European Central Bank - eurosystem
Možnosti vyhledávání
Home Média ECB vysvětluje Výzkum a publikace Statistika Měnová politika Euro Platební systémy a trhy Kariéra
Třídit podle
V češtině není k dispozici.

What is the BIRD?

The Banks’ Integrated Reporting Dictionary (BIRD) aims to alleviate the reporting burden for banks, foster cooperation in the field of regulatory reporting, and improve the quality of data reported to the authorities. This is achieved through the collaborative development of a harmonised data dictionary and a harmonised data model that specifies how data can be extracted from the banks’ internal IT systems to generate the reports required by authorities. The BIRD also provides transformation rules which can be applied to the data extracted from the banks’ internal IT systems to produce the above-mentioned regulatory reports.

The BIRD content, i.e. the BIRD database and other documentation, is a “public good” and therefore freely available for download.

A more detailed introduction to the BIRD methodology, as well as an overview of the BIRD member organisations, can also be found on the following pages:

BIRD MethodologyBIRD Steering Group and Experts GroupCall for expressions of interest to contribute to the BIRD

Why is the BIRD useful?

Currently, whenever a new statistical, prudential, or resolution regulatory framework is introduced or an existing one is updated, banks are expected to independently interpret the framework, extract data from their internal systems, and then transform the data to obtain the final figure required by the regulation. The BIRD offers an integrated dictionary with a redundancy-free input, where the concepts are identified and described once. This feature makes it less burdensome for banks to determine which source data to use and how to correctly process it.

Differences in how banks interpret specific regulations can have a negative impact on the quality of their output data and make it more difficult to compare different banks’ data. It is also costly and time-consuming for each bank to study new or revised legal acts separately.

Additionally, statistical frameworks vary from country to country and therefore major country-specific adjustments would currently be required to implement the BIRD at the national level. The Integrated Reporting Framework (IReF) will remedy this by establishing a common collection layer for statistical data from banks across euro area countries and it will be developed in close alignment with the BIRD. Reporting agents will then be able to directly understand and use the BIRD for statistical reporting without the need to make national adjustments.

How can banks use the BIRD?

The purpose of the BIRD is to provide a service to banks. The BIRD is a “public good” and therefore it is freely available to banks and all interested parties, such as software vendors that develop application packages for financial reporting. The adoption of the BIRD by banks is completely voluntary. It can be used to obtain additional documentation (related to regulations and guidelines) or as an “active dictionary” for procedures developed internally by banks.

The BIRD represents an “input approach” as it is not limited to the regulatory reporting requirements (i.e., the output) and it also covers the data stored in banks’ internal IT systems (i.e., the input).

The BIRD process offers two input components.

  • The Logical Data Model (LDM) is a highly normalised data model. It describes the business domain, i.e., the information (metadata), and the logical relationships of that data that are relevant for fulfilling the reporting requirements. The LDM semantically underpins the other BIRD components.
  • (Enriched) Logical Data Model (ELDM)
  • The Input Layer (IL) is a less normalised model than the LDM. It is designed to support the actual physical implementation of the BIRD by the banks. The IL is directly derived from the LDM via what is known as a “forward engineering” process.
  • (Enriched) Input Layer (EIL)

Consequently, and depending on the internal approach, banks can start the BIRD process either from the Logical Data Model or from the Input Layer. See the BIRD Methodology for more details.

What does the BIRD not provide?

The BIRD is not an IT tool, therefore providing an IT tool is outside the scope of the initiative. The BIRD does not make any changes to the banks’ internal IT systems, and it does not and cannot cover the mappings of the data from the banks’ internal IT systems to the Logical Data Model or to the Input Layer.

The BIRD is not a regulatory act and does not introduce new requirements; It is an interpretation and transposition of the regulatory reporting requirements at a more operational level and in a collaborative environment. In other words, the BIRD provides one example, based on best practice, of how the requirements set out in reporting regulations may be met.

How do banks benefit from the BIRD?

There are several advantages:

  • Redundancy-free input. Different reports can be produced from a redundancy-free input, where the concepts are identified and described once. This reduces the reporting burden for banks, improves the consistency and quality of their data, and removes the need for them to manage each mandatory data collection separately.
  • Transformation rules. Well-defined transformation rules provide a univocal interpretation of regulations and greater clarity, increasing compliance with the regulatory requirements.
  • Less time and effort are needed to analyse and comply with new reporting requirements, which increases efficiency and reduces costs.
  • Better understanding of the data and how they are produced is achieved by banks. This means that they can manage and use their data more effectively.

Which frameworks does the BIRD aim to cover?

The BIRD can cover any statistical, prudential, and resolution reporting requirements, including:

  • The IReF collection layer (statistical reporting requirements):
    • the ECB’s collection of granular credit and credit risk data (AnaCredit).
    • the ECB’s Securities Holdings Statistics (SHS).
    • the ECB’s Monetary Financial Institutions’ Balance Sheet Items (BSI) Statistics.
    • the ECB’s Monetary Financial Institutions’ Interest Rate (MIR) Statistics.
    • other statistical information required for the compilation of the balance of payments and national accounts.
  • The additional requirements under the Single Supervisory Mechanism (SSM) and the Single Resolution Mechanism (SRM).
  • The EBA’s Implementing Technical Standards (ITS), including Common Reporting (COREP) and Financial Reporting (FINREP).

The BIRD Steering Group agrees on which frameworks are covered and in which order they are included in the BIRD.

