Skip to content
FinLexicon Wiki
Directory

Ifrs9 Cecl Credit Provisioning

IFRS 9 vs CECL Procyclicality in Stress & Basel CET1 Regulatory Capital Transitional Phase-Ins

Empirical COVID-19 and CCAR/EBA stress tests reveal an inverse trade-off: US GAAP CECL hits CET1 harder on Day-1 loan growth, whereas IFRS 9 triggers steeper Stage 2 cliff volatility at the onset of sudden macro recessions.

By Arthur Pendleton, CPA, JDReview not recorded · Updated
IFRS 9 vs CECL Procyclicality in Stress & Basel CET1 Regulatory Capital Transitional Phase-Ins — FinLexicon Wiki 2026 quantitative & supervisory data card
FinLexicon Wiki (IFRS 9 ECL Staging, ASC 326 CECL, Fair Value Hierarchy & Hedge Accounting): Stage 2 cliffs vs CECL origination drag, IRB EL shortfall cushions, and Tier 2 1.25% / 0.60% caps.Source: FinLexicon Wiki Institutional Research DeskLicense: CC-BY-4.0

Which expected credit loss standard—IFRS 9 or US GAAP ASC 326 (CECL)—is more procyclical during a systemic recession, and how do prudential capital filters prevent sudden provision spikes from choking bank lending? Empirical evidence from Q1–Q2 2020 COVID-19 bank filings, Federal Reserve CCAR tests, and ECB/EBA stress tests uncovered a fascinating structural trade-off governed by BCBS d401 / CAP90 regulatory CET1 transitional add-backs.

Two Different Flavors of Procyclicality: CECL’s "Origination Tax" vs IFRS 9’s "Stage 2 Cliff"#

Compare the cycle dynamics of Bank of America (CECL) against BNP Paribas and Bank of Montreal (IFRS 9) across the four phases of a macro cycle, building on our IFRS 9 & ASC 326 CECL cornerstone guide:

  1. Late-Expansion / Recovery Loan Growth (CECL More Restrictive): Every new $1B 10-year loan originated during a recovery forces a US GAAP CECL bank to book full 10-year lifetime losses upfront (-150 bps) before earning a dime of interest, whereas an IFRS 9 bank books only 12-month Stage 1 ECL (-35 bps)—making CECL an heavier capital brake on new credit supply coming out of recessions!
  2. Onset of a Sudden Shock (IFRS 9 Steeper Percentage Jump): When unemployment forecasts spike in a single quarter (e.g., Q1 2020), CECL banks already held lifetime reserves on 100% of their book (so reserves scale by Delta_Lifetime_PD), whereas IFRS 9 banks experience a double multiplier: rising PDs push 15%–25% of the loan book across SICR Stage 2 thresholds, jumping those loans from 12-month to Lifetime ECL simultaneously!
  3. Perfect Foresight vs Realistic Imperfect Foresight (BIS Working Paper Findings): Academic BIS simulations prove that if economists possessed perfect foresight 2 years ahead of recessions, provisions would build smoothly in booms; in real life, forecast surprises arrive abruptly at Quarter 1 of the recession, causing both IFRS 9 and CECL to front-load CET1 hits at the trough unless buffered by capital filters!
Transmission pipeline from IFRS 9 and ASC 326 CECL provision expense through Tax DTAs and IRB/SA-CR filters into CET1 and Tier 2
Transmission pipeline from IFRS 9 and ASC 326 CECL provision expense through Tax DTAs and IRB/SA-CR filters into CET1 and Tier 2Verified Exhibit

Standardized (SA-CR) vs Internal Ratings-Based (IRB) Accounting-to-Capital Transmission#

Never assume $1.00 of new IFRS 9 or CECL provision reduces Basel III CET1 capital identically across banks! Basel CAP10 / CRE35 splits transmission cleanly by credit RWA approach:

