By lender →
Official federal data → published methodology → reproducible calculation → FinanceRateCalc interpretation
We are not the source. We are the layer that makes the source usable — and every step above is checkable.
FRC Research · The 2025 Edition · 319 metros

FHA denial rates by metro area, 2025.
The only free FHA-specific metro table.

A note on what makes this table different: it is FHA-specific and 2025, with a decision-rate denominator (actions 1, 2, 3). Directory-style metro tables typically blend all loan types on 2024 data and divide by all applications — including withdrawn and incomplete files — which produces systematically lower, non-comparable rates.

Computed from the complete 2025 federal HMDA record: every FHA application reaching a credit decision, aggregated to 319 metropolitan areas and divisions. Among the 210 metros with at least 1,000 decisioned applications, the denial rate ran from 9.0% in Amarillo, TX to 32.9% in Bridgeport-Stamford-Norwalk, CT — a 3.6× spread by geography alone, before a single lender is chosen.

The metro small-loan penalty: the loan-size gap seen at the state level sharpens at metro resolution. In El Paso, TX, a small FHA application (<$150K) was denied 3.94× as often as a large one (35.1% vs 8.9%) in 2025 — the widest measured metro penalty. Existing metro-level mortgage studies track approvals or demographics; none publishes the FHA loan-size axis. This table does.
what these spreads are not

Each metro shows its highest-volume FHA lenders, up to five. The spread quoted is the range within that set — not the range across every lender operating there. A metro can contain a smaller institution denying at 95% that never enters a top-five-by-volume list, so these figures understate the full dispersion and should not be read as a ranking of which metro varies most. An independent computation across all lenders with 100+ decisioned applications produces larger figures in several markets: correction of 28 July.

The extremes (metros with 1,000+ applications)

Hardest five:

Softest five:

Biggest small-loan penalties:

The spread lives inside your city: in Cleveland, OH, high-volume FHA doors (lenders with ≥100 decisioned applications; five qualify here) ranged from 6.4% (CrossCountry) to 80.1% (AmeriSave) denial in 2025 — a 73.7-point gap inside one metro. Pittsburgh: 72.7 points. The lender choice moves your odds more than the city does.
⟦FRC:metro-gap-cleveland-oh-2025:73.7pp:369427ab:w2610⟧ verify · Cleveland gap receipt: the hash changes if this figure is ever corrected

New finding from these cells: the same national brand behaves differently by city — loanDepot ran 13.6% in Boise and 58.4% in Detroit. Same Lender, Different City →

