Votes Are Now Underwriting Inputs

The Signal

Florida voters decide Amendment 3 today, November 3. Approval requires at least 60%. If approved, the proposal would reduce the annual assessment-growth cap for non-homestead property, including commercial buildings, from 10% to 5% for non-school levies beginning January 1, 2027. It would not automatically reduce every commercial property’s tax bill.

The enhanced non-school homestead exemption would increase to $150,000 in 2027 and $250,000 the following year. Newly qualifying homeowners would receive a smaller $50,000 exemption during their first four exemption years and the enhanced exemption in the fifth. This is not five years without any exemption.

The state Revenue Estimating Conference’s July 10 fiscal analysis estimates local revenue losses of approximately $4.93 billion in fiscal 2027–28, $8.71 billion in fiscal 2028–29 and $11.83 billion by fiscal 2031–32. These are modeled government revenue effects, not guaranteed increases in a particular owner’s tax bill.

A study commissioned by housing and apartment groups estimates an average 14.8%, approximately $406 per apartment unit annually, in tax year 2028 if local governments replace the revenue loss through millage increases, subject to rate limits. Keep both the scenario and its sponsorship visible.

Separately, Ballotpedia lists 41 local data-center-related measures across 10 states for November 3. That is a count of measures, not distinct municipalities. The questions include bans, pauses, conditions, advisory questions and pro-development votes. They cannot be treated as 41 identical decisions on construction.

Our Read

Taxable value, millage and entitlement status deserve explicit assumptions in a property model. A tax bill generally equals taxable value multiplied by millage and divided by one thousand. Exemptions affect taxable value; an assessment-growth cap addresses a different part of the calculation. A lower cap does not guarantee a lower bill.

For an existing Florida owner, test unchanged millage and a supportable revenue-replacement scenario using the correct exemptions and non-school levies. For a buyer, verify the acquisition’s assessment treatment rather than assuming a prior owner’s cap benefit carries over. The proposed law and local rate decisions must be analyzed separately.

For data-center land, power and capital are necessary inputs, but neither establishes entitlement. Commercial Property Executive’s November 2 ULI report describes an audience poll that initially placed power at 50% and community opposition at 38%; opposition ranked first at the end, with final percentages unreported. That is conference color, not representative market data. The specific ballot text and permit path are the decision-relevant evidence.

Texas’s September 21 TCEQ directive is earlier policy context, not a new November action. The Texas Tribune’s October 19 tally of approximately $9 million in Abbott fundraising involves direct and indirect data-center ties, not developers alone. State incentives and local taxes vary by jurisdiction; do not assume universal coverage or exclusion.

Stakeholder Lens

Florida multifamily owners can use the 2028 study as one stress scenario, not their base-case forecast. Net-lease landlords should check tax pass-through terms and caps. Lenders should ask which levies, exemptions and millage assumptions sit behind projected coverage. Data-center developers should identify the exact local measure, zoning status, permit exposure and tax treatment before committing land.

Neither a favorable ballot label nor a national tally replaces property-level diligence. The practical response is to show alternative cases and the source of each assumption, while leaving the election outcome open.