{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp11_mechanism_context_external20_20260804","research_id":"eoa_inverse_innovation_exp11_external_scrutiny_20260804","cell_id":"layer_decay_and_expiration_management__economics_finance","opaque_id":"layer_decay_and_expiration_management__economics_finance__B","search_lanes":{"direct_problem":{"queries":["IFRS 9 management overlays post model adjustments review release governance stale overlays duplicated conservatism","valuation adjustments overlays reserves periodic review release superseded banking model risk","management overlay release review IFRS 9 governance"],"source_ids":["SRC3","SRC4","SRC5","SRC8"],"no_result_note":null},"closest_prior_art":{"queries":["post-model adjustment sunset overlay","temporary management overlays ECL governance removal","PRA SS1/23 post-model adjustments expert overlays governance temporary review"],"source_ids":["SRC1","SRC5","SRC6"],"no_result_note":null},"historical_terminology":{"queries":["CECL qualitative factor adjustments management overlays governance review release stale reserves","allowance loan losses qualitative factor adjustments Q factors governance documentation sunset","pre IFRS 9 management judgment overlays loan loss provisions qualitative adjustments"],"source_ids":["SRC2","SRC8"],"no_result_note":null},"products_practices_standards":{"queries":["PRA post model adjustments overlays governance review removal IFRS 9","ECB management overlays model overlays provisioning review expiry","Federal Reserve model overlays capital planning independent challenge governance"],"source_ids":["SRC1","SRC2","SRC3","SRC6"],"no_result_note":null},"non_english_regional":{"queries":["IFRS 9 Management Overlay Auflösung Überprüfung Modellanpassung","IFRS 9 ajustes post modelo provisiones revisión liberación","IFRS 9 ajustements post-modèle gouvernance revue suppression","PSAK 71 management overlay penyesuaian model tinjauan penghapusan"],"source_ids":["SRC7"],"no_result_note":null},"composition_subproblems":{"queries":["model overlay inventory owner premise supersession dependency review retirement audit trail","post model adjustment backtesting scenario replay sampling release criteria","credit loss reserve duplicate layering periodic independent validation"],"source_ids":["SRC1","SRC3","SRC6","SRC8"],"no_result_note":null}},"sources":[{"source_id":"SRC1","title":"Model risk management principles for banks — Supervisory Statement SS1/23","url":"https://www.bankofengland.co.uk/-/media/boe/files/prudential-regulation/supervisory-statement/2023/ss123.pdf","publisher":"Bank of England, Prudential Regulation Authority","date_or_year":"2023, amended 2026","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Firms should maintain a documented, firm-wide PMA process covering calculation, completeness, review, and reduction or removal.","Every PMA should receive materiality-proportionate independent review of continued relevance, assumptions, inputs, outputs, and root cause.","Documentation should specify removal criteria and prolonged-use triggers, while material PMAs require approval by an appropriate senior authority."]},{"source_id":"SRC2","title":"Federal Reserve Guidance on Supervisory Assessment of Capital Planning and Positions for Large and Noncomplex Firms","url":"https://www.federalreserve.gov/frrs/guidance/federal-reserve-guidance-on-supervisory-assessment-of-capital-planning-and-positions-for-large-and-noncomplex-firms.htm","publisher":"Board of Governors of the Federal Reserve System","date_or_year":"Effective 2016","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Model overlays should be documented, systematic, transparent, and reported both with and without their effects.","Overlay challenge should be organizationally independent from overlay setting.","Material overlays require approval and scrutiny commensurate with their impact, potentially including a board committee, and sensitivity analysis is expected for larger firms."]