You are looking at `F 01.01`, the Financial Reporting (FINREP) balance sheet template of the Data Point Model (DPM), row 010, column 010. Somewhere behind that cell sits an actual fact: a metric, a set of dimension members, a context, a value. The template does not tell you any of that directly. It tells you a position, a coordinate on a grid. The question worth asking, and the one an accountant reviewing tagged output must be able to answer, is what fact that coordinate actually resolves to.
The answer lives in the table linkbase. A table linkbase defines breakdowns along structural axes: commonly an x axis for columns, a y axis for rows, and a z axis for repeating the same template across something like currency or reporting scope, though a DPM table may carry several axes of a kind, including open ones. Each axis is built from nodes, and each node pins down part of the answer: a metric on one axis, a set of dimension members on another. Where the applicable axes intersect, you get a cell, and that intersection is what fixes the complete metric-plus-dimension vector a fact in that cell must carry. The template you see rendered is therefore not a separately hand-authored layout. Its logical arrangement is the table linkbase, walked and drawn, with only the final visual formatting left to the rendering software. This is worth contrasting with the presentation linkbase used elsewhere in XBRL (the eXtensible Business Reporting Language), which only orders concepts into a tree for display and has no concept of rows crossed with columns at all. A presentation linkbase can tell you that one concept comes before another; only a table linkbase can tell you that a cell sits at the intersection of a row and a column.
Not every table behaves the same way once you get there. A closed table, like `F 01.01` itself, fixes both its rows and its columns in the taxonomy: the row for cash balances at central banks and the column for carrying amount are defined in advance, and every filer's cell at that intersection means the same thing. An open table works differently: the filer enumerates the rows (or columns) at submission time. Large exposures templates and similar granular breakdowns are open for exactly this reason, producing one row per counterparty, each row keyed by a typed dimension carrying a value such as a counterparty identifier that the taxonomy could never have enumerated in advance. Reading an open table therefore means reading the typed-dimension key alongside the metric, because the key is what makes each row distinct.
Before any of those cells can be trusted, the submission has to declare which templates it actually contains. In current European Banking Authority (EBA) filings in xBRL-CSV (the comma-separated-values serialisation), that declaration is one `fi:filed` fact per template, identified by an `fi:template` typed dimension carrying the template code, per the XBRL Filing Indicators specification. Filings in xBRL-XML (the Extensible Markup Language serialisation), which remain current for the European Insurance and Occupational Pensions Authority (EIOPA) and its Solvency II returns and are legacy for the EBA, use `find:filingIndicator` elements, drawn from the Eurofiling `find` namespace, each carrying a template code such as `C 01.00`, `F 01.01`, or `S.02.01` as its content and a `filed` attribute. Whichever form a given return carries, the rules demand completeness: every template in the selected module needs an explicit indicator, positive or negative, and omitting one entirely violates EBA filing rule 1.6, which is a rejection. The unforgiving case is the wrong negative indicator. The file validates cleanly, the receiving system treats the template as deliberately out of scope, and nothing flags the gap at intake. A mis-scoped template is the kind of error a filer discovers on the next reporting cycle, from a supervisor's query, not from the validator that just accepted the file.
Arithmetic and consistency checks are handled separately, by validation rules, the "V-rules". Most intra-module V-rules ship as XBRL Formula assertions within the taxonomy itself, and on the EBA side DPM 2.0 formalises their authoring in DPM-XL, the metamodel's own rules language; a minority of the Implementing Technical Standards' (ITS) validation rules are enforced outside the taxonomy. Whichever form the rules take, they cover cardinal sanity, such as a balance sheet actually balancing; intra-template arithmetic, such as subcategories summing to a reported total; and cross-template consistency, such as a value reported in Common Reporting (COREP) agreeing with the corresponding value in FINREP. Filers run the taxonomy's assertions locally, typically with Arelle or a DPM-aware processor, and the National Competent Authority validates again at intake before any onward transmission to the EBA or EIOPA.
That gives a reviewer a concrete checklist when an agent or tool has produced a DPM return. Confirm the filing indicators, the `fi:filed` facts in an EBA xBRL-CSV return or the `find:filingIndicator` elements in an xBRL-XML one, are complete and carry the right polarity for every template in the module, since a wrong negative will validate and silently scope a template out. Confirm the V-rules ran and came back clean, rather than assuming a file that parses is a file that validates. For open tables, check that the typed-dimension keys are consistent and correctly cased, since string-typed member equality is textual and a stray case mismatch silently creates a duplicate counterparty. And when a cell's value looks wrong, trace it back through the table linkbase to its metric and dimension vector rather than trusting the row and column label alone, because the label is a rendering, not the fact.
The Table Linkbase 1.0 specification of XBRL (the eXtensible Business Reporting Language) defines tabular fact layouts: commonly an x axis (columns), y axis (rows), and z axis (sheets), though a Data Point Model (DPM) table may carry several axes of a kind, including open axes. Rendering software intersects the axes to produce the template's logical layout; final visual formatting is left to the renderer.
A template like F 01.01 (the Financial Reporting (FINREP) balance sheet) follows the table linkbase, not a separate hand-authored layout: each cell's row and column nodes pin down the metric and dimension members that a fact in that cell must carry.
Closed tables fix row and column membership in the taxonomy. Open tables let the filer enumerate rows at submission time (large exposures C 28-C 30, Solvency II asset lists) and rely on typed dimensions for the filer-supplied keys.
Filing indicators declare which templates a submission actually contains: in the European Banking Authority's (EBA) xBRL-CSV format (the comma-separated-values serialisation), one fi:filed fact per template, identified by a fi:template typed dimension; in xBRL-XML (the Extensible Markup Language serialisation — legacy for the EBA, still current for the European Insurance and Occupational Pensions Authority's (EIOPA) Solvency II), find:filingIndicator elements carrying template codes like C 01.00 or S.02.01. Every template in the module needs an explicit positive or negative indicator — omitting one violates EBA filing rule 1.6 and is a rejection; the insidious failure is a wrong negative indicator, which validates cleanly and silently scopes the template out.
Most intra-module validation rules ('V-rules') ship with the taxonomy as XBRL Formula assertions (on the European Banking Authority's side authored in DPM-XL under DPM 2.0); a minority of the Implementing Technical Standards' validation rules are enforced outside the taxonomy. Filers run the taxonomy assertions locally (Arelle or DPM-aware processors); the National Competent Authority (NCA) validates again at intake before onward transmission to the European Banking Authority (EBA) or the European Insurance and Occupational Pensions Authority (EIOPA).
F 01.01 (FINREP balance sheet), row 030, column 010
|
| table linkbase: y-axis node (row 030)
| x-axis node (column 010)
v
metric: "Carrying amount" (monetary, instant)
dimensions: main category = Cash balances at central banks
... rest of the dimensional vector ...
+
context: entity (LEI), reference date
value/unit: amount in EUR, decimals
|
v
one XBRL fact = one data point in the submission3 questions. Druk op 1 tot 4 om te kiezen, Enter om te bevestigen.
Door deze quiz af te ronden, markeer je deze sectie als voltooid.
Tip: kies met 1 tot en met 4 en bevestig met Enter.