HSBC Bank USA is a lender of large loans, with a typical loan of $945,000 against $205,000 at similar banks, and its loan count fell 16% in 2025 while similar lenders grew.
Large loans set it apart
Jumbo loans (larger than the conforming limit) were 41.7% of its lending against 1.7% at similar lenders (national banks making between half and twice as many loans), and its typical loan was the highest among them. So its 3,065 loans came to $3.82 billion, against a typical $751.6 million.
Fewer loans, and smaller ones
Originations fell 16%, while the typical similar lender and the whole market each grew 10%; its growth was lower than more than nine in ten peers. Its jumbo share also fell from 51.2% and its typical loan from $1.1 million. The decline could reflect a move away from its largest loans.
Lower share in San Jose
It ranks 6 of 459 lenders in San Jose-Sunnyvale-Santa Clara, CA, but its share fell to 4.0% from 6.8%, behind Wells Fargo Bank and Bank of America among others. This could reflect the same move away from its largest loans.
Few LMI borrowers, many majority-minority areas
Low- and moderate-income (LMI) borrowers took 2.4% of its loans against 27.2%, among the lowest tenth of similar lenders, and LMI neighbourhoods 8.2% against 14.7%. This could reflect loan sizes near its typical $945,000, which few LMI borrowers would take on. Majority-minority neighbourhoods took 48.5% against 8.7%, higher than more than nine in ten, which could reflect markets such as San Jose and Los Angeles.
Questions the data can answer next
How do its rates on jumbo and conforming loans compare with other large banks in San Jose and San Francisco?
Similar lenders: same lender type, originations between half and twice the lender's in the data year.
Written from the 2025 HMDA data and checked figure by figure against it.
FIG. 1 — LOANS ORIGINATED, 2018–2025
3,065 loans in 2025, against a peak of 7,377 in 2020.
Originated means the loan was made: applications denied, withdrawn or not accepted are not counted. HSBC Bank USA took 4,645 applications in 2025.
70.5% of its 2025 loans were purchase loans; 100.0% were Conventional.
Purchase 70.5% · Refinance 19.9% · Home improvement 6.4% · Other 3.3% Conventional 100.0% · of 3,065 loans. Purchase: buying a home. Refinance: replacing an existing mortgage, including cash-out. FHA: insured by the Federal Housing Administration. VA: guaranteed by the Department of Veterans Affairs. Conventional: neither.
SOURCE: FFIEC HMDA 2025, LOANS ORIGINATED
Where it lends, 2025
39 states, 130 metros, 227 counties
State
Loans
Share
California
1,526
49.8%
New York
538
17.6%
Washington
358
11.7%
New Jersey
144
4.7%
Florida
134
4.4%
Texas
60
2.0%
Connecticut
37
1.2%
Massachusetts
25
0.8%
North Carolina
24
0.8%
Illinois
22
0.7%
Largest metro areas, 2025
Loans originated
Metro area
Loans
San Jose-Sunnyvale-Santa Clara
786
New York-Jersey City-White Plains
480
Seattle-Bellevue-Kent
310
San Francisco-San Mateo-Redwood City
254
Oakland-Fremont-Berkeley
185
Los Angeles-Long Beach-Glendale
150
Miami-Miami Beach-Kendall
77
Nassau County-Suffolk County
71
Anaheim-Santa Ana-Irvine
55
Newark
53
Identifiers
LEI (Legal Entity Identifier, the global company code HMDA has used since 2018)
1IE8VN30JCEQV1H4R804
2018–2025
RSSD (the Federal Reserve's identifier for banks and their parents)
413208
2005–2023
Respondent ID (the HMDA filer code before 2018)
124522
2004–2010
Respondent ID (the HMDA filer code before 2018)
9413208
2011–2017
Established 1 July 2004. Reported under HMDA 2004-2025.
HSBC Bank USA 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.
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.
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.