As we talk to global and national banks, regulators and IT staff, we have been heartened by the amount of preparation in plans around implementing XBRL. Of course XBRL is nothing new to France. What is slightly different though is that the standards to be implemented will now be determined at a European level and implemented, in largely the same manner, by national regulators across the EU.
This ushers in a two stage regulatory analysis. At a National level, there is, as ever, the requirement for micro-analysis and country level macro-analysis. Now however, there is also the intention of further analysis at a macro European level. The comparability and transparency afforded by the common use of XBRL as the Implementation Technical Standard will enable such analysis to take place.
“No more Mr Nice Guy”
Why does this matter? It matters because the regulators have now gone beyond using XBRL principally as a means to transfer data for the sole use of a particular country to one whose purpose is to facilitate regulatory analysis between regulatory agencies and countries. We have already seen examples in one of our customers of European and Far East regulators comparing submissions. In the event of such post submission regulatory enquiries, responses will be demanded within 24 to 48 hours. This has far reaching implications. In the next twelve months, although regulators will be finding their feet with this new level of data, they are getting more efficient more quickly. Although the ‘soft landing’ phase relevant to the SEC’s implementation has lasted some three years, there are now widespread indications of a toughening regime emerging. The financial press in the US, as you would expect, are warning of the SEC’s toughening up their regime by briefings on “no more Mr Nice Guy” (on a lack of quality in the data submitted) and on the use of “Robocop” (XBRL Analysis) to more quickly identify those who are less proficient at the use of XBRL and need ‘encouragement”.
More data, more often
In terms of re-defining how XBRL is used, make no mistake that the implementation of CRD IV is a major step change. CoReP and FinReP are largely based on Basel III standards and therefore expect further convergence in financial regulatory reporting to accelerate. As the EBA said at a recent meeting when asked if they felt that a quarterly cycle of submission would remain the long-term norm, their answer was “No – we see more data – more often being more likely”. If they haven’t yet thought of it – a great way to collect data for the upcoming Asset Quality Review of European banks would be to leverage the information within the CoReP Taxonomies.
We see and predict that the widespread use of XBRL will only increase. At the moment, Banks are concentrating in Europe on CRD IV. Swift and DTCC are however three years into a trial of XBRL being used to remove ‘interpretation and timing risk from the communication of Corporate Actions through the implementation of ISO 20022. In the UK, some 1.6m firms have, for the last two years, been filing Corporate Tax returns in XBRL. The project is viewed very positively by all its stakeholders and furthers the march of XBRL. So too in Australia, Standardised Business Reporting (SBR) takes the concept much further by enabling software applications to automatically generate reports from SBR supporting applications direct to ten different agencies based on one Taxonomy. What we are seeing is that XBRL is becoming increasingly pervasive, highly productive but most importantly absolutely persistent. No regulatory body or agency has ever turned an XBRL based system off!
What’s happening is that multiple agencies are now asking for the same data multiple times. This leads to re-stating the same data again and again for multiple agencies. This is sometimes referred to as a “build many-report many” model. One large global bank that we know deals with over 470 regulators and their bill for Regulatory Reporting in this way is now in excess of $1Bn.
Another approach that firms can take though, is the “build-once-report many” model. The good news is that XBRL is a key enabler in this approach and the savings can average between 40% and 60% over a three-year timeframe from ‘day one’ costs.
Driving the effectiveness in utilising XBRL
Therefore, given the increase in the prevalence of XBRL on an international scale, we have outlined below several key areas that over the last twelve years that we have been involved in XBRL-based reporting have proven to be important areas in driving the effectiveness in utilising XBRL. The list of points is not necessarily meant to be prescriptive, but in our experience, the more a firm has these points implemented, the more likely they are to be approaching ‘best-in-class’ and prepared for the future. The opposite is of course possible for those who don’t yet have them in place!
