Vacant Residential Land in the Twin Cities, by City

Published August 26, 2026 · Parcel data as of April 1, 2026 · 92 municipalities · Method and sources

Across the seven-county metro, the share of residential parcels sitting vacant runs from 0.08% in Richfield to 24.52% in Corcoran — roughly a 307-fold difference between two cities about twenty-five miles apart. The median city is at 3.28%. 22 of these 92 cities are below 1%; 11 are above 10%.

Those figures come from applying one consistent test to every municipality in the regional parcel dataset. The test is stated in full below, and it is deliberately narrow: it finds parcels the county assessor records as residential land carrying no building. It does not find buildable lots, and the gap between those two things is wide enough that we would rather lead with it than bury it.

What this measures

A parcel is counted as vacant residential land when all of the following are true in the county assessor's own record:

  • No meaningful building value. The assessor records no building value on the parcel, or less than $1,000 of it.
  • Land value above zero. It is a real taxable parcel, which excludes most road right-of-way, open water and public land.
  • A residential or undeveloped use class. The assessor's use classification reads as residential, vacant, agricultural, rural or lakeshore.
  • Not classed as something else. Parcels classed commercial, industrial, apartment, church, charitable, common area, utility or railroad are excluded.

That is the whole test. It runs identically on all 186 municipalities in the dataset, from a single extract dated April 1, 2026.

What this is not

Specifically, the measure:

  • Cannot see land that could become buildable. Assembling neighbouring parcels, subdividing an oversized one, and taking down an existing house are how most new homes appear in a built-out city. None of them are visible here, which is why a city near 0% is not a city that has stopped building.
  • Counts a farm field and an infill lot the same. One parcel is one parcel whether it is a fifth of an acre or forty acres. The outer-edge cities at the top of this table hold larger parcels on average, so their share overstates how many individual homes the land represents relative to a city of small lots.
  • Includes platted outlots. An outlot is the drainage pond, wetland buffer or common area a developer plats alongside the buildable lots, and it usually cannot be built on. In the 40 cities where we have checked parcel by parcel, about 7% of these vacant parcels carry an outlot designation.
  • Is a snapshot. Parcel records change continuously. Treat one city's number as of April 1, 2026, not as a live count.

We do run a further buildability screen — checking geometry, frontage, flood and wetland indicators and a setback envelope — but only in the 40 cities we have ingested in detail. There it retains 72% of these vacant parcels (14,138 of 19,643), removing about a quarter. That screen has not run on the other cities in this table, so its result is reported here as its own scoped figure and is never mixed into the comparison.

The pattern the ordering shows

The ranking is not random, and that is the main reason it is worth publishing. The cities at the tight end are the fully built-out post-war suburbs of the first ring — Richfield, New Hope, Saint Louis Park, Brooklyn Center, Robbinsdale — where the residential land was platted decades ago. The cities at the loose end sit on the developing edge, where agricultural parcels have not yet been converted: Dayton, Rogers, Linwood Township, Nowthen, Corcoran.

For anyone planning a build, that ordering is the practical finding. It says where buying a lot is likely to mean competing for a scarce one, and where it is more likely to mean choosing among several — without claiming to know how many of either exist.

All 92 cities

Sorted by vacant share, lowest first. Choose a column heading to re-sort. Cities with a link have individual lot pages in our lot finder (37 of 92). Municipalities under 2,000 parcels are left out: with a denominator that small a single new plat moves the percentage several points, and a number that unstable is not worth quoting.

