Primary tabs

De Nederlandsche Bank NV - Statistics Division

Initially the scope of an integrated reporting framework should focus on credit institutions. We expect their benefits to be the highest. Moreover, such focus would be in line with projects -under the umbrella of the ESCB- already underway. Other supervised financial intermediaries (e.g. investment firms) may be considered later.
As per the data collections, we expect all reporting requirements stemming from ESCB, EBA, and SRB legislation addressed to credit institutions to be considered by the feasibility study, including the statistical requirements covered by the ESCB Integrated Reporting Framework (IReF), EBA reporting requirements regarding supervision and resolution, data requirements from international organisations (e.g. BIS and IMF) and ESAs. In addition, ad-hoc requests should -to the extent possible- follow the standard data dictionary. However, supervisors and resolution authorities should maintain flexibility in requesting ad-hoc requests as this is the nature of ad-hoc requests.
In general we agree with the issues identified and proposals made in the discussion paper, although some issues, notably the push-pull approach and the centralisation of analysis and dissemination, seem to go beyond the original assignment. In addition, we would like to note that any centralisation of analysis and dissemination should be considered only after the common data dictionary phase has been completed.
Not relevantSomewhat relevantRelevantHighly relevant
Training / additional staff (skills)X   
IT changesX   
Changes in processesX   
Changes needed in the context of other counterparties / third-party providersX   
Time required to find new solutionsX   
Other (please specify)X   
Confirmed
Highly agreeAgreeSomewhat agreeDon’t agree
Data Dictionary - Semantic levelX   
Data Dictionary - Syntactic levelX   
Data Dictionary - Infrastructure levelX   
Data collection - Semantic levelX   
Data collection - Syntactic levelX   
Data collection - Infrastructure levelX   
Data transformation - Semantic levelX   
Data transformation - Syntactic levelX   
Data transformation - Infrastructure levelX   
Data exploration - Semantic levelX   
Data exploration - Syntactic levelX   
Data exploration - Infrastructure levelX   
In general, we agree with a holistic approach, although the focus should be clearly on the parts of the reporting process within the scope of article 430c of the CRR. Hence, most devotion should be given to data definition, data collection and data transformation. Moreover, these are typically the areas that impact on reporting agents and the reporting burden.

In addition, we would like to point out that any integrated reporting system should be based on a coherent suit of underlying legal requirements in terms of level 1 frameworks (such as accounting and prudential, statistical and resolution) in order to achieve an integrated system that significantly reduces the reporting burden experienced by the industry. Therefore, we recommend the Commission to perform in parallel an analysis of level 1 legislation aiming at a higher level of alignment. We assume the Commission will find that some revisions to the level 1 legislation will be necessary. Thereafter, the revised level 1 legislation could serve as the basis for enhancing the common data dictionary.

At present, we see no need to discuss and assess any centralisation of data analysis, and dissemination (i.e. exploration phase). These are typically tasks for national central banks and national competent authorities.
We agree to large extent with the assessments and text regarding the data dictionary. We feel that a standard data dictionary is key to the process to collect data. In fact, one should start with semantical harmonisation before moving forward with any logical and technical harmonisation. Hence, a common data dictionary is a pre-requisite for any integrated reporting system.

A data dictionary should be clear, concise, unique, and redundancy free. Therefore, definitions should be described in detail and, possibly, at (more) granular level. Any overlap or 'similarity' between definition should be documented and, in future, be discussed between users/legislators in order to overcome these (little) difference. One would need to ask: can I also perform my analysis/job with 'similar' data. In addition, clear definitions may -in the short run- also prevent overlapping (ad hoc) requests from different authorities, decreasing the reporting burden. Both paragraphs 74 and 75 seem to be in line with these views, although the importance of a common data dictionary should be stressed more strongly.

We would also like to point at the transitional arrangements when designing a common data dictionary when a data dictionary is based on so-called principle-based underlying legal requirements. Both accounting and prudential and resolution frameworks are principle-based and allow banks to apply a range of outcomes under one and the same legal requirement. For instance, banks may use internal models for expected credit losses in accounting and under the CRR and multiple outcomes are possible in terms of data attributes. When designing a common data dictionary such flexibility should be maintained and therefore the transition to the rules-based data dictionary deserves attention.
SignificantlyModeratelyLow
Understanding reporting regulationX  
Extracting data from internal systemX  
Processing data (including data reconciliation before reporting)X  
Exchanging data and monitoring regulators’ feedbackX  
Exploring regulatory dataX  
Preparing regulatory disclosure compliance.X  
Other processes of institutionsX  
Option 3 seems at this stage impossible to pursue, because the collection/submission of prudential data at a granular level is not sufficient: prudential/accounting aggregates and indicators are still needed for several reasons (e.g. compliance with principles and accounting data subject to expert judgement).