  • Standardized Approach (SA-CR) Banks (Including ALL U.S. Banks on Risk-Based Collins Floor!): Specific/Stage 3 allowances reduce risk-weighted exposure net of provisions, while General Allowances (ASC 326 ACL in U.S. / General provisions) hit Retained Earnings (-100% CET1 after tax) and may only be added back into Tier 2 Capital up to 1.25% of Standardized Credit RWA! (Crucially, in the United States, ALL ASC 326 CECL Allowances for Credit Losses (ACL) count as Tier 2 general allowances up to the 1.25% RWA cap!)
  • Internal Ratings-Based (IRB) Banks (European / UK / Canadian G-SIBs): Compare Total Eligible Provisions against *1-Year Regulatory Expected Loss (`ELIRB = PDTTC LGD_Downturn * EAD`)**:
  • If Provisions < EL_IRB (Shortfall): 100% of the shortfall was already deducted from CET1—so raising IFRS 9 provisions up to EL_IRB is CET1-NEUTRAL (+$1 shortfall relief offsets -$1 retained earnings)!
  • If Provisions > EL_IRB (Excess): Excess hits CET1 (-$1 after tax) and enters Tier 2 Capital up to 0.60% of IRB Credit RWA!

Dynamic CET1 Transitional Phase-Ins: U.S. "25% CECL Scaling" vs EU CRR "Quick Fix" Art. 473a#

In March 2020, both transatlantic regulators enacted emergency 5-Year Dynamic CET1 Add-Back Filters (BCBS d498):

  • United States (12 CFR 217.301 — 2020 Interim Final Rule): Allowed banks to defer 100% of Day-1 CECL adoption impact PLUS 25% of the post-Day-1 increase in ACL during Years 1–2 (2020–2021) (calibrated because 25% approximated the excess of lifetime CECL build over 12-month Stage 1!), followed by a 3-year linear phase-out (75% in 2022, 50% in 2023, 25% in 2024, 0% fully phased-in from January 1, 2025!).
  • European Union (CRR Quick Fix Regulation 2020/873 Art. 473a) & UK PRA: Reset dynamic Stage 1 + Stage 2 non-credit-impaired ECL add-backs to 100% in 2020–2021, stepping down 75% (2022), 50% (2023), 25% (2024), to 0% fully phased-in from January 1, 2025!
Quantitative Distribution · IFRS 9 (EU / UK / Canada / APAC)
IFRS 9 (EU / UK / Canada / APAC)US GAAP ASC 326 CECL (United States)
01Steady-State Reserve LevelLower (~0.5%–0.…02New Loan Origination Capita…Mild (12-month …03Recession Onset Shock (Q1 C…Sharp Stage 2 S…04Transitional CET1 Add-Back …Expired (0% add…
Stress & Capital DimensionIFRS 9 (EU / UK / Canada / APAC)US GAAP ASC 326 CECL (United States)Post-2025 Prudential Reality
Steady-State Reserve LevelLower (~0.5%–0.9% performing)Higher (~1.3%–2.1% performing)U.S. banks carry thicker baseline ACL
New Loan Origination Capital DragMild (12-month Stage 1 ECL)Heavy (100% Lifetime Day-1 tax)Penalizes rapid post-crisis loan growth
Recession Onset Shock (Q1 Crisis)Sharp Stage 2 SICR Cliff (3x–8x jump)High dollar build on R&S macro revisionBoth front-load losses at recession start
IRB Expected Loss (EL) BufferYes — IRB EL shortfall absorbs hit!Collins Floor (SA-CR) prevents IRB offset!European IRB banks enjoy initial CET1 shield
Transitional CET1 Add-Back (2025+)Expired (0% add-back from Jan 1, 2025)Expired (0% add-back from Jan 1, 2025)100% live CET1 volatility globally

Why Federal Reserve CCAR Stress Tests Still Delayed Full CECL Incorporation#

Recognizing that embedding forward-looking CECL inside a 9-quarter severely adverse CCAR Stress Capital Buffer (`/wiki/stress-capital-buffer-scb-ccar-nine-quarter-cet1-depletion`) simulation risks double-penalizing Quarters 8–9 terminal reserves alongside Quarter-1 shocks, the Federal Reserve maintained specialized supervisory stress loss modeling rules.

Net CET1 capital impact (in basis points) of a +$2.0B recession provision shock across Standardized CECL vs IRB IFRS 9 with EL Shortfall
Net CET1 capital impact (in basis points) of a +$2.0B recession provision shock across Standardized CECL vs IRB IFRS 9 with EL ShortfallVerified Exhibit

Permanent Post-2025 Equilibrium Without Transitional Shields#

With both U.S. and EU 5-year COVID transitional filters expired at 0% in 2025, every dollar of Stage 2 migration or CECL R&S forecast revision flows straight into post-tax CET1 across the IFRS 9, CECL & Fair Value Accounting topic hub.

