the European Commission will endorse this standard. Making business registers interoperable also requires the use  of  XML, or any other usable language  that enables business registers to communicate amongst each other.[iv] The subject has already been put on the agenda –and one hopes that gradually the Commission will take an initiative to support the effective implementation of these ideas. Directly linked to the previous subject is the distribution of financial information. Different levels have to be distinguished here: first the publication of the annual accounts by companies, listed or not, secondly making available financial and other disclosures by companies whose securities are traded on regulated markets. 

According to the 4^th company law directive, companies limited have to draw up annual accounts and make these publicly available. Depending on the national laws, the rule applies to all companies limited or only to the largest ones. The Commission has launched a public consultation whether the smallest companies should not be freed from this double obligation, but it has not yet rendered its opinion public.

At least in some member states, all annual accounts by all companies limited – both NV and BV, SA and Sarl, or AG and GmbH, etc – have to be filed with a central repository, which is not necessarily the business register, and made publicly accessible. In some states access consists of obtaining physical copies, in other it is remote, usually under PDF format. These national databases are not yet linked. One easily sees the tremendous opportunities that this basic system could offer. On the one hand the national databases could be linked, on the other by introducing XBRL tagging, searches can be made across systems and countries and help to make the cross border trade more reliable, more safe. The raw material for such a project is present, only its implementation requires a lot of effort, political will – and subsidies!

There are already some interesting developments in place. They are different from the US Edgar, as relating to all companies, public or private. Linking to the business registers has not yet been undertaken. 

Excuse me for citing the Belgian case as an example, but this is the one I know the best.

Since many years – somewhere in the 1980 – all annual accounts by all Belgian companies have to be deposited with the central bank, which makes them publicly accessible. The central bank is using the data as part of its statistical tasks for consolidating them into the “national accounts”. Since 2007 these accounts – 300.000 of them – are filed with the Bank in XBRL format, what already 75% of the companies have done. Consolidation of these data takes place with XBRL technology. The bank envisages making the technology available to the public what would allow anybody to make its own searches and calculations. Some of the taxonomies that could be introduced would allow business partners to determine on the basis of the well-known calculation models to determine the failure risk of a counterparty with whom they are opening negotiations or maintaining credit lines. This perspective is particularly significant: each firm could establish credit lines for dealing with any other firm, assuming the role of one’s own “private credit rating” organisation. The often described role of the annual account as instruments to protect business partners would become effective.

In the field of financial information relating to listed companies, several developments have to be mentioned.

On the basis art 21 (2) of the Transparency directive stating that “each home state shall ensure that there is at least one officially appointed mechanism for the central storage of regulated information “ the Member states are expected to organise the centralised storage and distribution of the financial information, published according to the directive by listed companies. CESR, the Committee of European Securities Regulators has been invited by the Commission to develop ideas relating to this subject. The actual situation in the different member states is quite different, some having installed a registration and disclosure mechanism that is centralised at the level of the securities supervisor, some even using XBRL for communication and retrieval – e.g. in Spain – while others are waiting until final decisions are made at the Community level. After long deliberations CESR has advised the Commission that the most effective way would be to organise a decentralised system, complemented by a hub and spoke approach, whereby CESR itself would organise the hub, a list of all listed companies, that would allow users to click to the national mechanisms, the spokes. In fact this list exists already as in implementation of the Mifid CESR publishes on its website the list of all shares – about 7200 - that are traded on one of the regulated markets in Europe. This list contains details and data about the status of these shares, e.g. indicating the most liquid market in these shares. It is projected that this list will be supplemented by a click to the national disclosure systems. How the national systems are going to be organised will depend on the national regulator: some have already established a national storage mechanism where all company related mandated information is being stored ( ex. Portugal, Spain using XBRL) , while other states are reflecting about whether to centralise this information on a server maintained by the local securities supervisor, or to store the information on a protected page of the issuer’s website. The choice between the approaches would be left to the member states, the use of clicks allowing in practice to obtain the same end result, except for one additional click. The last mentioned approach-  i.e. storing the information on the issuer’s website - would be the least costly and the most close to the markets, as investors are more likely to search for information on the issuer’s website, than on that of the national regulator. It also solves some delicate liability questions. The role of the securities supervisor would then be to supervise whether the information is available on the issuer’s site, and conform to its instructions as to content, format and stability. Companies would be urged to use XBRL, allowing for focusing access to core data. With today’s advanced search technology, there can be no problem for making searches on the websites of several issuers and have the selected data consolidated by using XBRL. One will be able to extend the search to information in the annual reports, e.g. for comparing directors’ remuneration in a number of selected companies.  This implies that the issuers all have to use the same language, meaning that not only XBRL tags have to be appended to the data in the data base, but that the content of the tagged information has to be comparable, e.g. that “basic salary” corresponds to the same definition. The use of XBRL would also allow to deal with the vexing language issue: many European companies still publish only in their national language. Even if it might be difficult to have a somewhat rough automated translation for the entire annual report, automated translation could at least be introduced for the more standardised data, such as the annual accounts, or specific numerical data in the annual report.

