General Instructions

The BIRD input layer is intended to support credit institutions to generate reports as required by regulatory authorities.

The unit populating the BIRD input layer may be:

  1. A head office, including domestic branches
  2. A foreign branch
  3. A legal entity
  4. A group or part of a group

Input data comprise the relationships between the head office and foreign branches, among foreign branches, and the relation among head of the group and its subsidiaries.

The BIRD can manage the report at the level of a branch, legal entity, and group.

The consolidation process is not envisaged to be managed by the BIRD, however the input layer can be fed with information at consolidated level or at individual level

Example : An illustrative example about the points above
Suppose there are two legal entities A and B, where B is subsidiary of A. A is composed by a Head office (the domestic branch) and an EU branch. The entity A and B belong to the group A. The following chart shows the relations between the entities:

The BIRD input layer may refer to:

  • The group A
  • The legal entity A, in which case it contains information on all the branches. The input layer is built in such a way that it is also possible to generate the reports for the units included in the legal entity.
  • The Headquarter alone or any of the foreign branches alone.
  • The legal entity B.


The following section describes the conventions applied in the BIRD model.

Sign convention

The BIRD input cubes generally contains one or more measures. A measure is a variable that quantitatively represents a phenomenon and in general points towards a monetary (MNTRY) domain.

In order to correctly represent a phenomenon a sign convention has to be followed in order to assess if a measure should be reported in the input layer as positive or negative.

The double-entry bookkeeping system is a standard used in accounting to report economic transactions. Each economic transaction is reported in two different accounts called debit and credit.

Under Debit assets and expense accounts are reported, while in Credit liability, equity and income accounts are recorded.

Debit Credit
Asset Liability, Equity
Expense (cost) Income (revenue)

The following table explains the sign convention to follow:

Balance/Increase Reversal /Decrease
Asset Positive (+) Negative (-)
Liabilities Positive (+) Negative (-)
Equity Positive (+) Negative (-)
Income Positive (+) Negative (-)
Expense Positive (+) Negative (-)

To report the values with the correct sign it is necessary to understand:

  1. The CUBE related account (Asset, Liabilities, Equity, Income, Expense)
  2. The nature of the measure used within the cube (Balance/Increase, Reversal/Decrease)

However there are exceptions that are on a case by case basis and are pointed out whenever they arise.Illustrative examples.

Case 1:

Suppose there is a loan measured at amortized cost. The loan is unimpaired and is reported in the cube other loans (OTHR_LN) and belongs to asset accounts.

The measures are represented by the following variables:

GRSS_CRRYNG_AMNT_E_INTRST Gross carrying amount excluding interest + 1000
ACCRD_INTRST Accrued interest + 10
ACCMLTD_IMPRMNT Accumulated impairment – 15

The amounts are represented in the asset accounts, respectively by:

  • A balance in assets of gross carrying amount excluding interest (positive value)
  • A balance in assets of accrued interest (positive value)
  • A decrease in assets by a loss allowance (negative value)

Suppose now that, subsequently, the loan is credit impaired, and the entity recognizes a loss of a 100 in the balance sheet. The entity shall recognize a decrease in the asset account via accumulated impairment of 100. The result for the accumulated impairment should decrease of 100 and the amount shall be:

ACCMLTD_IMPRMNT Accumulated impairment -15 + (-100) = – 115

Case 2:

Suppose the entity issues a 10 year bond measured at fair value with a par value of 1000 with annual fixed coupon rate. The interest rate is consistent with market rates for bonds with similar characteristics.

In subsequent reporting periods the following market prices for the bond are observed and the accumulated changes in fair value are calculated:

DESCRIPTION FV (Fair value) ACCMLTD_CHNG_FV (Accumulated changes in fair value)
Fair value + 1000 0
Fair value + 900 – 100 = 900 – 1000
Fair value + 1050 + 50 = (1050 – 900) + (-100)

The amounts are represented, for each reporting period, in the liability accounts, by:

  • A balance in liability of fair value (positive value)
  • In period 1, a decrease in liability with respect to period 0 (negative value)
  • In period 2, an increase in liability with respect to period 0 (positive value)
  • Case 3:

On 01/01/20X1 the entity enters a long position forward derivative contract with the following characteristics:

UNDERLYNG 1000 shares of stock XYZ

Assume that the annually compound risk free rate is constant at 1% and, for the sake of simplicity, no dividend yield is payed during the period.

In the following table are represented the fair values calculated for each reporting period:

0 01/01/20X1 100 0
1 31/03/20X1 120 + 19760
2 30/06/20X1 150 + 49510
3 30/09/20X1 90 – 10750
4 31/12/20X1 80 – 21000

Because of the nature of the derivative contract, the reported value could be either an asset or liability depending on its fair value at reporting period.