},{"source_id":"SRC3","title":"Overlays and in-model adjustments: identifying best practices for capturing novel risks","url":"https://www.bankingsupervision.europa.eu/press/blog/2023/html/ssm.blog230526~29af0452d6.sl.html","publisher":"European Central Bank Banking Supervision","date_or_year":"2023","source_type":"OFFICIAL_GUIDANCE","language":"English with regional-language site navigation","claims_supported":["An ECB targeted review of 51 banks found overlays widely used and comprising a material, highly variable share of performing-loan provisions.","Observed practices include scenario simulation, sensitivity analysis, representative client sampling, and reporting of significant overlays.","Weakly governed or aggregated overlays can obscure distinct risks, while overly extensive reliance correlated with lower provision coverage and stage-2 ratios."]},{"source_id":"SRC4","title":"Post-implementation Review of IFRS 9—Impairment: Project Summary and Feedback Statement","url":"https://www.ifrs.org/content/dam/ifrs/project/pir-9-impairment/pir-ifrs9-projectsummary-feedbackstatement.pdf","publisher":"IFRS Foundation, International Accounting Standards Board","date_or_year":"2024","source_type":"OFFICIAL_STANDARD","language":"English","claims_supported":["Stakeholders reported increasing use of PMAs and management overlays for emerging risks.","The review found diversity in recognition and repurposing practices and a general lack of transparency about rationale and magnitude.","The IASB concluded that PMAs may be necessary when relevant information cannot be incorporated into statistical models, but their use remains judgmental."]},{"source_id":"SRC5","title":"Entering a “new normal” for IFRS 9 ECL measurement","url":"https://www2.deloitte.com/content/dam/Deloitte/gr/Documents/risk/gr_IFRS_9_ECL_measurement_in_the_Covid19_era_onwards_noexpa.pdf","publisher":"Deloitte IFRS 9 Working Group","date_or_year":"2021","source_type":"TRADE_PROFESSIONAL","language":"English","claims_supported":["Leading-bank practice links each overlay to a model limitation, independently challenges it, and establishes criteria for reduction or change.","Banks are described as actively managing a portfolio of PMAs, with scope, duration, and unwinding conditions under regular review.","Long-term overreliance on PMAs creates control costs and regulatory risk, while short-term overlays need an exit strategy and root-cause remediation."]},{"source_id":"SRC6","title":"Standard Chartered PLC — Additional Financial Information Part 1","url":"https://www.investegate.co.uk/announcement/rns/standard-chartered--stan/additional-financial-information-part-1/9443522","publisher":"Standard Chartered PLC, disseminated by Investegate","date_or_year":"2026 reporting on year-end 2025","source_type":"OTHER","language":"English","claims_supported":["Standard Chartered publicly describes PMA identification, calculation, independent validation, committee approval, and remediation tracking.","Model-performance PMAs are removed when estimates return within monitoring or validation thresholds.","Judgmental adjustments undergo quarterly review and reapproval and are released when their risks cease to be relevant, demonstrating routine lifecycle retirement and identifiable committees."]},{"source_id":"SRC7","title":"Zwischenbericht zum 30. Juni 2024","url":"https://ircenter.handelsblatt.com/download/companies/DeutscheBank/Quarterly%20Reports/DE0005140008-Q2-2024-EQ-D-00.pdf","publisher":"Deutsche Bank AG, hosted by Handelsblatt IR Center","date_or_year":"2024","source_type":"OTHER","language":"German","claims_supported":["Deutsche Bank reports regular review of IFRS 9 methods, model changes, inaccuracies, uncertainties, assumptions, and the need for overlays.","The bank reduced a management overlay after implementing a model refinement that captured the relevant ECL effect.","The disclosure demonstrates regional first-party practice in which overlays are added, reduced, reviewed, and linked to planned model remediation."]},{"source_id":"SRC8","title":"Policies and Internal Controls — Allowance for Credit Losses","url":"https://publishedguides.ncua.gov/examiner/content/examinersguide/AllowanceCreditLoss/RiskManagement/PoliciesInternalControls.htm","publisher":"National Credit Union Administration","date_or_year":"Updated 2023","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Credit-loss controls should verify each reporting period that layering has not occurred.","The NCUA defines layering as recording more than one amount for the same probable loan loss, directly establishing duplicated reserve conservatism as a recognized control problem.","Periodic evaluation should define responsibilities, inform the board, preserve documentation, and receive independent validation."]