US SBA 7(a) small-business lending pulse (quarterly)
Quarterly US small-business credit-health intelligence from the SBA 7(a) FOIA loan file (keyless data.sba.gov CSV, approvals 2019-10-01 -> 2026-06-30, data as of 2026-06-30, U.S. federal public domain). Per (quarter, borrower state): net credit extended (loan count, approval volume, SBA-guaranteed volume, average loan/term/rate, jobs supported), vintage charge-off rate and loss rate with a 12-quarter seasoning flag, volume/count year-over-year momentum, an 8-quarter volume z-score, a documented 0-100 lending-health composite with per-quarter state ranks, expansion/contraction/stress flags, and top NAICS-2 sector concentration. Who joins this: a subscription business joins quarterly SMB-customer churn or expansion revenue to (quarter, state_code) to model how small-business credit availability moves its revenue; a sales team joins by state_code to prioritize territories where SMB lending is expanding or under stress. Caveat: the most recent quarter in any vintage can be incomplete (SBA posts data about a month after quarter-end), so contraction flags on the latest quarter deserve a second look. Source: U.S. Small Business Administration.
- Rows
- 1,447
- Columns
- 34
- Source cadence
- Quarterly
- Last refreshed
- Sep 29, 2026
- Theme
- finance
| Column | Type | Description |
|---|---|---|
| quarter | string | Calendar quarter of loan approval: first day of the quarter, ISO date. Primary join key with state_code. (unit: ISO date (quarter start)) |
| country_code | string | ISO alpha-3 country code (USA for every row). (unit: ISO 3166-1 alpha-3) |
| state_code | string | Borrower state, 2-letter USPS code (50 states + DC, GU, PR, VI). Primary join key with quarter. (unit: code) |
| state_label | string | Human-readable state/territory name. |
| loan_count | integer | Net loans approved in the quarter: real credit exposure only (statuses EXEMPT/active, PIF/paid in full, CHGOFF/charged off). Cancelled and undisbursed-commit approvals are excluded. (unit: count) |
| cancelled_count | integer | Approvals cancelled before disbursement (CANCLD); reported separately, excluded from volume. (unit: count) |
| committed_count | integer | Approved but undisbursed commitments (COMMIT); reported separately, excluded from volume. (unit: count) |
| approval_volume_usd | float | Sum of GrossApproval over net loans in the quarter: credit actually extended. (unit: USD) |
| sba_guaranteed_usd | float | Sum of SBAGuaranteedApproval over net loans: the SBA-guaranteed portion of the quarter's credit. (unit: USD) |
| guarantee_share_pct | float | SBA-guaranteed share of the quarter's volume, percent. (unit: percent) |
| avg_loan_usd | float | Mean GrossApproval over net loans in the quarter. (unit: USD) |
| avg_term_months | float | Mean TermInMonths over net loans in the quarter. (unit: months) |
| avg_interest_rate_pct | float | Mean InitialInterestRate over net loans in the quarter. (unit: percent) |
| jobs_supported | float | Sum of JobsSupported reported on net loans in the quarter. (unit: count) |
| active_share_pct | float | Share of net loans still active (EXEMPT), percent. (unit: percent) |
| paid_in_full_share_pct | float | Share of net loans paid in full (PIF), percent. (unit: percent) |
| chargeoff_count | integer | Net loans in the approval cohort currently charged off (CHGOFF). (unit: count) |
| chargeoff_rate | float | Vintage charge-off rate: CHGOFF / (EXEMPT + PIF + CHGOFF) for the approval quarter. Trustworthy only where seasoned_flag = 1. (unit: rate) |
| chargeoff_loss_rate | float | Vintage loss rate: sum(GrossChargeOffAmount) / sum(GrossApproval) for the cohort. (unit: rate) |
| seasoned_flag | integer | 1 when the approval quarter is at least 12 quarters before the file's AsOfDate: the vintage has had time to default, so its charge-off rate is meaningful. (unit: binary) |
