Commercial Mortgage Statistics
Last updated August 3, 2026. Compiled and maintained by RefiLoop. County recording data is current through August 3, 2026; bank regulatory data reflects the Q1 2026 Call Report cycle.
Most commercial mortgage statistics you will find online are recycled from the same handful of industry surveys, and are quoted without a date. This page is built the other way round. Every figure below comes from a primary public source we collect ourselves — quarterly bank Call Reports filed with federal regulators, SBA 7(a) loan approvals, and county deed recording indexes — and every figure carries the source and the period it covers directly beneath it.
Where a number is our own calculation on primary data rather than a published statistic, we say so and show the method. Where our data cannot support a claim — and there are several such places, which we flag rather than paper over — we say that too.
1. Bank CRE concentration
Commercial real estate concentration is the single most watched number in bank CRE supervision. Under the 2006 Interagency Guidance, a bank whose CRE loans exceed 300% of its total risk-based capital is expected to demonstrate heightened risk management. It is a screening threshold that draws supervisory attention, not a legal limit and not a judgment about any individual bank's condition.
Two precision notes, because both are easy to state wrongly. First, the guidance's second criterion pairs the 300% ratio with a CRE growth test of 50% or more over 36 months. The counts below apply the ratio threshold only, so read them as a screen — not as a count of banks meeting the full supervisory criterion. Second, a bank being above the threshold says nothing about its charter: we group banks by size and scope, and those labels are not charter types.
The ratio has a precise definition that is easy to get wrong: it counts construction and land development, multifamily, and non-owner-occupied non-farm non-residential loans, and it excludes owner-occupied commercial mortgages — the loans where the borrower operates a business out of the building. Including owner-occupied loans, as some calculations do, inflates the ratio substantially. Every figure below uses the regulatory definition, computed from each bank's own filing.
1.393 US banks — 9.2% of the 4,252 that filed a Call Report in Q1 2026 — hold commercial real estate loans above the 300% supervisory screening threshold.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
2.87 US banks carry CRE concentration above 400% of total risk-based capital, and 19 exceed 500%.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
3.The average CRE concentration ratio across all 4,242 reporting banks is 147% of risk-based capital, with a median of 132% — meaning the typical US bank sits at less than half the supervisory screening threshold.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
4.1,723 banks — 41% of all filers — hold CRE loans equal to less than 100% of their risk-based capital, the largest single group.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
5.The number of banks above the 300% threshold has fallen from 400 to 393 over the eight quarters from Q2 24 to Q1 26, and the number above 400% has fallen from 101 to 87 — a 14% decline in the count of the most concentrated group. This is a count of banks, not a measure of exposure: aggregate CRE balances across reporting banks grew over the same period, so the concentrated tail is thinning while total CRE lending is not.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q2 24–Q1 26.
6.1,621 banks — 39% of all filers — reduced their commercial real estate balances quarter over quarter in Q1 2026, while the median bank grew its CRE book 1.1%.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
7.Banks above the 300% threshold are geographically clustered: five states account for 142 of the 393 — 36% — while the remainder is spread across the rest of the country.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
8.The most CRE-concentrated bank in the country is River City Bank of Sacramento, CA, at 651% of risk-based capital — roughly 4.4 times the national average.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
How concentrated are US banks in CRE? (Q1 2026)
Distribution of 4,242 banks by CRE loans as a share of total risk-based capital. The two shaded bars are above the 300% supervisory screening threshold.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop. Q1 2026 (filed March 31, 2026).
View the data behind this chart
| CRE concentration | Banks | Share of filers |
|---|---|---|
| Under 100% | 1,723 | 40.6% |
| 100–200% | 1,210 | 28.5% |
| 200–300% | 916 | 21.6% |
| 300–400% | 306 | 7.2% |
| 400–500% | 68 | 1.6% |
| 500%+ | 19 | 0.4% |
Bank CRE concentration is easing, not building
Number of US banks above the 300% and 400% CRE concentration thresholds, by quarter. These are counts of banks in the concentrated tail, which has thinned — not a measure of total CRE exposure, which has grown over the same period.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop. Q2 24–Q1 26.
