SBA Loan Outcome Check · method and verification

How was the SBA Loan Outcome Check checked, and what did the independent check find?

A separate reader recomputed the figures from the raw files and matched them with no mismatches. The method was written and hashed on 12 November 2026 before any figure was published.

What was fixed before any figure was published?

The method was frozen on 12 November 2026. The file's SHA-256 hash begins 8e3661619a8e. The tool refuses to build if the method file no longer matches its hash.

What does the independent check do?

  1. Every row lands in exactly one status or is counted as excluded (cancelled or committed), and the totals reconcile with the file's row count.
  2. A separate reader recomputes, for a random sample of 200 franchise and industry groups, the loan counts and charge-off counts straight from the raw CSV rows, and the data must equal them.
  3. A name search that matches a group returns the same record the page shows.
  4. A mutation control: removing a charged-off loan from a group must change its rate (the control must be able to fail).
  5. No borrower, lender or address field appears in any output file (checked by a column scan of every output).

What did it find?

MeasureResult
Groups sampled300
Seed99
Mismatches0

A separate Python reader recomputes loan counts, charge-off counts and seasoned loan counts for the sampled franchise and industry groups straight from the raw CSV rows.

The checking has so far led to no amendments to the method.

What is still not covered?

Only loans the SBA guaranteed are in the file; loans that did not go through the program are not. The FranchiseCode is assigned by the SBA and a brand can appear under more than one code, or be renamed. A loan to a franchisee is not a measure of the franchisor's performance. Industry codes reflect what was reported at approval. A charged-off loan can later be recovered in part; the gross charge-off amount is the amount charged off, not the final loss.

How can you check it yourself?

The source data is public: see the SBA 7(a) loan program. The dataset, the method and the code that rebuilds it are packaged for deposit on Zenodo. Open the SBA Loan Outcome Check to see the result.

Questions about how the check works

What does "frozen before the first publication" mean?

The method was written, hashed and saved before any figure from the data was published. A change after that is an amendment with a date, and the original hash is kept, so anyone can see what was decided and when.

Why does an independent reader matter?

A tool can agree with itself and still be wrong. A separate reader, written without the builder's code, recomputes the figures from the raw files; where the two disagree, one of them has a defect.

Does a clean check mean the tool is always right?

No. It means the numbers match an independent recomputation under the stated method. The limits section lists what the method does not cover.

Further reading: Zenodo, where the datasets are deposited; the FAIR data principles; the National Academies on reproducibility; the author's ORCID record; the SBA 7(a) loan program.

Compiled by Vincent Wesley Couey (ORCID 0009-0005-6869-308X).

☆ Save