fdic-ncua-health-rollup
Screen banks and credit unions for financial stress using FDIC/NCUA data, including capital ratios, CRE concentration, deposit runoff, and QoQ deltas to flag at-risk institutions.
Instructions
Bank & Credit-Union Financial Health API — FDIC/NCUA QoQ. Bank financial-stress screen on keyless FDIC data: capital ratios, CRE concentration (2006 guidance two-prong test), deposit runoff, ROA/ROE/NIM and asset quality per institution, with quarter-over-quarter deltas, peer-percentile scoring and health flags (deposit outflow, low ROA, rising NPL). Reads live from the official government source. COST AND SIDE EFFECTS: read-only with respect to the government source — it never writes to any external system — but each call starts a metered run on YOUR Apify account, billed $0.008 per result ($8 per 1,000). Lower on paid Apify plans, down to $2.40 per 1,000. Nothing is charged when a run fails. Store page: https://apify.com/malonestar/fdic-ncua-health-rollup
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| mode | No | Every mode returns the SAME full field set — capital ratios, uninsured deposits, unrealized losses, CRE concentration, credit quality and quarter-over-quarter deltas. Mode changes the ordering only. snapshot = largest institutions first. delta = biggest quarter-over-quarter deposit move first (the headline run-risk signal). score = highest peer asset percentile first. stress = most health flags first, the triage view. Example: "stress". Applied by default if omitted: "snapshot". | |
| state | No | US state to scope the cohort, e.g. CA, TX, NY. Strongly recommended: it focuses the run and makes peer percentiles state-level. Empty = the entire country (slower; national peer scoring). Example: "TX". | |
| benchmark | No | Adds asset-weighted benchmark ratios and this bank's distance from them, in percentage points: benchmark_uninsured_deposit_ratio, benchmark_cre_to_tier1_pct, benchmark_unrealized_loss_to_equity_pct plus uninsured_vs_benchmark_pts, cre_vs_benchmark_pts, unrealized_vs_benchmark_pts. 'national' compares against all ~4,350 FDIC-insured banks; 'state' against the banks in your state. Costs exactly ONE extra request thanks to server-side aggregation — not a second full download. 'none' skips it. Example: "national". | |
| creGrowth | No | The 2006 interagency CRE guidance is TWO tests: construction >= 100% of capital, OR (CRE >= 300% of capital AND CRE grew >= 50% over 36 months). Leaving this on fetches the quarter from 12 quarters ago — one extra request — and fills cre_growth_36m_pct, cre_total_loans_36m_ago, cre_baseline_date and cre_guidance_prong, plus the cre_guidance_both_prongs flag. Turn it off to skip that request; the level tests still run. Example: true. | |
| maxAssets | No | Only include institutions with at most this many total assets, in thousands of dollars. 0 = no ceiling. Combine with minAssets to score within an asset-size peer band (e.g. community banks $250M–$1B). Applied by default if omitted: 0. | |
| minAssets | No | Only include institutions with at least this many total assets, in thousands of dollars (FDIC reports assets in $000s, so 1000000 = $1B). Use with maxAssets to build a peer band. 0 = no floor. Applied by default if omitted: 0. | |
| peerBasis | No | What counts as a 'peer' when computing peer_asset_percentile, peer_roa_percentile, peer_cre_percentile and peer_uninsured_percentile. 'cohort' scores against everything you pulled (a state cohort mixes a $27M agricultural bank with a $200B trust bank, so the percentile means little). 'business_line' uses the FDIC SPECGRP business-model peer group — the cut a bank examiner uses. 'asset_band' uses FFIEC-style size bands. 'community_bank' splits on the FDIC community-bank research flag. Groups with fewer than 5 institutions fall back to the full cohort rather than ranking a bank against two neighbours. Example: "business_line". Applied by default if omitted: "cohort". | |
| maxResults | No | Maximum number of institution-health records to return after filtering, scoring, and ranking. This is your cost cap: one record = one billable result. 500 covers a full mid-size state; TX has ~380 banks, CA ~180. Example: 500. Applied by default if omitted: 1000. | |
| priorItems | No | Optional. In delta mode, an array of institution rows from a previous run (each needs id, total_assets, total_deposits) to diff the current quarter against, instead of auto-fetching the prior quarter. Lets you compare two arbitrary runs. Applied by default if omitted: []. | |
| institutionType | No | Which institutions to include. 'bank' = FDIC-insured banks, fully supported, the only option that returns data today. 'credit_union' = NCUA — NOT AVAILABLE YET (ships in v1.3); selecting it alone fails the run immediately and bills nothing, rather than quietly handing back bank data. 'all' = runs the bank half now and picks up credit unions automatically the moment v1.3 lands. Example: "bank". |