View the data behind this chart
| Quarter | Banks above 300% | Banks above 400% |
|---|---|---|
| Q2 24 | 400 | 101 |
| Q3 24 | 397 | 100 |
| Q4 24 | 402 | 101 |
| Q1 25 | 409 | 98 |
| Q2 25 | 406 | 97 |
| Q3 25 | 401 | 96 |
| Q4 25 | 398 | 95 |
| Q1 26 | 393 | 87 |
Most CRE-concentrated US banks, Q1 2026
These are public regulatory filings. A high ratio indicates a business model concentrated in commercial real estate lending; it is not an indication of distress, and nothing here should be read as a comment on any institution's soundness.
| Bank | Headquarters | CRE / capital | CRE book |
|---|---|---|---|
| River City Bank | Sacramento, CA | 651% | $4.15B |
| SmartBiz Bank, National Association | Northbrook, IL | 639% | $0.07B |
| Nano Banc | Irvine, CA | 633% | $0.27B |
| OneUnited Bank | Boston, MA | 580% | $0.40B |
| Freedom Bank | Maywood, NJ | 573% | $0.56B |
| Marquette Bank | Chicago, IL | 570% | $1.09B |
| Tioga-Franklin Savings Bank | Philadelphia, PA | 569% | $0.02B |
| Crown Bank | Edina, MN | 554% | $0.19B |
| DMB Community Bank | Madison, WI | 553% | $0.55B |
| Hingham Institution for Savings | Hingham, MA | 550% | $2.96B |
| HCN Bank | Riverside, CA | 550% | $0.74B |
| RiverHills Bank | Milford, OH | 548% | $0.23B |
| Ocean Bank | Miami, FL | 522% | $5.00B |
| Five Star Bank | Roseville, CA | 521% | $3.55B |
| State Bank of Texas | Irving, TX | 514% | $2.21B |
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop. Q1 2026 (filed March 31, 2026). Bank names are not unique — several distinct institutions share a name across states — so headquarters city and state are part of the identification.
Look up any bank's CRE concentration
Search 9,023 indexed institutions. The 4,252 that file quarterly FFIEC Call Reports carry a commercial real estate concentration ratio — CRE loans as a percentage of total risk-based capital — shown here for Q1 2026 against the 300% supervisory screening threshold and the 147% average across all filers.
Enter at least two characters to search.
Ratios are computed from the institution's own Q1 2026 Call Report filing using the 2006 Interagency Guidance definition: construction and land development, multifamily, and non-owner-occupied non-farm non-residential loans, divided by total risk-based capital. Owner-occupied commercial mortgages are excluded, as the guidance specifies. A ratio above 300% is a supervisory screening threshold that prompts closer regulatory attention — it is not a limit, a violation, or a statement about any institution's condition.
3. Size and composition of the CRE loan book
Aggregating every filed Call Report gives the total commercial real estate exposure sitting on US bank balance sheets, and — more usefully — how that exposure splits between property categories. The owner-occupied versus investor split matters because the two behave very differently in a downturn: an owner-occupied loan is underwritten against an operating business, an investor loan against a rent roll.
9.US banks held $3.03 trillion in commercial real estate loans in Q1 2026, across 4,252 reporting institutions.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
11.Investor-owned commercial property loans make up 63% of banks' non-farm non-residential mortgage book ($1.21 trillion), with owner-occupied loans making up the remaining 37% ($700 billion).
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
12.Multifamily lending accounts for $660 billion of bank CRE exposure, and construction and land development for $455 billion.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
13.Concentrated CRE lending is almost entirely a small-bank phenomenon: of the 393 banks above the 300% threshold, 376 are community banks and 17 are regional banks by our size classification. None of the largest nationwide banks appear. ("Community", "regional" and "national" here describe size and footprint, not charter type — many banks above the threshold are OCC-chartered national associations.)
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
What US bank CRE exposure is made of (Q1 2026)
$3.03 trillion in bank-held commercial real estate loans, by category.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop. Q1 2026 (filed March 31, 2026).
View the data behind this chart
| Category | Balance ($B) | Share of CRE |
|---|---|---|
| Non-owner-occupied (investor) | 1,214 | 40.1% |
| Owner-occupied | 700 | 23.1% |
| Multifamily | 660 | 21.8% |
| Construction & land | 455 | 15.0% |
4. Credit quality and delinquency
Call Reports separate loans that are merely late from loans the bank has stopped accruing interest on. Nonaccrual is the more meaningful signal: it is the bank's own judgment that full repayment is in doubt. The figures below are aggregate CRE nonaccrual and past-due balances as a share of the total CRE book.
14.1.14% of US bank commercial real estate loans were on nonaccrual status in Q1 2026 — the bank's own assessment that full repayment is in doubt.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
15.A further 0.38% of bank CRE loans were 30–89 days past due and 0.05% were 90 or more days past due but still accruing in Q1 2026.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q1 2026 (filed March 31, 2026).
5. Observed rates and loan terms
Published “commercial mortgage rates” are usually indicative quotes. The figures below are different: they are rates on loans that actually closed. SBA 7(a) approvals carry the contract rate on each individual loan, and bank Call Reports let us compute the realised yield on each bank's real estate loan book.