In this case a different convention is used and a positive fair value will be considered under asset accounts and a negative fair value will be considered under liabilities account.

Null / not applicable

In specific cases it is necessary to indicate that a particular concept (i.e. a variable) is not applicable or does not exist. In the BIRD input layer the following convention applies:

For enumerated variables (i.e. based on an enumerated domain) the member “Not applicable (0)” is used (please note that this also applies for Boolean variables). For other variables (e.g. numeric, string or date) the value null indicates that the concept is not applicable or does not exist in this case. Please note that the second case also includes identifiers (e.g. Protection identifier).

(Input) Parameters

In the BIRD input parameters (set by the reporting agent) are necessary such that the system is able to process the (input) data in a sound manner. These parameters are described in a specific cube called Parameters (PRMTRS).

An example of such a parameter is the variable Is carrying amount derived. Its value indicates if the BIRD derivation rule for the carrying amount should be applied or not.

Another example is the date of the opening balance that may differ from the first day of the year.


Time in the BIRD model means, that there exists the functionality such that, given a specific point in time, we are able to univocally extract information about an instance (e.g. a specific loan) from the BIRD model.

It is established by the variable Reference date (DT_RFRNC) in the input layer cubes, allowing the extraction of one specific instance (e.g. one specific loan) at a particular point in time from the BIRD input layer. E.g. a loan for different points in time is represented as indicated in the following table:

Instrument unique identifier Reference date Carrying amount
exampleLoan 31/01/2018 31
exampleLoan 28/02/2018 29
exampleLoan 31/03/2018 23
exampleLoan 30/04/2018 19

Table 1: a loan for different points in time

Please note that the variable Reference date (DT_RFRNC) acts as a dimension in the input cubes.

Unique identification via identifiers

The reporting agent needs to ensure the unique identification of objects via its identifier. For example an instrument needs to be uniquely identified by its Instrument unique identifier. The same holds true for credit facilities and their Credit facility unique identifier or protections and their Protection identifier.

Please note that the Instrument unique identifier is not to be confused with the Instrument identifier reported in AnaCredit. The first one acts as a dimension in the instrument cubes allowing the identification of a record in these cubes while the later only acts as an observation.

BIRD scope with respect to Individual- & Consolidated reporting

The BIRD process for data generation covers different output requirements with respect to the data that is taken into consideration when generating the reports, e.g. the AnaCredit output layer comprises individual content while FinRep may comprise individual and / or consolidated content.

In general banking group structures are in scope of the BIRD although “consolidation logic” meaning that creating consolidated numbers from individual figures is not in scope of the BIRD. Therefore consolidated figures (e.g. the carrying amount of a loan as it contributes in the consolidated balance sheet) need to be provided in the input layer together with individual figures.

The following example illustrates a group structure and related instruments and indicates what is in scope of the BIRD and what is not in scope of the BIRD.

Table 2: Group structure example

Where the group (ABCDE or BCD) itself and each member of the group apply IFRS except for D who applies nGAAP.

Any composition of the group ((ABE, AB,E,C) can be described within the BIRD In particular in the Cube Group and related entities (BIRD_GRP_2) it is store for each group structure a specific group ID (GRP_ID)

As an example for the group (ABCDEF) and the different perimeter ABE

Cube Group
Group ID Legal entity ID Consolidation method Scope of consolidation
ABCDEF A 1 Prudential
ABCDEF B 1 Prudential
ABCDEF C 1 Prudential
ABCDEF D 1 Prudential
ABCDEF E 1 Prudential
ABCDEF F 2 Accounting
ABE A 1 Prudential
ABE B 1 Prudential
ABE E 1 Prudential

Perspective, objective, subjective information

Due to the fact that the BIRD covers individual and group related reporting requirements using the same input layer it is necessary to be able to perceive an instrument or an aggregate from different perspectives (e.g. individual / solo perspective or the perspective from group G).

By objective information we refer to information that is independent of the perspective taken, e.g. the currency of a loan is independent of the perception in the sense that it is the same for solo and consolidated reporting. On the other hand, subjective information is dependent on the taken perspective, e.g. the accounting classification of a loan may be different with respect to solo and consolidated reporting.

In the BIRD input layer such a loan is presented in the following way:

Instrument unique identifier Perspective identifier Currency Accounting classification
singleLoan D Euro nGAAP: Trading financial assets
singleLoan ABCDE Euro IFRS: Financial assets held for trading

Table 3: a single loan perceived from two different perspectives

Please note that only subjective information may change (e.g. Accounting classification) while object information (e.g. Currency) is static. A detailed description of subjective information is defined in the section dedicated to auxiliary table.