Formula or model from the articleNot independently verified
Delta_CET1_Standardized = -( 1 - Tax_Rate ) * Delta_ECL_General [ assuming Timing DTAs remain below 10% CET1 cap ]
  • DeltaCET1IRB = -( 1 - TaxRate ) * DeltaECLTotal + min( PreShockELShortfall , DeltaECL_Total )
  • Tier2AddBackEligible = min( ExcessGeneralACL , 0.0125 * RWACreditSA ) + min( ExcessIRBProv , 0.0060 * RWACreditIRB )
  • Worked IRB Shortfall Offset Proof ($100B IRB Credit RWA Bank; Pre-Shock IRB EL = $1,800M vs Pre-Shock IFRS 9 ECL = $1,100M => $700M CET1 Deduction):
  • Recession triggers +$1,200M IFRS 9 Stage 2 ECL surge (25% Tax Rate => Retained Earnings falls -$900M):
  • New IFRS 9 ECL = $2,300M > $1,800M IRB EL => Pre-existing $700M IRB Shortfall CET1 Deduction drops to $0 (+$700M CET1 relief!)
  • Net CET1 Change = -$900M (Retained Earnings) + $700M (Eliminated IRB Shortfall) = -$200M (-20 bps vs -90 bps without IRB offset!)

Frequently asked questions

Why did the 2020 U.S. regulatory rule add back exactly 25% of CECL reserve builds into CET1?

Federal Reserve, OCC, and FDIC empirical calibration estimated that during a sudden crisis, roughly 75% of a bank’s CECL reserve surge reflects near-term 12-month credit losses (comparable to legacy/IFRS 9 Stage 1–3), while 25% reflects the incremental multi-year lifetime tail unique to CECL!

What happens if a massive recession ECL charge creates Deferred Tax Assets (DTAs) that breach the 10% CET1 threshold?

Double CET1 amplifier! Provision expenses create temporary-difference DTAs (Tax_Rate * Delta_ECL); if cumulative timing DTAs exceed Basel CAP30’s 10%/15% CET1 threshold basket (or require a GAAP/IFRS valuation allowance due to operating losses!), the tax shield itself gets deducted 100% from CET1!

Does the BCBS d424 72.5% Output Floor erode the IRB Expected Loss cushion for European banks?

Watch how CRE99 / EU CRR III computes the 72.5% Output Floor: **total capital ratios compare CET1 (after IRB shortfall deductions) against max(RWA_modeled, 0.725 * RWA_SA)**—requiring careful tracking of general vs specific credit risk adjustments inside the Standardized output-floor denominator!

Can supervisors reactivate dynamic IFRS 9 / CECL CET1 phase-in filters in a future crisis?

Not automatically without legislation or rulemaking! Both 12 CFR 217.301 and EU CRR Art. 473a contained fixed statutory sunset dates ending December 31, 2024; re-introducing dynamic filters in a future shock would require fresh agency/parliamentary action (unlike automatic CCyB buffer releases!).

Publication record

Published
14 November 2026
Updated
15 November 2026
Author
Arthur Pendleton, CPA, JD
Review
Review not recorded
Automated fact check
Passed in the stored pipeline record
Source URLs
4 recorded

AI assistance recorded: anthropic/claude-opus-4.6. This field does not establish human review or source verification.

Academic & Institutional Attribution

Cite This Research & Embed Live Data Card

Open Attribution License (CC BY 4.0)

Standard academic & institutional research paper citation format.

Arthur Pendleton, CPA, JD. (2026, November 14). IFRS 9 vs CECL Procyclicality in Stress & Basel CET1 Regulatory Capital Transitional Phase-Ins. FinLexicon Wiki. https://finlexicon.org/wiki/ifrs-9-vs-cecl-procyclicality-cet1-regulatory-capital-phase-in
Explore Ifrs9 Cecl Credit Provisioning Hub (20) →

FinLexicon Wiki Data, Benchmarks & Contributor Records

IFRS 9 vs CECL Procyclicality & Basel CET1 Phase-In Rules | FinLexicon Wiki