FY 2027 Fair Market Rents: Where HUD’s Rent Benchmark Rose and Where It Fell
On October 1, 2026, a new set of Fair Market Rents took effect across the United States. They are the figures the U.S. Department of Housing and Urban Development (HUD) uses for the next twelve months to decide, among other things, how much rent a housing voucher will cover in each area. HUD posts them as spreadsheets on its Fair Market Rents data page, one row per county.
We downloaded the new file and last year’s and matched them row by row. The short version: the two-bedroom benchmark went up in 3,416 places and down in 1,338. But the places where it fell include some of the biggest and most expensive rental markets in the country, and that’s the part worth a closer look.
Every count below is ours, built from HUD’s files. HUD published the data, not these conclusions. How we count is written down once in our Editorial Methodology.
First, what a Fair Market Rent actually is
It is not the average rent, and it isn’t what a landlord is allowed to charge. The regulation that defines it, 24 CFR 888.113, describes Fair Market Rents (FMRs) as “estimates of rent plus the cost of utilities, except telephone,” and says the figure is drawn from the rents of homes “occupied by recent movers,” generally at the 40th percentile.
Put simply: line up what people who moved recently are paying in an area, utilities included. The FMR sits a bit below the middle. About four in ten of those homes cost less, about six in ten cost more.
That makes it a modest benchmark by design, and a useful one for a reason that has nothing to do with vouchers: it exists for every county in the country, it’s calculated the same way everywhere, and it comes out every year. Few other rent figures can say all three.
“FY 2027” is HUD’s fiscal year 2027, which runs from October 1, 2026 to September 30, 2027. Unless we say otherwise, the figures here are for a two-bedroom home.
The national picture: seven in ten went up
The FY 2027 county file has 4,764 rows, and so does the revised FY 2026 file. We could match 4,758 of them between the two years using HUD’s location code; six rows in each file had no partner in the other and are left out of everything that follows.
Of those 4,758 rows:
| Two-bedroom FMR, FY 2026 to FY 2027 | Rows | Share |
|---|---|---|
| Went up | 3,416 | 71.8% |
| Went down | 1,338 | 28.1% |
| Did not change | 4 | 0.1% |
The typical row moved by 3.4 percent: half rose by more than that or fell by less, half the other way. In 603 rows, about one in eight, the benchmark jumped 10 percent or more in a single year. In 429 it dropped 5 percent or more.
One caution about the word “row.” In most states a row is a county. In the six New England states it’s a town, which is why Maine alone has 527 rows. So “3,416 places” means 3,416 rows of the file, not 3,416 counties. Where that difference changes the story, we say so.
Counting areas instead of rows
Several counties often share one FMR area, and every county in it gets the same rents. The greater Chicago area is one example; so is metropolitan Washington. Count each area once and the file holds 2,606 of them. On that basis the split is wider: 2,135 areas up, 467 down, 4 unchanged. Roughly eight areas in ten saw the benchmark rise.
Why does the share of increases grow when you count areas? Because many of the decreases sit in New England, where a single area is spread across dozens of town rows. New Hampshire is the clearest case: 207 of its 259 rows fell, yet those rows belong to a small number of areas.
Big metros split almost down the middle
Here is where the national average hides the most. We took the FMR areas whose rows add up to at least one million people, using the 2024 population figure HUD includes in the file. There are 64. In 37 the two-bedroom benchmark rose. In 27 it fell.
