← Credit Union Merger Forecast DataVisuals AI Labs

Model Card · NCUA 5300

Credit Union Merger & Liquidation Risk

A quarterly model that ranks every federally insured U.S. credit union by its risk of disappearing — merging or liquidating — within roughly the next year, learned from seven years of NCUA 5300 Call Report filings.

0.30
PR-AUC (validation)
6.5× base rate
0.80
ROC-AUC
~1 yr
Prediction horizon
4 quarters
4,336
CUs scored
as of Q1 2026

Scores are a relative risk ranking, not calibrated probabilities. Read the limitations before acting on an individual score.

01What it predicts

For each active credit union in a given quarter Q, the model estimates the likelihood that the institution stops filing within the next four quarters — the observable signature of a merger or liquidation. The label starts from disappearance in the Call Report corpus — a CU that files in Q but is absent thereafter — and is then cross-checked against NCUA's official merger and liquidation records (see limitations). Confirmed re-charters (a CU continuing under a new number) are removed as false positives.

Output is a score in [0, 1], bucketed into three bands by rank within the scored population:

▲ High top 10% · 434 CUs ◆ Medium next 20% · 867 CUs ● Low remaining 70% · 3,035 CUs

02How it works

The pipeline is a deterministic sequence, rebuilt each quarter as new filings publish:

  1. 01
    IngestQuarterly NCUA 5300 ZIPs (2019-Q1 → present) parsed into a DuckDB time series. Account codes are resolved from AcctDesc.txt at ingest, never hard-coded.
  2. 02
    LabelAny CU whose last filing precedes the latest quarter is marked disappeared_after = its last-seen date.
  3. 03
    Engineer featuresPer-quarter capital, profitability and delinquency ratios, plus trailing 4- and 8-quarter changes, computed as SQL window functions.
  4. 04
    TrainA gradient-boosted classifier (HistGradientBoosting) on a time-based split — older quarters train, recent quarters validate — with balanced class weights for the rare event.
  5. 05
    ScoreThe latest quarter's active CUs are scored and written back for the dashboard and downstream consumers.

03What drives a score

Permutation importance on the validation set (contribution to PR-AUC). The signal is economically coherent: small, shrinking, thinly-capitalized, unprofitable credit unions are the ones that get absorbed.

FeatureReads asImportance
roaReturn on assets — profitability
log_assetsInstitution size
net_worth_ratioCapital cushion vs. assets
total_membersMembership scale
assets_chg_4qAsset growth, trailing year
members_chg_4qMembership trend, trailing year

04How well it performs

Disappearance is rare — about 4.7% of CU-quarters in the validation window. Against that base rate, PR-AUC of 0.30 is a 6.5× lift, and ROC-AUC of 0.795 indicates solid rank-ordering. Accuracy is not reported: a model that called everything "safe" would score 95% and be useless.

At a working threshold that flags the riskiest 5% of institutions, the model captures ~32% of actual disappearances at ~30% precision — more than six times better than chance, and a practical shortlist for an examiner or analyst to review. Cross-checking labels against NCUA records (v2) lifted PR-AUC from 0.29 and recall from ~30% by removing false-positive re-charters from training.

05Data lineage

Source: NCUA 5300 Call Report quarterly files, 2019-Q1 through 2026-Q1 — 29 quarters, ~142,000 CU-quarter observations across 5,569 distinct credit unions. Core metrics (assets, members, shares) resolved for 100% of filers; the net-worth ratio uses NCUA's reported figure (ACCT_998) where present. Labels are reconciled against the NCUA Insurance Report of Activity (approved mergers, with merging and continuing charter numbers) and the conservatorships & liquidations list. Everything is reproducible from raw files via the pipeline.

Labels assume corpus completeness through the latest ingested quarter — a "negative" is trusted only where the full forward horizon is observed, and disappearance detection relies on every published quarter being present.

Limitations & responsible use

Request access to the tool →