The 60 largest FHA metros

Metro / divisionDecisioned appsDenial rateSmall-loan denialSmall-loan penalty
Houston-The Woodlands-Sugar Land, TX33,79817.7%38.9%2.29×
Atlanta-Sandy Springs-Roswell, GA30,56827.3%55.4%2.37×
Phoenix-Mesa-Scottsdale, AZ27,10218.2%55.5%3.38×
Dallas-Plano-Irving, TX22,53317.1%46.7%2.97×
Riverside-San Bernardino-Ontario, CA20,75921.3%52.4%2.61×
Chicago-Naperville-Evanston, IL17,59130.0%50.9%2.02×
Tampa, FL15,62421.7%51.3%2.67×
Orlando-Kissimmee-Sanford, FL14,86324.5%52.3%2.37×
San Antonio-New Braunfels, TX14,61516.3%30.6%2.10×
Charlotte-Concord-Gastonia, NC-SC13,57321.9%45.4%2.41×
Denver-Aurora-Lakewood, CO12,02223.1%55.9%2.56×
Indianapolis-Carmel-Anderson, IN11,92123.6%37.8%1.84×
Los Angeles-Long Beach-Glendale, CA11,91924.9%55.1%2.31×
Fort Worth-Arlington, TX11,39616.4%45.8%2.88×
Las Vegas-Henderson-Paradise, NV11,09418.1%47.1%2.93×
Baltimore-Columbia-Towson, MD10,84223.5%47.7%2.39×
Nashville-Davidson--Murfreesboro--Franklin, TN10,67319.3%47.3%2.85×
St. Louis, MO-IL10,45820.9%32.9%2.30×
Lakeland-Winter Haven, FL9,62825.8%43.7%1.78×
Cincinnati, OH-KY-IN9,46321.2%36.6%2.22×
Jacksonville, FL9,25222.2%44.1%2.30×
Kansas City, MO-KS8,53019.5%35.7%2.41×
Austin-Round Rock, TX8,50315.5%——
Columbus, OH8,39022.2%40.6%2.17×
Minneapolis-St. Paul-Bloomington, MN-WI8,19821.5%47.3%2.52×
Cleveland, OH8,02825.0%37.1%1.93×
Washington, DC-MD7,96126.8%63.1%2.51×
Warren-Troy-Farmington Hills, MI7,49727.6%38.3%1.64×
Marietta, GA7,37021.3%51.8%2.80×
Oklahoma City, OK7,18819.1%37.7%2.31×
New York-Jersey City-White Plains, NY-NJ7,14429.1%——
Virginia Beach-Norfolk-Newport News, VA-NC7,09021.4%51.9%2.93×
Detroit-Dearborn-Livonia, MI7,05230.7%39.2%1.65×
Cape Coral-Fort Myers, FL7,02123.2%54.6%2.62×
Miami-Miami Beach-Kendall, FL6,97529.4%68.2%2.52×
Arlington-Alexandria-Reston, VA-WV6,93523.6%——
Memphis, TN-MS-AR6,76023.9%44.5%2.66×
Louisville/Jefferson County, KY-IN6,59321.6%41.6%2.62×
Fort Lauderdale-Pompano Beach-Deerfield Beach, FL6,52331.9%81.7%2.86×
Pittsburgh, PA6,46426.2%36.8%1.88×
Birmingham-Hoover, AL6,40222.9%42.3%2.58×
Richmond, VA6,38822.4%54.6%2.98×
Philadelphia, PA6,36027.5%47.1%2.37×
Sacramento--Roseville--Arden-Arcade, CA6,27722.5%——
Portland-Vancouver-Hillsboro, OR-WA6,05017.9%——
Camden, NJ6,04422.4%48.0%2.54×
San Juan-Carolina-Caguas, PR6,03916.9%18.0%1.21×
Providence-Warwick, RI-MA5,66621.4%48.8%2.52×
Lakewood-New Brunswick, NJ5,64924.8%47.9%2.08×
Newark, NJ-PA5,53826.6%——
Greenville-Anderson-Mauldin, SC5,51419.7%39.2%2.35×
Raleigh, NC5,45521.1%49.7%2.76×
Tucson, AZ5,41019.2%49.6%3.28×
Columbia, SC5,32520.7%40.6%2.23×
Tulsa, OK5,27917.0%37.0%3.01×
Bakersfield, CA5,03820.1%47.9%2.74×
Deltona-Daytona Beach-Ormond Beach, FL4,95122.7%54.3%2.76×
North Port-Sarasota-Bradenton, FL4,89023.4%57.9%2.68×
Charleston-North Charleston, SC4,88318.6%48.1%2.99×
Salt Lake City, UT4,84117.3%——

Full 320-metro dataset (CC BY 4.0): JSON · Hugging Face (one-line load). Rows in orange: small applications at least 2.5× likelier to be denied than large ones.

The loan-size penalty by state → Printable one-pager for agents → By lender: all 100 → The Two Walls →

First published 2026-07-22 by FinanceRateCalc · free to reuse with attribution (CC BY 4.0) · license & publication record

Prefer it while you shop? The Chrome extension shows a metro's FHA record on any Zillow, Redfin or Realtor.com listing.

The Denial Dispatch
One finding a week from the federal mortgage record.
One chart, three paragraphs, every Saturday. Measured, not assumed.
Get the Dispatch →
AI Accuracy Index The Door Effect The Denial Map Open Data About 184 Metro Gaps Evidence Navigator Hallucination Files Press Newsletter
FinanceRateCalc · Independent analysis of the complete federal HMDA record · Measured, not assumed. · No lender or AI vendor funds or previews this work.