| volume_yoy_pct | float | Year-over-year percent change in approval volume vs the same quarter one year earlier. Null for the first 4 quarters (warm-up) and when the base is zero. (unit: percent) |
| count_yoy_pct | float | Year-over-year percent change in loan count vs the same quarter one year earlier. Null for the first 4 quarters (warm-up) and when the base is zero. (unit: percent) |
| volume_8q_z | float | Z-score of the quarter's approval volume against the trailing 8 quarters (exclusive). Null until 8 quarters of history exist. (unit: z-score) |
| loss_rate_for_score | float | Loss rate entering the health score: the cohort's own chargeoff_rate when seasoned, else the state's most recent seasoned vintage rate carried forward. (unit: rate) |
| loss_component_estimated | integer | 1 when loss_rate_for_score is carried forward from an older seasoned vintage (unseasoned cohort). (unit: binary) |
| lending_health_score | float | Documented 0-100 composite: 100 * (0.45 * min-max of volume_yoy_pct clipped to [-50, +50] + 0.30 * min-max of count_yoy_pct clipped to [-50, +50] + 0.25 * (1 - min-max of loss_rate_for_score clipped to [0, 0.20])). High = credit expanding with low vintage losses. Null until YoY history exists. (unit: 0-100 score) |
| health_rank | integer | Per-quarter dense rank of lending_health_score across states, 1 = healthiest. (unit: rank) |
| expansion_flag | integer | 1 when volume_yoy_pct exceeds +15 (credit expanding). (unit: binary) |
| contraction_flag | integer | 1 when volume_yoy_pct is below -15 (credit contracting). (unit: binary) |
| stress_flag | integer | 1 when the vintage is seasoned and chargeoff_rate exceeds 0.10 (elevated credit losses). (unit: binary) |
| top_sector_naics2 | string | 2-digit NAICS code with the largest share of the cohort's approval volume. Null where no NAICS code is present. (unit: code) |
| top_sector_volume_share | float | Share of the cohort's approval volume in top_sector_naics2. (unit: rate) |
| as_of | string | The FOIA file's AsOfDate (YYYY-MM-DD); identical across rows and across runs on the same vintage. (unit: ISO date) |
| row_hash | string | Deterministic 16-hex row id: sha256('SBA7A|<state_code>|<quarter>'). (unit: hash) |
First 10 sample rows — a preview, not the complete dataset.
| quarter | country_code | state_code | state_label | loan_count | cancelled_count | committed_count | approval_volume_usd | sba_guaranteed_usd | guarantee_share_pct | avg_loan_usd | avg_term_months | avg_interest_rate_pct | jobs_supported | active_share_pct | paid_in_full_share_pct | chargeoff_count | chargeoff_rate | chargeoff_loss_rate | seasoned_flag | volume_yoy_pct | count_yoy_pct | volume_8q_z | loss_rate_for_score | loss_component_estimated | lending_health_score | health_rank | expansion_flag | contraction_flag | stress_flag | top_sector_naics2 | top_sector_volume_share | as_of | row_hash |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2019-10-01 | USA | AK | Alaska | 23 | 1 | 0 | 11,207,300 | 8,307,100 | 74.12 | 487,274 | 141.13 | 8.12 | 377 | 56.52 | 43.48 | 0 | 0 | 0 | 1 | — | — | — | 0 | 0 | — | — | — | — | 0 | 44 | 0.26 | 2026-06-30 | 03587bff5f47fac0 |
| 2019-10-01 | USA | AL | Alabama | 81 | 9 | 0 | 42,364,800 | 32,484,960 | 76.68 | 523,022 | 131.14 | 7.38 | 839 | 38.27 | 48.15 | 11 | 0.136 | 0.096 | 1 | — | — | — | 0.136 | 0 | — | — | — | — | 1 | 81 | 0.21 | 2026-06-30 | f187996d477e66dd |
| 2019-10-01 | USA | AR | Arkansas | 64 | 10 | 0 | 19,766,800 | 14,395,005 | 72.82 | 308,856 | 119.2 | 6.8 | 344 | 26.56 | 70.31 | 2 | 0.031 | 0.012 | 1 | — | — | — | 0.031 | 0 | — | — | — | — | 0 | 72 | 0.29 | 2026-06-30 | c23e880f752fd23f |
| 2019-10-01 | USA | AZ | Arizona | 210 | 31 | 0 | 114,624,000 | 85,198,390 | 74.33 | 545,829 | 152.31 | 7.39 | 2,104 | 36.67 | 60 | 7 | 0.033 | 0.009 | 1 | — | — | — | 0.033 | 0 | — | — | — | — | 0 | 62 | 0.17 | 2026-06-30 | d2d4d3a1b0a33204 |