}],"problem_evidence":{"status":"PARTLY_SUPPORTED","finding":"Public evidence supports the mechanism but not the proposal's frozen prevalence and materiality thresholds. Regulators recognize duplicate reserve layering as inappropriate; the IASB reports increasing, diversely repurposed and insufficiently transparent PMA use; and practitioners warn that long-lived overlay reliance creates cost and regulatory risk. No retained source measures ownerless, explicitly superseded, or fully duplicated overlays in the proposed eight-quarter population.","source_ids":["SRC4","SRC5","SRC8"],"uncertainty":"Public disclosures generally report aggregate overlay governance and balances rather than overlay-level owners, premises, dependencies, review hours, or duplication. The proposed 10% prevalence and materiality gate therefore remains unverified."},"adopter_evidence":{"status":"SUPPORTED","finding":"The authorizers are identifiable: the PRA specifies senior management, risk or audit committees for material PMAs; Federal Reserve guidance assigns material decisions to senior management or board committees; and Standard Chartered identifies independent validation, its Credit Model Assessment Committee, and IFRS 9 Impairment Committee.","source_ids":["SRC1","SRC2","SRC6"],"uncertainty":"Exact mandates and compliance or records-retention approvals remain institution- and jurisdiction-specific."},"implementation_evidence":{"status":"PARTLY_SUPPORTED","finding":"The lifecycle core is implemented in public practice: independent continued-relevance review, documented reduction/removal criteria, prolonged-use triggers, quarterly reapproval, threshold-based removal, risk-expiry release, remediation tracking, and observed reduction after model refinement. The exact frozen prevalence audit, five-state ledger, dependency-clearance dossier, one-business-day restoration drill, and paired 90-day shadow noninferiority test were not found.","source_ids":["SRC1","SRC5","SRC6","SRC7"],"uncertainty":"Public sources cannot show whether internal implementations include ownerless-overlay detection, supersession links, restoration logs, or the proposal's quantitative pass thresholds."},"prior_art":{"disposition":"ESTABLISHED_PRACTICE","closest_analogues":[{"name":"PRA SS1/23 Principle 5.1 PMA governance and retirement","source_ids":["SRC1"],"same_problem":true,"same_causal_lever":true,"overlap":"Requires a firm-wide PMA process, independent review of continued relevance, documented reduction/removal criteria, prolonged-use triggers, root-cause remediation, transparency, materiality scaling, and approval by an appropriate authority.","remaining_difference":"Does not prescribe the proposal's frozen prevalence gate, overlay-level stale taxonomy, paired shadow arm, one-day restoration drill, or quantitative safety-and-efficiency thresholds."},{"name":"Standard Chartered PMA and judgmental-adjustment governance","source_ids":["SRC6"],"same_problem":true,"same_causal_lever":true,"overlap":"Operationally tracks PMAs and remediation, removes model-performance PMAs when thresholds normalize, and quarterly reviews, reapproves, and releases judgmental overlays when risks are no longer relevant.","remaining_difference":"The public disclosure does not document a controlled shadow-retirement experiment, review-hour endpoint, signed rollback procedure, or census/stratified sampling rule."},{"name":"Deloitte active PMA-portfolio lifecycle practice","source_ids":["SRC5"],"same_problem":true,"same_causal_lever":true,"overlap":"Describes leading banks as actively managing PMA portfolios through independent challenge, explicit scope and duration, regular review, unwinding conditions, exit strategies, and model remediation.","remaining_difference":"It is professional guidance rather than evidence that the proposal's exact audit and noninferiority protocol has been implemented."},{"name":"NCUA anti-layering control","source_ids":["SRC8"],"same_problem":true,"same_causal_lever":false,"overlap":"Recognizes duplicate reserving for the same probable loss as inappropriate and requires periodic controls and independent validation to detect it.","remaining_difference":"It addresses duplicate loss amounts, not a complete overlay lifecycle with premise expiry, dependency clearance, archival restoration, and shadow testing."