In turn, there is an increasing demand for granular data (which can be easily combined with other sources). To limit the overall reporting burden for institutions and authorities, reporting more granular data instead of aggregates is desirable. Therefore, for data other than aggregated data required to comply with regulations or data that require expert judgement in the compilation, option 2 seems to be preferred once new granular data requirements in that field are defined. Overall it should be clear that option 2 (and option 3) require a completely new legal framework as aspects of data, data ownership and data responsibilities should be defined before such a system becomes operational.

As per the expected costs for the definition and maintenance of derivation rules, experiences from the BIRD project indeed confirm these can be high. More in particular when the requirements change often. If transformation rules are defined jointly by reporting agents and authorities, the former should clearly remain responsible for the implementation of those. It is not the task of the authorities to ensure accuracy of the reports.
Highly (1)Medium (2)Low (3)No costs (4)
Collection/compilation of the granular dataX   
Additional aggregate calculations due to feedback loops and anchor valuesX   
Costs of setting up a common set of transformations*X   
Costs of executing the common set of transformations**X   
Costs of maintaining a common set of transformationsX   
IT resourcesX   
Human resourcesX   
Complexity of the regulatory reporting requirementsX   
Data duplicationX   
Other: please specifyX   
Highly (1)Medium (2)Low (3)No benefits (4)
Reducing the number of resubmissionsX   
Less additional national reporting requestsX   
Further cross-country harmonisation and standardisationX   
Level playing field in the application of the requirementsX   
Simplification of the internal reporting processX   
Reduce data duplicationsX   
Complexity of the reporting requirementsX   
Other: please specifyX   
In our view the real value added for all stakeholders is the harmonisation and integration of the reporting frameworks across countries and domains, rather than having a central data collection point. Therefore, the costs to replace legacy (systems) and develop a central data collection point potentially outweigh the benefits for all stakeholders. In addition, we fear that pursuing such a drastic change from the current approach would delay the entire endeavour, not allowing for any benefits in the short run.

Therefore, we prefer to follow a stepwise approach, with two processes in parallel: first the IReF covering the statistical requirements. Second, the integrated collection of resolution and prudential data. Ultimately, both work streams could converge into a fully integrated system as soon as a common data dictionary and common data model are developed and implemented, and there is a positive business case. During the transition period, it is i) necessary to enhance the framework for sharing data among authorities and ii) to develop a single business register of (reporting) entities, allowing for an assessment of who reports what.
not valuable at allvaluable to a degreevaluablehighly valuable
Data definition – Involvement    
Data definition – Cost contribution    
Date collection – Involvement    
Date collection – Cost contribution    
Data transformation – Involvement    
Data transformation – Cost contribution    
Data exploration – Involvement    
Data exploration – Cost contribution    
Data dictionary – Involvement    
Data dictionary – Cost contribution    
Granularity – Involvement    
Granularity – Cost contribution    
Architectures – Involvement    
Architectures – Cost contribution    
Governance – Involvement    
Governance – Cost contribution    
Other – Involvement    
Other – Cost contribution    
In our view the focus should be clearly on the common data dictionary, data model and -possibly- an integrated framework. Thereafter one can assess the pro's and con's, as well as costs and benefits, of the respective modalities. Hence, this is point is of secondary importance at this stage.

At any rate, any integrated system to be developed in the future should be 'future-proof': against the background of changing circumstances and technological advances, the system should be flexible enough to accommodate for various ways to manage data in Europe.
There is a clear need for a thorough assessment of any/all legal obstacles, stemming from both (primary) European law and national legislation, for the various proposals in the discussion paper.

It goes without saying that authorities should aim at lowering the reporting burden, and reuse to the extent possible existing information. However, (national) competent authorities should maintain full control of their own data requests. In addition, any coordination mechanism of ad-hoc data requests, should not limit authorities' flexibility to collect prudential and statistical information in case of (policy) needs. For statistical data collection, authorities should retain the right to align with international statistical standards such as the SNA and BPM.

As per the set-up of a joint committee, ahead of any formal establishment we would welcome an informal coordination mechanism among authorities in order to create a forum between authorities to discuss the main aspects of the Integrated reporting system (i.e. the development of the common data dictionary and data model). In addition, given the current state of legal structures in the European union, the Commission could make use of existing bodies mandated to deliver legal requirements at level 1, 2 and 3 (Council/EC/EP, ESA's and national authorities in coordination with the ESCB statistical function where appropriate). A first step could be to align the legal mandates of all relevant bodies to ensure the bodies need to consult each other before deciding on new reporting requirements, and ideally start assessing the level 1 legislation aiming at a higher level of alignment.
Yes
Fabienne Fortanier (Director Statistics Division)
D