The largest increases among those 64 are concentrated in the Midwest and the Mid-Atlantic:
| FMR area (1 million people or more) | FY 2026 | FY 2027 | Change |
|---|---|---|---|
| Kansas City, MO-KS | $1,358 | $1,546 | +13.8% |
| Chicago-Joliet-Naperville, IL | $1,781 | $2,011 | +12.9% |
| Columbus, OH | $1,430 | $1,602 | +12.0% |
| St. Louis, MO-IL | $1,218 | $1,349 | +10.8% |
| Providence-Fall River, RI-MA | $1,729 | $1,879 | +8.7% |
| Birmingham-Hoover, AL | $1,266 | $1,375 | +8.6% |
| Washington-Arlington-Alexandria, DC-VA-MD | $2,246 | $2,438 | +8.5% |
| Baltimore-Columbia-Towson, MD | $1,857 | $2,006 | +8.0% |
Chicago and Baltimore both crossed $2,000 for a two-bedroom this year. And the decreases? They read like a list of the places that got expensive fastest over the past several years:
| FMR area (1 million people or more) | FY 2026 | FY 2027 | Change |
|---|---|---|---|
| Rochester, NY | $1,573 | $1,416 | -10.0% |
| Salt Lake City, UT | $1,747 | $1,615 | -7.6% |
| Dallas, TX | $1,931 | $1,807 | -6.4% |
| San Jose-Sunnyvale-Santa Clara, CA | $3,483 | $3,272 | -6.1% |
| San Diego-Chula Vista-Carlsbad, CA | $3,001 | $2,823 | -5.9% |
| Phoenix-Mesa-Chandler, AZ | $1,839 | $1,734 | -5.7% |
| Milwaukee-Waukesha, WI | $1,338 | $1,263 | -5.6% |
| Santa Ana-Anaheim-Irvine, CA | $3,236 | $3,057 | -5.5% |
Houston and the Riverside-San Bernardino area fell too, by 4.8 and 4.5 percent. A lower benchmark in San Jose still leaves it at $3,272, more than double Kansas City’s after its jump. A falling FMR doesn’t make a place cheap. It means this year’s estimate came in below last year’s.
State by state: the widest swings
Looking at the typical row within each state, the strongest increases were in North Dakota, where the middle row rose 16.2 percent and 46 of 53 went up; Vermont, at 14.5 percent, with all 256 of its town rows rising; and Alabama, at 13.2 percent. In South Dakota every one of the 66 rows increased.
At the other end, the typical row fell 8.4 percent in New Hampshire and 7.8 percent in Arizona, where 14 of 15 rows went down. Nevada followed, with 15 of its 17 rows lower. In Florida, 42 of 67 counties saw a decrease.
The same state can look different depending on how you weigh it. Illinois is the best example: its typical row rose only 1.5 percent, because most of its counties are rural and barely moved, but the Chicago area’s 12.9 percent increase covers most of the state’s population. If you rent in Chicago, the statewide middle tells you nothing useful. Check your own area.
Why so many decreases are exactly 10 percent
Sort the file by the size of the drop and something odd shows up: 108 rows fell by exactly 10 percent, and none by more. That isn’t a coincidence, and it isn’t the market.
The same regulation says HUD sets each FMR at the 40th percentile rent “but no lower than 90 percent of the previous year’s FMR for the FMR area.” The rule is a brake: however far the estimate falls, the published number can’t lose more than a tenth in a year.
Those 108 rows belong to 29 FMR areas. Seventy-four of the rows are New Hampshire towns in three county areas: Belknap, Cheshire and Grafton. The rest include Flagstaff and Yuma in Arizona, Salinas and the Oxnard-Thousand Oaks-Ventura area in California, eleven Florida counties, Rochester in New York, Reading and Erie in Pennsylvania, and Charlottesville in Virginia.
What should you read into a 10 percent drop? Only that the underlying estimate fell by at least that much. It may have fallen further, and the floor may hold the published figure up again next year. It is not evidence that rents in Rochester are a tenth lower than they were.
The most and least expensive benchmarks
Across the 50 states and the District of Columbia, the highest two-bedroom FMR for FY 2027 is in the Santa Cruz-Watsonville area of California, at $4,254. San Francisco follows at $3,697, then Nantucket County, Massachusetts, at $3,583, Napa at $3,375 and San Jose at $3,272. Boston sits at $3,008 and the New York City area at $2,971.