Those two series currently move in opposite directions, and the divergence is the most interesting thing on this page. New small-business loans are being written at steadily lower rates, while the average yield banks earn on their existing real estate book has risen — older, cheaper loans are still repricing upward as they mature and reset. A borrower coming off a loan written five years ago is refinancing into a lower rate than a year ago, but still a higher one than they are leaving.
17.The median observed rate on SBA 7(a) loans approved in 2026 Q1 was 9.50%, across 1,286 approvals.
Source: SBA 7(a) loan approvals, compiled by RefiLoop, 2026 Q1.
18.Median SBA 7(a) rates have fallen 175 basis points over seven quarters, from 11.25% in 2024 Q3 to 9.50% in 2026 Q1.
Source: SBA 7(a) loan approvals, compiled by RefiLoop, 2024 Q3–2026 Q1.
19.Fixed-rate SBA 7(a) loans carry a median rate of 7.75% against 10.25% for variable-rate loans — a 250 basis point gap, though fixed-rate loans are the minority at 734 of 9,242 approvals.
Source: SBA 7(a) loan approvals, compiled by RefiLoop, 2024-07 to 2026-03.
20.The median SBA 7(a) loan approval is $293,300, with a median term of 10 years.
Source: SBA 7(a) loan approvals, compiled by RefiLoop, 2024-07 to 2026-03.
21.The median US bank earned a 6.20% yield on its real estate loan portfolio in Q1 26, up from 5.88% in Q2 24 — the existing book is still repricing upward even as new loan rates fall.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q2 24–Q1 26.
22.Median bank real estate portfolio yield peaked at 6.33% in Q4 25 and has since eased to 6.20%.
Source: FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop, Q2 24–Q1 26.
New loan rates are falling while bank book yields rise
Median observed SBA 7(a) contract rate on newly approved loans, against the median yield banks earn on their existing real estate loan portfolios.
Sources: SBA 7(a) loan approvals, compiled by RefiLoop; FFIEC Call Reports (Schedule RC-C), compiled by RefiLoop.
View the data behind this chart
| Quarter | Median SBA 7(a) rate | Median bank RE portfolio yield |
|---|---|---|
| 2024-Q3 | 11.25% | 6.05% |
| 2024-Q4 | 10.75% | 6.10% |
| 2025-Q1 | 10.25% | 6.02% |
| 2025-Q2 | 10.25% | 6.18% |
| 2025-Q3 | 10.00% | 6.30% |
| 2025-Q4 | 9.75% | 6.33% |
| 2026-Q1 | 9.50% | 6.20% |
6. Who is writing commercial mortgages now
Call Reports tell you what a bank holds. They do not tell you who is writing loans this month. For that we index county deed records directly — every commercial mortgage instrument recorded in the counties we cover, with the lender name as it appears on the recorded document. We publish this as the RefiLoop CRE Lending Activity Index, updated monthly with a permanent archive so older citations keep resolving.
Two limits are worth stating before the numbers rather than after them. First, this is a count of recordings, not dollar volume — fewer than 20 of the instruments we index state a loan amount in the recording index, so we publish no dollar figures from county data at all. Second, these are new originations, not maturities. County records show loans being written; they say nothing about loans coming due.
23.3,981 commercial mortgage instruments were recorded across the 60 counties we index in July 2026.
Source: County recording indexes, compiled by RefiLoop, July 2026.
24.Across June 2026 and July 2026, we indexed 5,981 commercial mortgage recordings in 60 counties across 5 states.
Source: County recording indexes, compiled by RefiLoop, June 2026–July 2026.
25.The most active commercial mortgage lender by recording count in July 2026 was KIAVI FUNDING INC, with 76 recorded commercial mortgage instruments across 16 counties.
Source: County recording indexes, compiled by RefiLoop, July 2026.
26.Commercial mortgage origination is strikingly fragmented: the busiest lender in our July 2026 index accounted for only 1.9% of recorded commercial mortgages, and the top 20 lenders together accounted for 15%.
Source: County recording indexes, compiled by RefiLoop, July 2026.
27.Non-bank and private lenders appear consistently among the most active commercial mortgage originators by recording count, alongside regional and community banks — a mix invisible in bank regulatory data, which by construction only covers banks.
Source: County recording indexes, compiled by RefiLoop, July 2026.
RefiLoop CRE Lending Activity Index — July 2026
Most active commercial mortgage lenders by count of recorded instruments among the 60 counties we index. Counts, not dollar volumes.
Source: County recording indexes, compiled by RefiLoop. July 2026. Lender names appear as recorded on the instrument and are not consolidated across affiliates.
