Derived table · PPP
PPP per-lender totals
Loan count, approved dollars, average loan, reported jobs, and share of the program for each of 4,680 PPP originating lenders, with the Womply and Blueacorn aggregator networks consolidated into rows of their own.
Download CSV 358 KBData dictionary CSVAt a glancePreviewColumnsHow to useBefore you quote a row
At a glance
The file as published folds nine partner lenders into two 'aggregator network' rows, so those nine lenders have no row of their own in it and the two network rows are not values of SBA's originating-lender field. The lender pages and rankings now rank every originating lender on its own and show the networks unranked beside them; a table rebuilt that way follows in the next data release.
Measured on the served file on 2026-09-12; the checksum above is the file these numbers describe.
lender · uniqueloans column of ppp-lender-totals.csv for the ten rows with the largest values; the two aggregator-network rows are included, and each holds the whole PPP book of its partner lenders.The numbers behind this chart
| Lender | Loans |
|---|---|
| Womply (aggregator network) | 1.4M |
| Blueacorn (aggregator network) | 840K |
| Bank of America, National Association | 491K |
| Cross River Bank | 479K |
| JPMorgan Chase Bank, National Association | 453K |
| Wells Fargo Bank, National Association | 281K |
| Customers Bank | 250K |
| U.S. Bank, National Association | 175K |
| Itria Ventures LLC | 173K |
| Celtic Bank Corporation | 167K |
dollars column of ppp-lender-totals.csv for the ten rows with the largest values, in dollars.The numbers behind this chart
| Lender | Approved dollars |
|---|---|
| JPMorgan Chase Bank, National Association | $44.1B |
| Bank of America, National Association | $34.4B |
| Womply (aggregator network) | $24.2B |
| PNC Bank, National Association | $17.3B |
| Truist Bank | $16.2B |
| Wells Fargo Bank, National Association | $13.8B |
| Manufacturers and Traders Trust Company | $13.3B |
| Blueacorn (aggregator network) | $13.1B |
| Cross River Bank | $12.9B |
| TD Bank, National Association | $12.3B |
What it is
One row per originating lender in the SBA's PPP FOIA file, 2024-09-30 vintage. The rows sum to 11,468,171 loans and $792.6 billion in current approved amounts, 39 rows fewer than the 11,468,210 loans in the source release (see Before you quote a row). First-draw and second-draw loans are counted together.
Two rows are not lenders in the SBA file: Womply (aggregator network) and Blueacorn (aggregator network). The two fintech aggregators took applications and passed them to partner lenders, and the SBA file books each of those loans to the partner. Here the partner lenders' volume is consolidated into the two network rows so the aggregators can be compared with banks on the same footing. Ranked by loan count, the Womply network row is the largest row in the table at 1,413,323 loans, 12.3% of all PPP loans; the Blueacorn row is second at 840,139. The House Select Subcommittee's December 2022 staff report, held in this archive, puts the PPP loans funded through Womply at over 1.3 million, out of 2.58 million referred to lenders. Each network row is the whole PPP book of its partner lenders, including loans those lenders may have made without the platform, so it is an upper bound on platform-sourced volume, not a measured count of platform originations.
This archive's Womply count is built differently and stands at more than 1,293,428 loans: court-record figures for three partner lenders plus the 2021 loans of four more, set out on Womply loan count. The Womply row in this file is the whole-book figure described above.
How it was built
The thirteen SBA files were grouped on the OriginatingLender field, summing CurrentApprovalAmount and JobsReported. Lenders on the fixed Womply and Blueacorn partner lists were then folded into the two network rows. Averages and ratios are computed after consolidation. A partner lender's entire volume is attributed to its network, including any loans it made outside the aggregator channel.
Preview
| lender | loans | dollars | avg_loan | jobs | jobs_per_loan | dollars_per_job | pct_program_loans |
|---|---|---|---|---|---|---|---|
| JPMorgan Chase Bank, National Association | 452892 | 44111851716.69 | 97400.38 | 7100912.0 | 15.679 | 6212.14 | 3.9491 |
| Bank of America, National Association | 491034 | 34411968884.59 | 70080.62 | 4222159.0 | 8.5985 | 8150.33 | 4.2817 |
| Womply (aggregator network) | 1413323 | 24196074246.25 | 17119.99 | 2002818.0 | 1.4171 | 12081.01 | 12.3238 |
| PNC Bank, National Association | 119352 | 17321532877.14 | 145129.81 | 1755025.0 | 14.7046 | 9869.68 | 1.0407 |
| Truist Bank | 117956 | 16207232060.29 | 137400.66 | 1837734.0 | 15.5798 | 8819.14 | 1.0285 |
| Wells Fargo Bank, National Association | 280717 | 13777713764.69 | 49080.44 | 1747245.0 | 6.2242 | 7885.39 | 2.4478 |
| Manufacturers and Traders Trust Company | 90029 | 13310300403.42 | 147844.59 | 1413298.0 | 15.6983 | 9417.9 | 0.785 |
| Blueacorn (aggregator network) | 840139 | 13063325741.17 | 15549.01 | 862510.0 | 1.0266 | 15145.71 | 7.3258 |
| Cross River Bank | 478866 | 12894672292.93 | 26927.52 | 1438370.0 | 3.0037 | 8964.78 | 4.1756 |
| TD Bank, National Association | 133140 | 12266833153.27 | 92134.84 | 1321878.0 | 9.9285 | 9279.85 | 1.1609 |
First 10 of 4,680 rows · cells longer than 60 characters are cut with an ellipsis · shown as text, unformatted.
