It’s not uncommon for accounting and valuation practitioners assess ECL based on annual default studies from international rating agencies’ annual default study report (global data). To apply in ECL assessment of other receivables in Hong Kong or PRC, the benchmark is defensible— only if you calibrate. IFRS 9 requires ECL estimates to reflect reasonable and supportable information about past events, current conditions and forward-looking factors; the standard allows both internal and external sources of evidence. That means mapping an internally derived counterparty credit score to a well-understood global rating scale (Moody’s/S&P) can be acceptable provided you calibrate and validate that mapping to your portfolio and environment. (IFRS 9 points explicitly permit using internal and external ratings/statistics as inputs; the key is relevance and supportability.)

Why “lift-and-shift PDs from a report” is not enough. Merely assigning a Moody’s/S&P grade and plugging the agencies’ published PDs into ECL does not, by itself, satisfy IFRS 9. You still need to show (i) how the internal score ↔ rating linkage was built, (ii) how PDs were calibrated to your receivables mix, tenor, collateralization and jurisdictions, and (iii) that you considered forward-looking Hong Kong/PRC conditions. (Those are explicit IFRS 9 expectations.)

The Key Points to Note if Your PDs are Derived from International Credit Rating Agencies Annual Default Study

  • Scoring & linkage – You need to estimate counterparty credit scores using financial metrics, behavioral data (e.g., days-past-due), sector and size. You then need to map scores to Moody’s/S&P rating buckets using overlapping samples and statistical alignment so that mapped buckets produce observed default rates consistent with the agencies’ long-run experience.”
  • Calibration to local context. You need to re-scale those benchmark PDs using Hong Kong and Mainland indicators (sector cycles, SME concentration, payment behavior) and validate against local insolvency/default evidence.
  • Forward-looking overlays. You need incorporate near-term data relevant to Hong Kong/PRC so that the derived PD reflects better macroeconomic trends.

Valtech’s suggested enhancements (what we actually do)

  1. Location-specific baselines. Where available, we replace global long-run PD anchors with Hong Kong-specific and Mainland-specific studies/data, then keep the rating-linked structure for interpretability.
  2. Score-to-rating mapping with evidence. We use internal probability-of-default model and statistically map financial attributes of the target companies based on best available information.
  3. Receivable-specific calibration. For “other receivables,” we adjust PDs for exposure tenor, credit enhancements, intra-group vs. third-party status, industry, and payment performance. Where counterparties are PRC-based, we can apply a model based on Mainland data reflecting sectoral stresses assumptions.
  4. IFRS 9-consistent overlays & validation. Our assessment is based on forward-looking data, consistent with requirements IFRS 9, then back-test against observed delinquencies/write-offs and re-calibrate.