Cooperation between the authorities and banks

The ECB, European Banking Federation, the national central banks, and several participating commercial banks work closely together to develop and maintain the BIRD. The BIRD project is collaboratively steered by the banking industry and the authorities via the BIRD Steering Group. The involvement of commercial banks is crucial to ensuring that the BIRD documentation is correct and useful. Since its inception, various companies specialised in fields relevant to the BIRD, e.g., business and software solutions for bank reporting, have been involved in the BIRD project after having expressed an interest in contributing to it on a voluntary basis.

Project milestones

This section describes the BIRD project’s history, starting with the latest developments.


In the first half of 2021, the BIRD Steering Group (BIRD SG) revised the BIRD’s scope and objectives. The focus of this revision was on shaping the BIRD to be a beneficial and viable product for the banking industry and the authorities. The BIRD SG decided to include a wider set of BIRD components – Logical Data Model & Enriched Logical Data Model, Input Layer & Enriched Input Layer, Reference Output Layer & Non-reference Output Layer – and to describe transformation rules in a language that better supports the business experts. If additional resources are made available, the new BIRD process can be expanded to include add-ons, such as validation rules and technical transformation rules.

Please refer to the BIRD Methodology page for a more detailed explanation of the current BIRD components and add-ons.

In addition, the BIRD SG pointed out that the alignment of data models between the BIRD and IReF is strongly supported by the banking industry. Analysis of the IReF and its inclusion in the BIRD will be one of the project’s future priorities, together with the updates to COREP and FINREP (EBA ITS).


In 2020 the Work Stream on Data Modelling (WS DM) and Work Stream on Testing (WS T) put considerable work into reviewing and improving the BIRD process and documentation. The WS DM completed the investigation of the Input Layer and proposed introducing a Logical Data Model underpinning the BIRD layers, e.g., to ensure consistency and integrity. Meanwhile the WS T conducted testing activities based on test cases and test data on the BIRD transformation rules, written using the Validation and Transformation Language (VTL). Based on the feedback of the BIRD Experts Group and the results of both work streams’ activities, the BIRD SG initiated a discussion on the lessons learnt and possible enhancements; the conclusions prepared the ground for the review of the scope and objectives of the BIRD in 2021 (as detailed above).


In 2019 the BIRD content was expanded to incorporate more supervisory requirements, among others COREP credit risk, FINREP non-performing loans, and asset encumbrance. While these activities were ongoing the BIRD content grew at a fast pace, which led to inhomogeneous modelling choices for the BIRD layers. A dedicated group, the Work Stream on Data Modelling (WS DM), was created and tasked with reviewing and harmonising the modelling approach and defining best practice design principles. In 2019 the BIRD project was also supported by the Work Stream on Testing (WS T), which started testing the capabilities of the BIRD transformation rules by means of prototypes and test cases developed by volunteer external companies, e.g., software vendors and consulting companies.


In June 2018, following the success of the BIRD pilot, the European Banking Authority’s ITS regarding FINREP were also incorporated into the BIRD together with methodological enhancements.

In April 2017 the BIRD pilot was successfully concluded with the publication of comprehensive and detailed documentation describing how the requirements for banks, regarding the ECB Regulations on AnaCredit and SHS, can be met using data from the banks’ internal systems.

At the end of 2015, the Statistics Committee of the European System of Central Banks (STC) decided to launch a BIRD pilot for two types of statistics which had been suggested as a result of the ECB workshop with the banking industry (see below), i.e., AnaCredit and SHS. This pilot involved several volunteer national central banks together with banks contacted by each participating NCB.

In 2015 an ECB workshop with the banking industry took place. The conclusion was that the best way to launch the BIRD would be to first implement the new requirements related to the collection of granular credit and credit risk data (AnaCredit), as defined in Regulation (EU) 2016/867, and to statistics on holdings of securities (SHS), as defined in Regulation (EU) No 1011/2012.

In 2014 an STC task force recommended developing, in close collaboration with the banking industry, a European “input approach” model which would provide a standardised design of the banks’ internal data warehouses to support the process of reporting data to the authorities.

In 2013 the STC began investigating the possibility of promoting an integrated approach to supervisory and statistical data.

Legal Notice


The Banks’ Integrated Reporting Dictionary (BIRD) is a collaborative project between central banks of the European System of Central Banks and commercial banks. The aim of the dictionary is to support reporting agents in preparing datasets for transmission under the respective reporting frameworks to which they are subject.

The BIRD is made available for public purposes only. It does not constitute any type of professional advice to users, nor any official interpretation of any reporting requirement. The information it provides must not be understood as binding, or otherwise legally relevant concerning compliance with reporting obligations. The BIRD does not affect, amend, or replace the responsibilities of reporting agents towards the authorities to which they provide statistical information. Reporting agents remain fully responsible for organising their internal reporting systems and ensuring the correctness of their reports to the authorities.

The project sponsors shall not be liable for any gap, error, or inaccuracy in the content of BIRD, or for any action taken in reliance thereon including, without limitation, any reference made in contractual agreements. They expressly disclaim all warranties, expressed or implied, as to the accuracy of any of the content provided or as to the merchantability or fitness of the content provided in the BIRD.

The project sponsors do not endorse or take responsibility for any replication of the content of BIRD, or any other form of redistribution.


BIRD users may make free use of the information obtained directly from it. The original content of BIRD cannot be changed; however, the content may be used for the creation of new products and the commercial use thereof. Whenever re-using the content of BIRD for the creation of a new product the source “BIRD” must be quoted correctly. Where the information is incorporated in documents that are sold (regardless of the medium), the natural or legal person selling the information must inform buyers, both before they pay any subscription or fee and each time they access the information taken from this website, that the information may be obtained free of charge through the BIRD website.