In order to define the perspective identifier the cube Perspective information (BIRD_PRSPCTV_INF_1) need to be feed following the example of the group reported above

We can manage the information reported by the company D following accounting framework of the group (IFRS) and its own accounting framework (nGAAP)

Perspective information
Accounting framework Counterparty identifier Perspective identifier

Depending on the accounting framework different perspective identifiers Internal, intra-group instruments / aggregates will be reported in the instrument cubes.

Because the BIRD covers individual and group related reporting requirements it is necessary to identify internal (i.e. transactions between a legal entity and its foreign branches) and intra-group transactions (i.e. transactions between legal entities that are part of the same group).

For instruments the relevant counterparty is easily identifiable and using the information about the group structure (see cube Group (GRP) and Composition of the legal entity (CMPSTN_LGL_ENTTY)) internal and intra-group instruments can be identified and if necessary omitted (e.g. in case of individual FinRep reporting internal instruments should not be considered, in case of consolidated FinRep reporting intra-group instruments should not be considered).

For cubes comprising aggregated information (e.g. Other assets (OTHR_ASSTS)) this approach cannot be applied. In this case the perspective on the data is relevant, consequently the value of internal, intra-group aggregates is controlled by the Perspective identifier (PRSPCTV_ID).

Information regarding Accounting frameworks

Many dictionary elements and transformation rules are related to accounting standards and this either may refer to IFRS or national accounting frameworks (national Generally Accepted Accounting Principles or nGAAP). For the variable ACCNTNG_FRMWK, the member 2 has to be chosen for IFRS. Alternatively, for nGAAP the member 1 shall be used if the nGAAP bases on Bank Account Directive (BAD)2, or member 3 if the respective nGAAP is fully “compatible with IFRS”. Depending on the reporting institution, the accounting framework for the solo (ACCNTNG_FRMWK_SL) and, if applicable, also for the group reporting (ACCNTNG_FRMWK_GRP) needs to be populated.

For nGAAP compatible with IFRS, the IFRS dictionary elements and transformation rules apply. For nGAAP based on BAD the national central banks and supervisory authorities defined in close collaboration with the banking industry which data could be reported under nGAAP. This exercise was done for the FINREP templates, hence, on an aggregated level. The results of this exercise should also be used for BIRD, if applicable on a granular level. Further information on particular nGAAP issues will be provided in the subsequent sections of the BIRD technical guideline. In general, the reporters shall choose instead of IFRS related elements the respective nGAAP attributes. One example: If it was decided for nGAAP reporting that levels for fair value hierarchy (in the sense of IFRS 13) could not be provided, e.g. according to DE and FR nGAAP, since it does not exist, the variable FV_HRRCHY shall be populated with member “not applicable”.

Independent from the accounting standard, for some institutions the financial year might deviate from the calendar year. If this is the case it is particularly important to choose the correct reference date (variable DT_RFRNC). If an institution also produces consolidated financial information for FINREP, the reference date shall be adjusted accordingly, i.e. relating to the date to which the consolidated financial information refer. Similarly, the same applies to flow information which relate to a specific period, instead of end of period stock data.

Auxiliary table

The Auxiliary table are tables populated with specific values developed by the BIRDused for BIRD transformations.

The auxiliary cubes contain data and are not stored in the BIRD DB in which only metadata are described.

Auxiliary table on sector classification

One auxiliary cube has been developed to assign the sector classification of the International Organisation. The treatment is different according to different reporting framework. The auxiliary cube can be used to derive the proper classification.

In particular the International Organisation are stored in the input cubes (International organisation code) in counterparty cubes. Starting form this codification the distinction between International Organisation (IO) and Multilateral Development Banks (MDB), that are assigned to general government and to credit institution respectively by FinRep, is reported in column “classification for FinRep”.

The distinction between IO and MDB for FinRep classification has been derived first using the classification contained in CRR art 117 and 118(column BIRD classification following art 117 and 118 CRR ) , then the Balance of Payment statistics – ECB classification of International organisation sector classification (clumn Sector Classigfication) and finally using the list of “Non-bank financial institutions” published by the BIS in its Guidelines for reporting the BIS international banking statistics following the Q&A EBA nr 138.

Auxiliary table on subjective information

The following table provides an overview on all subjective variables and the reasoning why the variable may have different values (green columns):

  1. The fair value is calculated according to the formula {Number of shares}*({Spot price} – {Strike price}*(1+{risk free})^(-{time to maturity}) )
  2. Council Directive 86/635/EEC of 8 December 1986 on the annual accounts and consolidated accounts of banks and other financial institutions (OJ L 372, 31.12.1986, p. 1).