}],"contrastive_claim_remaining":"Beyond established overlay lifecycle governance, a frozen overlay-level audit followed by paired shadow retirement may identify an eligible stale subset and reduce review hours by at least 20% without any additional control breach or more than 5% relative degradation in valuation error or stress-tail-loss coverage, while demonstrating restoration within one business day.","contrastive_claim_falsifier":"The claim fails if the audit does not open the prespecified prevalence-and-materiality gate, no eligible subset exists, a retained source or documented internal rival already implements materially the same frozen paired protocol, or the pilot causes any extra breach, exceeds either 5% risk-degradation limit, misses the 20% review-hour reduction, or fails the restoration drill.","confidence":"HIGH","search_limitations":"The bounded eight-source search emphasizes public UK, EU, US, and German banking materials. It cannot inspect confidential overlay ledgers, internal audit workpapers, vendor implementations, all jurisdictions, patents, or every institution's committee procedures. Public evidence establishes routine lifecycle governance but not the prevalence of stale overlays or implementation of the exact experimental protocol."},"researchability_gates":{"externally_supported_problem":{"status":"PASS","rationale":"Independent official and professional sources support duplicate reserve layering, opaque or repurposed overlays, and the need to control prolonged reliance and unwinding, although scoped prevalence remains for the audit to determine.","source_ids":["SRC4","SRC5","SRC8"]},"identifiable_adopter_or_authorizer":{"status":"PASS","rationale":"Risk, model, impairment, audit, and board committees are explicitly identified as approval and review authorities in supervisory guidance and bank practice.","source_ids":["SRC1","SRC2","SRC6"]},"distinct_testable_incremental_claim":{"status":"PASS","rationale":"The frozen prevalence gate and paired shadow-retirement test with explicit risk, breach, efficiency, and restoration thresholds remain distinct from the established lifecycle controls found.","source_ids":["SRC1","SRC5","SRC6"]},"bounded_next_evidence_step":{"status":"PASS","rationale":"A census or stratified 100-overlay audit followed only by historical replay and a 90-day shadow pilot is bounded, measurable, reversible, and does not require changing booked values.","source_ids":["SRC1","SRC2","SRC3"]},"no_unresolved_safety_or_authority_stop":{"status":"PASS","rationale":"Independent review, senior committee authorization, materiality-sensitive controls, scenario analysis, and retention of the active baseline make the audit and shadow pilot permissible in principle. Live retirement or evidence deletion would require separate institution-specific authorization and is outside the proposed first step.","source_ids":["SRC1","SRC2","SRC3","SRC6"]},"adequate_search_evidence":{"status":"PASS","rationale":"All six required lanes were searched adversarially. Exactly eight retained sources were opened, spanning seven publisher groups and including six official, standard-setting, or regulator-issued sources plus public institutional practice in English and German.","source_ids":["SRC1","SRC2","SRC3","SRC4","SRC5","SRC6","SRC7","SRC8"]}},"strict_success":false,"screen_survival":false,"remaining_research_value":"MODERATE","recommended_next_step":"Treat lifecycle review and retirement as established practice rather than a new control package. First benchmark the scoped institution's current process against PRA SS1/23 and the documented quarterly-release rival; then run only the frozen overlay-level prevalence audit. Proceed to the shadow pilot solely if the problem gate opens and the institution lacks an equivalent paired evaluation, preserving the no-live-change and restoration constraints.","world_novelty_boundary":"This bounded public search supports an ESTABLISHED_PRACTICE disposition for the lifecycle-governance core and leaves a narrower experimental-evaluation claim. It does not establish world novelty, patentability, freedom to operate, market size, comprehensive regulatory compliance, or realized impact."}