As matters stand at present the member states are waiting for a clear sign from the EU Commission, to see how the Commission reads the directive, and what measures have to be taken at the national level. One can expect this opinion to contain a position on the use of XBRL for financial disclosures. But before this part of the directive can be fully rolled out, agreements will have to be reached on the content of the information to be treated in XBRL, and what the tags will exactly cover. CESR could be material in advising the IASB, on formulating  the definition and standard contents of the fields to which the tags will relate,  and this for all financial information in OAMs.  This need is the stronger  as in IFRS, there is no more a standard format for the annual accounts, as is the case for the solo accounts in national GAAP according to the Fourth Directive.

The work undertaken by CESR has mainly focused on the advice to the Commission on these storage mechanisms, in the jargon called OAMs or Officially Appointed Mechanisms. On XBRL, CESR has not undertaken systematic analysis and the member states have not all expressed their opinions. As to the implementation in the further future,CESR has organised a working group, headed by the chairman of the Dutch AFM, Mr. Hoogervorst, which is in charge of proposing the steps to be taken for the implementation of the Transparency directive in general. Depending on future decisions, one may expect that the further process on organising financial disclosure, including the follow up on the use of XBRL will be developed and monitored by the Transparency Working group. This may include the development of taxonomies, along with overseas regulators, especially with the SEC. Indeed as the SEC has announced that it will consider allowing US issuers to use IFRS,  it would contribute to facilitate transatlantic access to the securities markets if companies on both sides would be using IFRS,  thereby allowing issuers to use their accounts for whatever purposes, and without changes, and to allow users to make transatlantic searches and comparisons.

Depending on the choices made by the Commission, it may be necessary that expensive investments will have to be made: these will have to be funded out of the Union’s budget. I hope that we will have clarity on these subjects, including on the funding, before the end of this year.

Other important applications are being planned or implemented in the field of banking and insurance supervision. Several member states have decided to apply XBRL to the information streams relating to the application of FINREP and COREP   The use is still incomplete, as is illustrated in the date found on the website of CEBS

The use of XBRL is mandatory for Finrep, being consolidated financial reporting based on IFRS   in France, Belgium, Luxembourg, Poland and Lithuania.  It is optional in Ireland and Netherlands, while its use is planned in Germany, Bulgaria.

As to Corep relating to reporting the solvency ratio of credit institutions according to the Basel II framework: its use is mandatory in the same states, Spain included. It is optional in Germany, Ireland, Netherlands Norway and Cyprus, and planned in Bulgaria.  But the situation is changing rapidly and more and more jurisdictions sign up to the use of XBRL

One may readily see other applications of the XML technology in the financial field. As the European financial markets become more and more integrated, it will be necessary that market participants have a precise view on the application of the national laws in each of the jurisdictions. Initiatives have already been undertaken by CEBS, and are likely to be extended to the securities and insurance sectors as well. Having all national laws translated in one common language  - likely to be English - would constitute an important step in that direction. Translations are very labour intensive and cumbersome exercises: they could be based on preliminary work to be undertaken in XML, as already happens since several year for translating the Canadian tax code in French. The national regulators are likely to be increasingly confronted with language translation issues for which an XML technology could offer a useful support, especially as all regulation is now derived from the directives, that are available in different languages.

All these data indicate that if there is wide interest for using XBRL in Europe, the situation is not satisfactory, as it does not cover the entire disclosure field. In general, and from the sources I consulted, there is widespread support for the XBRL- like techniques. But some argue that XBRL is the first step in a new development, and that other, more efficient technologies may be developed. I think they rather wait for the final version of colour television. Others argue that it is better to wait for sufficient support taxonomies and programmes to become available. This is a chicken and egg problem: once the data become available, the markets will develop these tools. Here again, comparability with the US would be very helpful, even if the use of the data may be different. 

In this field as in several other fields of international cooperation, it is necessary to avoid that communication between the different languages would become burdensome, or even impossible. Imagine that before the tower of Babel, the entire word had kept one single language!

A pilot programme like the one undertaken in the US would be very helpful, although some states are already that far advanced that the programme may essentially have to deal with the cross border use of the tools.  A firm option to use XBRL should be adopted, to ensure that firms investing in its use would be sure that their investment would not be wasted.

Finally, there is a macro financial interest in this debate: if no European initiative would be forthcoming, European securities will become less attractive to analysts and investors, as they will not be willing to entrust their funds to issuers about which they are less informed. At the moment that overtures are made to create the integrated transatlantic capital market, it would be regrettable that Europe would adopt a lower level of transparency.