Files
| File | Rows | Size | SHA-256 | |
|---|---|---|---|---|
ppp-lender-totals.csvCSV, UTF-8, header row | 4,680 | 366,614 B | e673fe0cbb6c0b68c28791996eb9b2f1f030d772376df8c9f002ca2133812c20 | Download ↓ |
dictionary.csvdata dictionary, 8 rows | 8 | — | built with this page | Download ↓ |
Columns
| Column | Type, measured | Blank | Distinct | Range / values | Meaning |
|---|---|---|---|---|---|
lender | text | 0 | 4,680 | Lender name exactly as the SBA's OriginatingLender field spells it, or one of the two aggregator-network rows. | |
loans | integer | 0 | 1,835 | 1 – 1.4M · median 399 | Number of PPP loans, first and second draw combined. |
dollars | number | 0 | 4,680 | 2K – 44.1B · sum $792.6B · median $18.7M | Sum of CurrentApprovalAmount, in dollars. This is the approved amount after any modification, not the amount disbursed or forgiven. |
avg_loan | number | 0 | 4,680 | 2K – 1.4M · sum $304.6M · median $47K | dollars divided by loans. |
jobs | number | 0 | 3,531 | 1 – 7.1M · median 2K | Sum of JobsReported across the lender's loans, as borrowers stated it on the application. |
jobs_per_loan | number | 0 | 4,454 | 1 – 129 · median 6.16 | jobs divided by loans. |
dollars_per_job | number | 0 | 4,660 | 841.39 – 40K · sum $38.4M · median $8K | dollars divided by jobs; blank where no jobs were reported. |
pct_program_loans | number | 0 | 549 | 0 – 12.32 · median 0.00 | loans as a percentage of all 11,468,171 loans in the file. |
Type, blanks, distinct count, and range are measured on the file. The same table as CSV: dictionary.csv.
How to use it
Specific to this file, measured on it: keys, joins, encodings, units, and the values that will trip a naive count.
- Join on
lender, exact string. It is unique across the 4,680 rows (0 duplicates) and is the same spelling the two top-500 tables use, including the two aggregator-network rows. Lender names contain commas, so the field is quoted. - Dollars are dollars, not cents, with two decimals (
dollarsexample:44111851716.69).jobsis written as a float with a trailing.0(7100912.0); the top-500 tables write it as an integer. pct_program_loansis a percentage (0–100, not a fraction) and sums to 99.9968 across the file because of rounding to four decimals; theloanscolumn sums to 11,468,171, which is 39 fewer than the 11,468,210 loans in the 2024-09-30 source release (see Before you quote a row).- Line endings are CRLF, UTF-8, header row, no byte-order mark, no cell contains a line break. Nothing here needs a special CSV dialect.
How the lender pages use this table
Each lender page prints cells of this table and of its two top-500 companions, built from the SBA PPP FOIA loan-level release of 2024-09-30. The ranks on those pages are this archive's own ranking of the 4,688 named originating lenders in that release; a platform network is not one of them and carries no rank. The published top 500 by approved dollars and top 500 by loan count still carry the rank column of the 2026-07-10 data release, which folds the partner lenders of the two platform networks into two network rows, so their ranks differ from the ranks on the lender pages. The fee rows on the lender pages are modeled loan by loan from SBA's published fee schedule, because SBA never published what it paid each lender; they were recalculated from the loan-level file on 19 September 2026 and are ranked against every other originating lender in the fee ranking. Each lender page states the caveats below in one paragraph and links here for the rest.
Before you quote a row
- Population. The rows sum to 11,468,171 loans, 39 fewer than the 11,468,210 rows in the 2024-09-30 source release. 28 source rows carry a blank
OriginatingLender. The other 11 are the loans of one lender, $265,044.37 in all, whose name in the source file contains a byte that is not valid UTF-8; the reader that built this table drops those rows when it groups on the lender name (measured on the thirteen source files, 2026-09-21). This archive's own ranking of originating lenders counts them: 4,688 named lenders, 11,468,182 loans. The published table is left as it is and the difference stated here rather than adjusted away. - Jobs are self-reported.
JobsReportedis what the borrower wrote on the application; SBA did not verify it at origination, and for the smallest loans it is often 1 by construction. - Dollars are approved, not paid out. Cancelled and reduced loans are reflected in the current approved amount; disbursement and forgiveness are separate fields in the source file.
- Network rows absorb whole lenders. A bank on the Womply or Blueacorn partner list contributes all of its loans to the network row, not only the aggregator-sourced ones.
- Lender attribution is reconciled because the raw platform fields misattribute some volume.
PPP per-lender totals, built from SBA PPP FOIA loan-level data (2024-09-30 vintage). Pandemic Darlings, release pandemic-darlings-data-v1-2026-07-10, https://pandemicdarlings.com/data/ppp-lender-totals/, SHA-256 e673fe0cbb6c0b68c28791996eb9b2f1f030d772376df8c9f002ca2133812c20