Applications / Mortgage Analytics / Lenders / South Dakota / Citibank / Fair lending
Citibank
Sioux Falls, SD · National bank · OCC · $1.8 trillion assets · 15,812 loans in 2025
Citibank's data, $215Buy the data, $2,110Get Jenner, the SAS-compatible software
15,812
loans originated, 2025
33,966
applications, 2025
49
states and territories
0
gaps bigger than at similar lenders
How to read this section
For Black, Hispanic and Asian applicants, we look at how much more often Citibank turns them down than White applicants: the gap. Then we compare that gap with the gap at similar lenders, for similar applications: the same income, loan size, debt-to-income range, loan-to-value, loan type, kind of property and place.
"Bigger gap" means we are at least 99% sure the gap is bigger than at similar lenders. That is a reason to look closer, not proof of discrimination. The public data has no credit scores and none of the documents a lender reviews, and the kind of loans a lender makes can widen a gap for good reasons. Each result below gives the numbers and the explanations they allow.
Citibank turns down some groups of applicants more often than White applicants. It turned down 23% of Black purchase applicants in 2025, against 7% of White ones. Nearly every lender shows gaps like this, partly because the groups apply with different incomes, debts, loan sizes and properties. So the test does not ask whether there is a gap. It asks whether Citibank's gap is bigger than the gap at similar lenders, for similar applications. For every group and loan purpose the answer was no: no gap was clearly bigger (that is, bigger with at least 99% certainty).
FIG. 1 — GAPS AGAINST SIMILAR LENDERS
None of Citibank's gaps is clearly bigger than similar lenders'.
Each dot compares Citibank's gap with the gap at similar lenders: at 1x they are the same, at 2x its gap is twice as large. A gap is how much more likely the group is to be turned down than White applicants. The line around each dot shows the range the true value probably lies in (95%). Amber: groups where we are at least 99% sure the gap is bigger.
SOURCE: FFIEC HMDA 2025; JENNER ANALYTICS NATIONAL DENIAL MODEL
| Applicants | Turned down: group | Turned down: White | How sure the gap is bigger | Result |
|---|---|---|---|---|
| Hispanic purchase | 15% | 7% | 11.5% | Not clearly bigger |
| Hispanic refinance | 22% | 9% | 1.8% | Not clearly bigger |
| Black purchase | 23% | 7% | 28.5% | Not clearly bigger |
| Black refinance | 25% | 9% | 3.1% | Not clearly bigger |
| Asian purchase | 8% | 7% | 40.9% | Not clearly bigger |
| Asian refinance | 6% | 9% | 10.4% | Not clearly bigger |
"How sure the gap is bigger": how sure we are that the lender's gap is bigger than at similar lenders. We call it bigger at 99% or more.
How this is calculated: the fair-lending method.
The data and the software behind these pages
Citibank's data
$215
Citibank and its peers, 2018–2025: every application in the HMDA data, cleaned and corrected, as Parquet files, with the Jenner programs that made these pages.
Buy Citibank's data, $215Jenner Mortgage Analytics: all the data
$2,110
Every mortgage application reported under HMDA from 1990 to 2025, cleaned and corrected, every lender followed across name changes and mergers, with the market-share and fair-lending tables behind these pages and the Jenner programs that built them: the warehouse and the datamart, as Parquet files.
Buy the data, $2,110Jenner, the SAS-compatible software
See prices
Jenner runs programs written in the SAS language. It drew every chart on these pages and fitted the fair-lending model, PROC GLIMMIX on 3.5 million applications. Run the programs that come with the data, or your own.
Get JennerQuestions, or the data and Jenner together: [email protected]