An auditor opens a draft AFR and runs the calculation linkbase. The cash-flow statement, beautifully formatted in thousands, throws a single line back: `xbrl.5.2.5.2 calculation inconsistency: 1,234,000 vs 1,234,500`. The numbers tie. The footing in the PDF is correct. The validator still complains.
That is the calc-1.0 world in miniature. The original calculation rule from XBRL 2.1 section 5.2.5 is a strict summation-item check, blissfully unaware of how facts are actually consumed. Two facts for the same concept that differ by one in the last digit are treated as suspect. Rounding tolerances are computed per-fact from declared `decimals`, so a parent declared at thousands plus a child declared at units produces a permitted error band so loose that real defects hide inside it. Most "spurious" calc errors trace back to a single cause: `decimals` declared inconsistently across the calc tree.
Calculations 1.1 (the 2024 Recommendation, dated 2024-02-14) reasons in the terms of the Open Information Model. Duplicate facts that agree within declared precision no longer fire. Rounding tolerance is computed from the strictest `decimals` across operands, which removes the loose-band hiding place. Calc 1.1 is opt-in: the taxonomy declares the calculation-1.1 arcrole, and validators evaluate the new semantics only when that arcrole is present. Filings that "migrate" without updating the arcrole stay silently on 1.0 semantics. The check is one grep away, and worth doing.
ESEF accepts both versions; ESMA's filing rules are explicit about treating Calc 1.1 as a permitted upgrade. The SEC EDGAR profile keeps most filings on 1.0 because EFM already layers its own rounding rules (notably around `decimals="INF"` and consistent-duplicate handling) on top of the base check. Calc 1.1 is being incorporated selectively.
When you see `xbrl.5.2.5.2`, the first move is always the same. Open the calc tree. Look at `decimals` on the parent and on each child. Mismatched decimals are the cause four times out of five. Fix the decimals first; only escalate to a Calc 1.1 migration when the data is genuinely correct and the regulator allows the move.
Calculations 1.0 (XBRL 2.1 section 5.2.5) is the original summation-item check. It is brittle in the face of duplicate facts and rounding mismatches across decimals levels.
Calculations 1.1 (XBRL Recommendation 2024-02-14) reasons in OIM-semantic terms. Duplicate facts agreeing within declared precision no longer trip false errors. Rounding tolerance is computed from the strictest declared decimals across operands.
Calc 1.1 is opt-in. The taxonomy must declare the calculation-1.1 arcrole. ESEF accepts both; SEC EFM stays primarily on 1.0 with EFM-specific rounding rules layered on top.
When you see xbrl.5.2.5.2 calculation inconsistency, check decimals across the calc tree first. Mismatched decimals are the single most common cause of spurious errors.
Migrating a filing to Calc 1.1 without updating the taxonomy declaration silently keeps the filing on Calc 1.0 semantics. Always confirm the arcrole.
3 questions. Press 1 to 4 to select, Enter to confirm.
Completing this quiz marks the section done in your progress.
Tip: press 1 to 4 to select, Enter to confirm.