View the data behind this chart
| Lender (as recorded) | Recordings | Counties |
|---|---|---|
| KIAVI FUNDING INC | 76 | 16 |
| PROSPERITY BANK | 57 | 9 |
| SSB COMMUNITY BANK | 54 | 2 |
| FROST BANK | 48 | 10 |
| RFLF 4 LLC | 47 | 13 |
| FIRST FINANCIAL BANK | 40 | 12 |
| HUNTINGTON NATIONAL BANK | 34 | 7 |
| FIRST COMMONWEALTH BANK | 23 | 5 |
| CONSTRUCTION LOAN SERVICES II LLC | 21 | 4 |
| CITY BANK | 20 | 5 |
| SCF JAKE LP | 20 | 1 |
| CREBRID LLC | 18 | 2 |
| JPMORGAN CHASE BANK | 18 | 1 |
| MYLOANPATHWAY | 16 | 5 |
| SCHUMER-YARBROUGH DEVELOPMENT LLC | 16 | 1 |
| BPL MORTGAGE LLC | 15 | 6 |
| CAPITAL BANK | 15 | 3 |
| RBI PRIVATE LENDING LLC | 15 | 2 |
| SIMMONS BANK | 15 | 4 |
| AMEGY BANK | 14 | 2 |
Why we do not rank counties against each other
Our ability to tell a commercial mortgage from a residential one depends on what each county clerk publishes in its recording index. In some counties the instrument type and legal description make the distinction clean; in others it is largely absent. Detection rates therefore vary by a factor of several between counties, which means a county with more indexed commercial recordings may simply have a more informative index rather than a busier lending market. We publish lender rankings, which are computed within a consistent detection regime, and we do not publish county-versus-county comparisons or “busiest market” claims, because our data cannot honestly support them.
7. Methodology and limits
What is counted
Bank figures come from the FFIEC Call Report, the quarterly regulatory filing every US bank submits. We ingest all 4,252 filings for Q1 2026 and compute concentration ratios ourselves using the 2006 Interagency Guidance definition — construction and land development, plus multifamily, plus non-owner-occupied non-farm non-residential loans, divided by total risk-based capital. Owner-occupied commercial mortgages are excluded, as the guidance specifies. Balance sheet figures are reported in thousands of dollars and are aggregated across all filers.
Rate data
SBA figures cover 9,242 individual 7(a) loan approvals from July 1, 2024 to March 31, 2026, each carrying its own contract rate — these are closed-loan rates, not quotes. Bank portfolio yields are computed from Call Report interest income on real estate loans against average balances, so they describe the whole existing book rather than new originations. The two are not comparable to each other as “the rate” and we do not present them that way.
County recording data
We index commercial mortgage instruments recorded in 61 counties across Texas, Ohio, Pennsylvania, Colorado and Georgia, refreshed daily. Every indexed record carries both a county document number and a link to the clerk's record at 100% coverage, so any figure here can be checked against the original filing. We publish only records our classifier identifies as commercial; records it cannot confidently classify are excluded rather than assumed.
What this data cannot tell you
- No dollar volumes from county records. Loan amounts appear on fewer than 20 of the instruments we index, so we publish counts and rankings only.
- Originations, not maturities. County records show loans being written. Nothing here speaks to loans coming due.
- No county-versus-county comparison. Commercial detection rates vary widely by county index quality, so cross-county volume comparisons would be an artefact of the source data.
- No monthly trend claims yet. Multi-county collection began in June 2026. Two months is enough for a monthly snapshot and not enough for a trend, and we will not publish direction-of-travel claims from county data until more months accumulate.
- County coverage is not the national market. Every county figure is stated as being among the recordings we index in those counties, never as a national or market total.
- Credit unions have no CRE concentration ratio here. They file NCUA 5300 reports rather than FFIEC Call Reports, so the lookup tool returns an explanation rather than a number for the 4,344 credit unions in our registry.
Corrections
If you find an error, tell us and we will fix it and note the correction. Accuracy on this page matters more to us than any individual number in it.
8. Download the data
Both published tables are available as CSV, free to use with attribution. If you are writing about this and need a cut we do not publish, ask — we would rather you had the right numbers.
CRE concentration by bank (CSV)
All 4,242 banks with a computable Q1 2026 concentration ratio, with CRE balances, risk-based capital, total assets and quarterly growth.
CRE Lending Activity Index (CSV)
Monthly commercial mortgage recording counts by lender and county coverage, for every month in the published index.
Monthly archive: June 2026 · July 2026. Each archived edition keeps its numbers frozen at that month so older citations continue to resolve.
RefiLoop is a commercial mortgage brokerage. This page is published as reference material. It is not an offer of credit, a rate quote, a guarantee of terms, or a recommendation of any institution. Rates shown are historical observations for the periods stated and are not available terms. Figures describing named institutions are taken from their own public regulatory filings and are presented without characterisation.