When local voters got a direct say on data centers this week, they mostly said no. On Nov. 3, 2026, voters in 10 states decided 41 local ballot measures on data center development. Ballotpedia’s results table shows the side favoring limits winning at least 37 of them.
The breakdown is lopsided. Ballotpedia’s pre-election analysis grouped the 41 measures into 28 that would limit data centers, through bans, moratoriums, voter-approval requirements and similar rules, and 13 that would make development easier. Its results table marks 24 of the 28 restrictive measures as approved. Three more show majorities in favor but were not yet marked approved when we checked on Nov. 4. Just one restrictive measure, in Harwood, N.D., was voted down. All 13 measures that would have eased development lost, most by wide margins.
The counts are election-week tallies and are not final until local officials certify them. Nine of the 41 were advisory questions that do not bind local officials. Even so, they give a direct read on how residents in these places feel about the facilities. Ballotpedia tracked 50 data center measures in 13 states in 2026, and every one decided before Election Day also went the restrictive way.
Bans, from Southern California to rural Ohio
Ohio had 21 of the 41 measures, more than any other state, and most of its approved measures are outright prohibitions. Voters in Hubbard and Urbana banned data centers within city limits. Granville prohibited their construction by 82% to 18%. Trenton and Pataskala barred large-scale facilities. Conneaut, Oregon and Sunbury approved bans keyed to electricity demand. Piqua approved a charter amendment prohibiting construction. In Perry, 87% of voters approved a measure requiring hyperscale developers that seek special tax incentives to provide financial guarantees to the community.
Ohio also produced the clearest rejection of a specific site. In Wilmington, voters overturned four rezonings from rural residential to light industrial, covering about 546 acres combined, each by roughly 80% to 20%. Ballotpedia’s preview, citing WCPO News, said a specific data center company could build on that land if the rezonings stood.
Results elsewhere followed the same pattern:
- Alhambra, Calif.: Measure DB, which bans data centers in the city, passed 82% to 18%, with 12,496 yes votes.
- Edgewater, Fla.: An amendment prohibiting new data centers from being approved, built, expanded or operated in the city passed 77% to 23%.
- Kansas: Emporia (60%) and El Dorado (71%) approved bans on what both measures call “high impact” data centers, and El Dorado’s also covers Type 2 and Type 3 battery energy storage systems. Ottawa (72%) barred data centers using more than 25 megawatts of power.
- Maine: North Berwick banned data centers of 25,000 square feet or more. Wells, Peru and Wiscasset approved temporary moratoriums of 180 days or six months.
- Janesville, Wis.: Voters approved, 63% to 37%, a requirement for a public vote on any development of the former GM/JATCO site costing more than $450 million.
- Boulder City, Nev.: A measure to allow data centers in the Eldorado Valley Transfer Area failed, 71% to 29%.
Illinois had seven measures, all advisory. They point the same way: Joliet voters rejected, 79% to 21%, a question asking whether the city should approve AI data centers. Bloomington backed a moratorium on large-scale data centers, 78% to 22%. In four Madison County townships, 74% to 83% of voters rejected questions supporting new data centers.
Why it matters for real estate
Data centers have become one of the main sources of demand for industrial land and power-ready sites. They are a growing line of business for brokers, too: Cushman & Wakefield said Nov. 4 that data centers were part of the industrial leasing strength behind its third-quarter Americas leasing growth. Each local ban or moratorium takes land off the table for new projects. On our reading, that will push developers toward jurisdictions that still welcome them and make zoning risk a bigger part of land pricing.
Local resistance was already measurable before the vote. Data Center Watch estimated that local opposition stalled $68 billion of projects in the second quarter, and a pre-election poll put local opposition to data centers at 69%. Communities have also used moratoriums outside the ballot box. In August, Indianapolis finalized a moratorium through 2027. Ballotpedia notes that in July, New York Gov. Kathy Hochul issued what it describes as the first state-level moratorium on data center construction, barring state permits for new data centers expected to use 50 megawatts or more. As of April 2026, it says, 38 states offered tax incentives to attract the facilities.
On our reading, the results say more about who turned out on these questions than about national demand for computing capacity, which these votes do not measure. Many of the jurisdictions are small; in Monroe and Sprigg townships in Ohio, fewer than 700 ballots were cast on each zoning measure. But the margins were wide and the direction was uniform across very different places, from a Los Angeles-area suburb to Kansas towns and Maine coastal communities. For site selectors, that is now a documented pattern rather than an anecdote.
More votes are coming. On Dec. 1, voters in Clarksville, Ark., decide whether to uphold two ordinances authorizing up to $50 billion and up to $5 billion in taxable industrial development bonds for data center projects. On Dec. 15, Mandan, N.D., votes on whether to ban data centers using 25 megawatts or more. More coverage is in our technology section.
