TD Bank's originations fell 5% in 2025 while the whole market grew 10%, its third yearly fall in a row, from 32,964 in 2022 to 23,499.
Fewer loans, larger ones
Similar lenders (national banks making between a quarter and four times as many loans) grew a typical 6%; its change was lower than more than four in five of them. Yet lending by value rose from $8.28 billion to $8.93 billion, as its typical loan grew from $185,000 to $205,000.
A lender to existing homeowners, almost all conventional
Home improvement was 32.7% of its loans, refinancing 27.3% and purchase 25.2%. Conventional loans were 99.8%; FHA (insured by the Federal Housing Administration) was 0.2%, lower than about nineteen in twenty similar lenders. It turned down 22.1% of first-lien refinance applications on owner-occupied homes, the same as the typical peer.
Fewer LMI loans than most peers
Its LMI (low- and moderate-income) borrower share rose from 15.8% in 2020 to 20.0%, still lower than about three in four similar lenders. Loans in LMI areas were 12.8%, among the lowest tenth. Its jumbo loans (larger than the conforming limit), 15.8% against 9.0%, and big metro markets could mean more borrowers earn above LMI limits.
Seventh in New York and Philadelphia
It ranks seventh in both New York and Philadelphia, 15th in Boston and 19th in Miami, all served since before 1990. Its typical rate of 7.49% was higher than about four in five similar lenders; its many home-improvement loans, which can be priced differently, may explain part of that.
Questions the data can answer next
In which of its markets did originations fall most since 2022?
How are its rates spread across home-improvement, refinance and purchase loans?
Similar lenders: same lender type, originations between a quarter and four times the lender's.
Written from the 2025 HMDA data and checked figure by figure against it.
FIG. 1 — LOANS ORIGINATED, 2018–2025
23,499 loans in 2025, against a peak of 35,171 in 2020.
Originated means the loan was made: applications denied, withdrawn or not accepted are not counted. TD Bank took 46,634 applications in 2025.
32.7% of its 2025 loans were home improvement loans; 99.8% were Conventional.
Home improvement 32.7% · Refinance 27.3% · Purchase 25.2% · Other 14.7% Conventional 99.8% · FHA 0.2% · VA 0.0% · of 23,499 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
16 states, 114 metros, 417 counties
State
Loans
Share
New Jersey
6,695
28.5%
New York
4,324
18.4%
Florida
2,722
11.6%
Massachusetts
1,975
8.4%
Pennsylvania
1,803
7.7%
South Carolina
1,143
4.9%
Connecticut
1,002
4.3%
New Hampshire
997
4.2%
Maine
699
3.0%
North Carolina
555
2.4%
Largest metro areas, 2025
Loans originated
Metro area
Loans
New York-Jersey City-White Plains
3,694
Lakewood-New Brunswick
2,137
Nassau County-Suffolk County
1,437
Newark
1,286
Camden
1,210
Montgomery County-Bucks County-Chester County
937
Cambridge-Newton-Framingham
889
Philadelphia
582
West Palm Beach-Boca Raton-Delray Beach
461
Miami-Miami Beach-Kendall
439
Identifiers
LEI (Legal Entity Identifier, the global company code HMDA has used since 2018)
03D0JEWFDFUS0SEEKG89
2018–2025
RSSD (the Federal Reserve's identifier for banks and their parents)
497404
2004–2023
Respondent ID (the HMDA filer code before 2018)
318409
1990–1999
Respondent ID (the HMDA filer code before 2018)
124096
2000–2010
Respondent ID (the HMDA filer code before 2018)
9497404
2011–2017
Established 1 January 1852. Reported under HMDA 1990-2025.
Mergers, name changes, data editions and corrections: TD Bank's data.
The data and the software behind these pages
TD Bank's data
$215
TD Bank 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.