The purpose of XBRL for banks is to facilitate Regulatory Analysis
A surprising number of firms that we talk with are missing this axiomatic point. It is the only point of implementing XBRL. What does the information gathered say about your firm? Thinking through how you will respond if you are asked about data provenance should be a key area for those responsible for filing. Also, it’s worth reflecting that if firms themselves do not analyse the information – then the regulatory agencies will almost certainly know more about your firm than you do.
Establish an Enterprise-wide XBRL Strategy
If your Bank already has this than you are indeed in a good place. In such a bank, business ‘owns’ the data and IT has ‘stewardship’ of the data. You may even have a Chief Data Officer who understands the broad implications of the impact of XBRL including data quality, process around the data and matching process with regulatory controls. You will also have people who understand where the regulators globally are taking XBRL and are feeling relaxed about that. Remember that the Reserve Bank of India is implementing an Automated Data Flow XBRL-based regime. XBRL defines the data and the Tier 1 banks in India are required to automate populating it. No need to ‘push’ data – the regulator will ‘pull’ it from the bank. If banks want a vision of where XBRL reporting can go – this is one of the more ambitious.
Understand the difference between Templates and Taxonomies
In some cases we are seeing firms relying on vendors to interpret the regulators published taxonomy and after vendor analysis, those vendors send their customers templates which facilitate creation of XBRL filings. This is almost certainly not that great an option going forward. The reason being that we know that taxonomies for CRD IV will be quarterly and will change. The ability of any vendor to hard code these taxonomies accurately for differing firms is difficult enough, but on a quarter after quarter basis makes no sense for the vendors as their costs for hard coding templates will escalate and ultimately will be less accurate in line with the XBRL Conformance Suite 2.0 (Your vendor does use the conformance suite from XII?)
Acquire a Taxonomy Rendering Tool and teach the business how to use it
One of the first things in this new order is to actually understand what any given taxonomy is actually asking for. The way that seems to us the most effective way for firms going forward, is to be able to simply download a given taxonomy and use the rendering tool to render it to a human readable format. In this way, what regulators are asking for can be interpreted by the business immediately, not only in terms of what data is being asked for, but also how the regulator intends that data should be checked or ‘validated’. By rendering, firms have the ability to do this immediately when the regulator releases a new taxonomy. What results is firms have far more time to prepare for changes rather than sit and wait for a vendor to define what they might be for them. These tools also provide a crucial capability going forward and that is the capability to compare one quarter’s taxonomy to another. Knowing the net difference between two taxonomies speeds up mapping of any new data requirements and the ‘sun-setting’ of any data no longer required.
Apply external Data Validations as close to point of extraction from source systems
One of the issues firms are grappling with is that a number of firms already have a regulatory ‘data store’ or similar. The data in these data stores was never prepared or exposed to the external regulatory checks or ‘validations’. When firms try to validate the data previously gathered, these external regulatory validations will almost certainly fail to validate – the data was prepared without any reference to these checks. The issue then is how to correct the error? Is it prudent to manually change the number at the reporting level (and lose lineage) or is it better to correct further back in the information supply chain? As the heading above suggest, in our experience, the closer to the point of extraction that these validations can be applied, the less errors from external regulatory validations will result in calculation errors.
Think of building a secure archive for regulatory filings
The resulting regulatory filings are increasingly complex and comprehensive and open to easy analysis using freely available XBRL ‘readers’. One bank recently admitted that existing (non-XBRL) filings currently reside on a shared departmental drive. Given the completeness of the financial data in these Instance Documents (as the filings are called) it is important that when completed, it is defined who can access them, who accessed them and for what purpose. This is important, as there is often a gap between banks having to respond to regulators and when they are required to report to the market. Without proper security, the data can and may be compromised.
Define and develop a Regulatory Post-submission enquiry process – before you need one!
Finally, going back to our starting point that the purpose of XBRL-based filing is to facilitate regulatory analysis, it is important to review your existing processes and procedures in the light of the style, substance and speed of likely enquiries that the regulator may bring back to you. Being given a regulatory enquiry is not the time to start thinking of what the process should be and who should be involved.