| 2019-10-01 | USA | CA | California | 1,325 | 209 | 0 | 731,074,900 | 538,902,315 | 73.71 | 551,755 | 145.1 | 7.29 | 13,877 | 39.77 | 55.47 | 63 | 0.048 | 0.012 | 1 | — | — | — | 0.048 | 0 | — | — | — | — | 0 | 72 | 0.23 | 2026-06-30 | f1baea1539322e39 |
| 2019-10-01 | USA | CO | Colorado | 266 | 24 | 0 | 170,481,100 | 126,253,205 | 74.06 | 640,906 | 147.1 | 7.12 | 3,058 | 38.35 | 57.89 | 10 | 0.038 | 0.011 | 1 | — | — | — | 0.038 | 0 | — | — | — | — | 0 | 72 | 0.24 | 2026-06-30 | e89b24a7cb43ed35 |
| 2019-10-01 | USA | CT | Connecticut | 122 | 19 | 1 | 30,620,500 | 21,325,305 | 69.64 | 250,988 | 126.49 | 7.52 | 918 | 32.79 | 62.3 | 6 | 0.049 | 0.013 | 1 | — | — | — | 0.049 | 0 | — | — | — | — | 0 | 81 | 0.16 | 2026-06-30 | b8acffa627eb25cf |
| 2019-10-01 | USA | DC | District of Columbia | 31 | 5 | 0 | 8,990,500 | 6,443,925 | 71.67 | 290,016 | 122 | 7.9 | 226 | 38.71 | 58.06 | 1 | 0.032 | 0.054 | 1 | — | — | — | 0.032 | 0 | — | — | — | — | 0 | 72 | 0.28 | 2026-06-30 | 1a16b3eb88e59179 |
| 2019-10-01 | USA | DE | Delaware | 33 | 3 | 0 | 12,432,600 | 9,100,025 | 73.19 | 376,745 | 124.82 | 9.09 | 140 | 36.36 | 63.64 | 0 | 0 | 0 | 1 | — | — | — | 0 | 0 | — | — | — | — | 0 | 45 | 0.38 | 2026-06-30 | d0092d12bc2efb03 |
| 2019-10-01 | USA | FL | Florida | 680 | 78 | 0 | 359,714,000 | 267,688,615 | 74.42 | 528,991 | 139.63 | 7.85 | 6,649 | 39.12 | 52.79 | 55 | 0.081 | 0.02 | 1 | — | — | — | 0.081 | 0 | — | — | — | — | 0 | 72 | 0.14 | 2026-06-30 | 481acdf26abaf690 |
- Current
20260929T143608Z-8b97cee30f7d · sha256 8b97cee30f7d…
1,447 rows · first snapshot
Point any LLM at the metadata endpoint — the documentation above is machine-readable too (JSON-LD + Croissant).
curl "https://datazimuts.com/v1/datasets/sba_lending_pulse/us_sba_7a_lending_pulse_quarterly" | jq '{title, rows, columns_count, license}'import requests
ds = requests.get("https://datazimuts.com/v1/datasets/sba_lending_pulse/us_sba_7a_lending_pulse_quarterly").json()
print(ds["title"], ds["rows"], "rows")
# Sample rows for an LLM context window
for row in ds.get("sample_rows", [])[:5]:
print(row)API endpoint: https://datazimuts.com/v1/datasets/sba_lending_pulse/us_sba_7a_lending_pulse_quarterly
Tip: fetch /llms.txt for the full machine-readable catalog.
Where this data comes from and what was made from it. Other people's work shows as counts; only shared projects are named.
Cite this snapshot
Pinned to snapshot 20260929T143608Z-8b97cee30f7d and its content hash, so readers get exactly the data you used.
U.S. Small Business Administration. (2026). US SBA 7(a) small-business lending pulse (quarterly) [Data set, snapshot 20260929T143608Z-8b97cee30f7d, sha256 8b97cee30f7d]. Datazimuts. Retrieved 2026-09-30, from https://datazimuts.com/en/datasets/sba_lending_pulse/us_sba_7a_lending_pulse_quarterly?snapshot=20260929T143608Z-8b97cee30f7d
@misc{dz_sba_lending_pulse_us_sba_7a_lending_puls_8b97cee3,
title = {{US SBA 7(a) small-business lending pulse (quarterly)}},
author = {{U.S. Small Business Administration}},
year = {2026},
publisher = {Datazimuts},
howpublished = {\url{https://datazimuts.com/en/datasets/sba_lending_pulse/us_sba_7a_lending_pulse_quarterly?snapshot=20260929T143608Z-8b97cee30f7d}},
note = {Snapshot 20260929T143608Z-8b97cee30f7d, sha256 8b97cee30f7dd259a5debd62fb6483d93c6477172e15949c57cd9443e7a8cb8d; accessed 2026-09-30}
}Embed a table or a chart
Paste this into any page. The embed is pinned to the same snapshot, follows the reader's light or dark setting, and always shows the source, license and a link back.
<iframe src="https://datazimuts.com/embed/chart?dataset=sba_lending_pulse%2Fus_sba_7a_lending_pulse_quarterly&lang=en&theme=auto&snapshot=20260929T143608Z-8b97cee30f7d&x=quarter&y=loan_count&agg=avg" title="US SBA 7(a) small-business lending pulse (quarterly)" width="100%" height="380" style="border:0" loading="lazy"></iframe>
Ask about this dataset. Answers come only from its catalog record, measured profile and change history, and list the facts they used.