Richfield Hennepin 9 11,927 0.08%
New Hope Hennepin 12 6,145 0.20%
Saint Louis Park Hennepin 53 17,662 0.30%
Brooklyn Center Hennepin 35 8,913 0.39%
Robbinsdale Hennepin 21 5,284 0.40%
Champlin Hennepin 37 8,978 0.41%
Bloomington Hennepin 139 31,131 0.45%
Brooklyn Park Hennepin 113 24,812 0.46%
New Brighton Ramsey 42 7,625 0.55%
Spring Lake Park Anoka + Ramsey 14 2,471 0.57%
Minneapolis Hennepin 832 131,129 0.63%
Circle Pines Anoka 14 2,083 0.67%
Mounds View Ramsey 29 4,236 0.68%
Hopkins Hennepin 37 5,252 0.70%
Apple Valley Dakota 125 16,019 0.78%
Edina Hennepin 166 21,372 0.78%
Columbia Heights Anoka 60 7,564 0.79%
Eden Prairie Hennepin 211 23,019 0.92%
Coon Rapids Anoka 209 22,583 0.93%
Golden Valley Hennepin 82 8,771 0.93%
Oak Park Heights Washington 20 2,084 0.96%
North Saint Paul Ramsey 41 4,245 0.97%
Eagan Dakota 208 20,299 1.02%
Stillwater Washington 91 8,622 1.06%
Crystal Hennepin 91 8,286 1.10%
Shoreview Ramsey 119 10,779 1.10%
Burnsville Dakota 187 16,568 1.13%
White Bear Lake Ramsey + Washington 102 9,025 1.13%
Plymouth Hennepin 378 29,201 1.29%
Anoka Anoka 82 5,985 1.37%
Roseville Ramsey 182 13,097 1.39%
Saint Anthony Hennepin + Ramsey 45 3,019 1.49%
Maplewood Ramsey 218 13,967 1.56%
Savage Scott 193 11,689 1.65%
Minnetonka Hennepin 352 20,971 1.68%
Arden Hills Ramsey 57 3,378 1.69%
Saint Paul Ramsey 1,475 84,950 1.74%
Oakdale Washington 196 11,055 1.77%
Chanhassen Carver + Hennepin 202 10,854 1.86%
Vadnais Heights Ramsey 113 5,571 2.03%
White Bear Township Ramsey 113 5,270 2.14%
Woodbury Washington 631 29,433 2.14%
Mahtomedi Washington 73 3,339 2.19%
Blaine Anoka + Ramsey 664 26,342 2.52%
Maple Grove Hennepin 909 28,405 3.20%
Mendota Heights Dakota 142 4,382 3.24%
Chaska Carver 342 10,437 3.28%
Prior Lake Scott 386 11,667 3.31%
Fridley Anoka 343 9,917 3.46%
West Saint Paul Dakota 210 5,996 3.50%
South Saint Paul Dakota 281 7,959 3.53%
Mound Hennepin 171 4,802 3.56%
Hastings Dakota + Washington 295 7,742 3.81%
Andover Anoka 521 12,339 4.22%
Shorewood Carver + Hennepin 147 3,329 4.42%
Shakopee Scott 706 15,869 4.45%
Cottage Grove Washington 773 16,494 4.69%
Ramsey Anoka 561 11,503 4.88%
Forest Lake Washington 446 8,958 4.98%
Victoria Carver 262 5,229 5.01%
Inver Grove Heights Dakota 598 11,000 5.44%
Hugo Washington 427 7,757 5.50%
Little Canada Ramsey 246 4,130 5.96%
Lake Elmo Washington 373 6,107 6.11%
Lakeville Dakota 1,651 26,088 6.33%
Waconia Carver 369 5,658 6.52%
Farmington Dakota 577 8,440 6.84%
Saint Paul Park Washington 165 2,364 6.98%
Lino Lakes Anoka 664 9,289 7.15%
Jordan Scott 182 2,443 7.45%
Orono Hennepin 341 4,477 7.62%
Grant Washington 168 2,166 7.76%
Saint Francis Anoka 245 3,135 7.81%
Ham Lake Anoka 583 6,900 8.45%
North Oaks Ramsey 204 2,364 8.63%
Belle Plaine Scott 270 3,070 8.79%
Carver Carver 258 2,790 9.25%
Minnetrista Hennepin 436 4,700 9.28%
Columbus Anoka 231 2,421 9.54%
Oak Grove Anoka 395 4,092 9.65%
Elko New Market Scott 202 2,083 9.70%
Credit River Scott 254 2,421 10.49%
East Bethel Anoka 629 5,564 11.30%
Rosemount Dakota 1,210 10,642 11.37%
Medina Hennepin 440 3,751 11.73%
May Township Washington 275 2,045 13.45%
Scandia Washington 403 2,907 13.86%
Dayton Hennepin 705 4,911 14.36%
Rogers Hennepin 1,083 6,862 15.78%
Linwood Township Anoka 471 2,833 16.63%
Nowthen Anoka 382 2,142 17.83%
Corcoran Hennepin 1,173 4,784 24.52%

Method, and where every number comes from

The source is the regional parcel dataset, which aggregates the seven metro counties' assessor and GIS records. We applied the test in What this measures to every municipality in one pass from a single extract dated April 1, 2026, so no city's figure is computed differently from another's.

