Comerica Bank is mainly a refinance and home-improvement lender rather than a home-purchase lender, and its originations fell 3% in 2025 while similar banks grew, to about half their 2020 level.
Little home-purchase lending
Refinancing made up 40.8% of Comerica's loans, home improvement 28.6% and other purposes 22.3%; purchases were 8.4%. It also bought 299 loans, against 1 for the typical peer. Its main state is Michigan, though its head office is in Dallas.
Fewer loans, more dollars
Among similar lenders (state member banks making between half and twice as many loans), its 3% fall compares with a 3% rise for the typical peer and 10% for the whole market. Yet the amount it lent rose from $852.4 million to $1.10 billion; one possibility is a shift towards larger loans, as its typical loan rose from $155,000 to $175,000. It made 6,773 loans in 2021; with refinancing its largest purpose, its count may move with rates, and its typical rate was 2.74% in 2020.
Lower price than most peers
Its typical rate, 5.99%, was among the lowest tenth of peers, whose typical rate was 6.75%. Higher-priced loans (a rate well above the average for similar loans) were 0.1%, against 1.1%.
More lending in majority-minority areas
Majority-minority areas took 32.3% of its loans, higher than nearly nine in ten peers, which may reflect its large markets, including Detroit-Warren-Dearborn and Los Angeles. Its shares to LMI (low- and moderate-income) borrowers, 24.1%, and LMI areas, 15.0%, were near the middle.
Questions the data can answer next
Which Michigan, Texas and California counties account for its refinance lending?
How have its loan sizes changed by purpose since 2021?
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
3,126 loans in 2025, against a peak of 6,773 in 2021.
Originated means the loan was made: applications denied, withdrawn or not accepted are not counted. Comerica Bank took 7,959 applications in 2025.
40.8% of its 2025 loans were refinance loans; 99.4% were Conventional.
Refinance 40.8% · Home improvement 28.6% · Other 22.3% · Purchase 8.4% Conventional 99.4% · FHA 0.4% · VA 0.2% · of 3,126 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
17 states, 80 metros, 172 counties
State
Loans
Share
Michigan
1,286
41.1%
California
975
31.2%
Texas
663
21.2%
Arizona
111
3.6%
Florida
68
2.2%
Colorado
4
0.1%
Georgia
4
0.1%
North Carolina
4
0.1%
Washington
3
0.1%
Illinois
1
0.0%
Largest metro areas, 2025
Loans originated
Metro area
Loans
Warren-Troy-Farmington Hills
657
Detroit-Dearborn-Livonia
293
Dallas-Plano-Irving
269
Houston-Pasadena-The Woodlands
238
Los Angeles-Long Beach-Glendale
217
Anaheim-Santa Ana-Irvine
145
San Diego-Chula Vista-Carlsbad
124
Phoenix-Mesa-Chandler
107
Oakland-Fremont-Berkeley
100
San Jose-Sunnyvale-Santa Clara
98
Identifiers
LEI (Legal Entity Identifier, the global company code HMDA has used since 2018)
70WY0ID1N53Q4254VH70
2018–2025
RSSD (the Federal Reserve's identifier for banks and their parents)
60143
2004–2023
Respondent ID (the HMDA filer code before 2018)
260143
1990–2010
Respondent ID (the HMDA filer code before 2018)
960143
2011–2017
Established 5 March 1849. Reported under HMDA 1990-2025.
Comerica 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.