The lowest is $885, shared by 13 rural counties in South Carolina that HUD groups under the same figure, with Allendale, Marion and Williamsburg among them. Nearby come Bamberg County, South Carolina, at $889, and two Louisiana parishes at $892. The file also covers five U.S. territories, where figures run lower still: several areas of Puerto Rico are at $517.
So the top benchmark in the states is almost five times the bottom one. In all, 542 rows of the file are at $2,000 or more for a two-bedroom, and 826 are under $1,000.
The small-county problem
The single biggest increase in the whole file is Jackson County, Colorado, where the two-bedroom figure went from $989 to $1,485, a jump of 50.2 percent. It would make a striking headline. It would also be misleading.
HUD’s file lists Jackson County’s population as 1,372. In a place that small there are very few rentals to measure, so the estimate can swing sharply on little data. The next largest increases are in similar places: Grant County, South Dakota, up 26.9 percent, and Mountrail County, North Dakota, up 26.6 percent, each with fewer than 10,000 people.
That’s why the tables above use areas of a million people or more. A 13.8 percent rise in Kansas City rests on far more data than a 50 percent rise in a county of 1,400, and it reaches far more renters.
Metro and rural areas moved about the same
HUD flags each row as inside or outside a metropolitan area. Of the 1,971 metropolitan rows we matched, 1,473 rose and 496 fell. Of the 2,787 non-metropolitan rows, 1,943 rose and 842 fell. The typical metropolitan row increased 3.0 percent and the typical non-metropolitan row 4.0 percent.
Rural rows rose a little faster in percentage terms, from a lower starting point. There’s no sign in this file of a gap opening between city and country benchmarks this year.
What this means if you rent
If you use a Housing Choice Voucher, the FMR is the starting point for the payment standard your local housing agency sets, so a change in your area’s figure can change how much rent your voucher reaches. The agency decides the actual amount and the timing, not HUD’s spreadsheet. Ask your housing agency what its payment standards are for the new fiscal year.
If you don’t, the FMR is still a free, official reference point when you’re sizing up a listing. Three things to keep in mind when you use it:
- It includes utilities. Compare it with rent plus what you’d pay for electricity, gas, water and heating, not with the advertised rent alone.
- It’s below the middle of the market on purpose. A listing above the FMR isn’t overpriced for that reason. Most recent rentals in the area are above it.
- It covers a whole area. One figure can span a city center and suburbs an hour away. For many metropolitan areas HUD also publishes Small Area FMRs by ZIP code, which are closer to what a neighborhood costs. We measured how far they spread inside a single metro in Same Metro, Double the Rent.
And one thing it can’t do: it won’t tell you whether your landlord may raise your rent, or by how much. That depends on your lease and on state and local law, which this file doesn’t touch.
How to look up your own county
You don’t need a spreadsheet. On HUD’s Fair Market Rents page, choose the fiscal year, then your state and county, and the site shows the figure for each home size from a studio to four bedrooms, along with the steps HUD took to calculate it. If you’d rather see the whole country at once, the same page has the county file we used here.
The limits of this count
This comparison says how HUD’s benchmark changed between two fiscal years. It doesn’t measure what happened to asking rents this month, because the estimates are built on survey data collected earlier and then adjusted forward. It compares FY 2027 with the revised FY 2026 file, so it won’t match articles written last fall from the original release. And HUD can still revise FY 2027; if it does, we’ll rerun the numbers and note the change here.
Sources
- HUD USER, Fair Market Rents: the FY 2027 county-level file and the revised FY 2026 county-level file. Every count, share, ranking and dollar figure in this article comes from these two files; the calculations are ours.
- 24 CFR 888.113, Fair market rents for existing housing: Methodology: the definition of an FMR, the 40th percentile and recent movers, and the rule that an FMR is set no lower than 90 percent of the previous year’s.
Figures checked against the sources on October 5, 2026.