Two things are worth stating because they would otherwise be invisible. Seven cities cross a county line — Blaine, Chanhassen, Hastings, Saint Anthony, Shorewood, Spring Lake Park and White Bear Lake — and appear in the source data once per county. Their rows here are combined, because reporting only the larger half understates the denominator: Saint Anthony reads 1.96% split and 1.49% whole. Seven other pairs share a name with a neighbouring township — Waconia and Waconia Township, and similar — and those are genuinely different places, kept apart here.

Claim on this page Where it comes from
Richfield 0.08%, Corcoran 24.52%, and every row of the table Parcel and vacant-parcel counts per municipality, regional parcel dataset extract dated April 1, 2026, grouped by municipality name.
92 municipalities shown, of 186 in the dataset Those with at least 2,000 residential parcels, the stated floor.
Median 3.28%, 22 cities below 1%, 11 above 10% Computed from the table on this page at build time, not entered separately.
Buildability screen retains 72% 14,138 retained of 19,643 screened across the 40 cities where the screen has run. Not applied to the rest of the table.
About 7% of vacant parcels are platted outlots Plat designation on screened parcels in the same 40 cities. Not measured for cities outside that set.
Saint Anthony 1.96% split, 1.49% combined 45 vacant parcels against 2,295 in Hennepin alone, versus 3,019 across Hennepin and Ramsey.

If you are checking a figure and it does not reconcile, we would rather hear about it than not — get in touch.

Where to go from here

Our lot finder maps individual candidate parcels in 37 of the cities above, with what we know and what we do not stated on each one. If you are further along, our cost-to-build guide covers what land scarcity does to a Twin Cities budget, and the builder directory lists builders by the cities they actually work in.

If you would rather see finished new construction in the outer-edge cities that still have land, the Fall 2026 Parade of Homes runs September 11–27, and several cities at the loose end of this table have homes on the tour.

Frequently Asked Questions

Does this mean Richfield has only 9 buildable lots left?

No, and that is the most important thing on this page. 9 is the number of Richfield parcels the county assessor records as residential land carrying no building. Whether any of them can actually be built on depends on setbacks, zoning, easements, wetlands, access and utility service, none of which this measure checks. It also cannot see the land that produces most infill housing in a built-out city: teardowns, splitting an oversized lot, or assembling neighbours. A fully built-out city can keep producing new homes for decades with a number like this.

Why compare cities if the absolute numbers are limited?

Because the same test is applied to every city, from one dataset, on one date. Whatever the measure misses, it misses everywhere at roughly the same rate, so the ordering carries information the individual counts do not. Richfield at 0.08% and Corcoran at 24.52% is a real 307-fold difference in how much undeveloped residential land the assessor sees.

What counts as a parcel here?

A tax parcel in the regional parcel dataset, which is the assessor record rather than a lot on a plat. One parcel can be a quarter-acre city lot or a forty-acre farm field, and this measure counts each as one. That is why the share is expressed as a percentage of a city’s own residential parcels rather than as acreage, and why the two are not interchangeable.

How current is this?

The parcel extract is dated April 1, 2026 and covers all seven metro counties. Parcel records change continuously as land is platted, sold and built on, so treat any single city’s figure as a snapshot rather than a live count.

Can I cite or reuse this?

Yes. Please link to this page and state the measurement date, and please carry the distinction between vacant residential parcels and buildable lots with it — that distinction is the reason the numbers are trustworthy enough to quote.

Building on a lot in one of these cities?

Tell us the city and the shape of your build, and we'll match you with builders who work there.

Start My Free Match