FL SB 484 / signed

Why it matters

Four product-relevant items, layered. (1) This is the first US *state-level statute* (not utility tariff filing, not commission-approved rate schedule) putting the 'large load = own cost of service' principle into binding law. The ≥50 MW threshold is meaningfully lower than the ≥100 MW thresholds Ohio AEP and Indiana NIPSCO have used in their tariff filings, so Florida captures a much wider tail of mid-size data-center proposals — anything roughly 17 MW IT load and up (figuring 50 MW total facility = ~17 MW IT × ~3x for cooling/power overhead at modern PUE) is now in scope. Cliff's regulatory-knowledge graph needs Florida added to the 'large load tariff' state cluster (currently OH, IN, OK, MD-pending, VA-debating) at a *statutory* level rather than a tariff-filing level — the legal hierarchy matters because statute is much harder to undo than a rate schedule and applies to every Florida-regulated utility (FPL, Duke FL, Tampa Electric, Gulf Power) automatically. (2) The July 1 effective date stacks on top of the Virginia DEQ APG-576 July 1 cutoff covered yesterday, meaning two separate state regulatory regimes flip on the same day across two of the largest data-center markets. The submission-validator product needs a *Florida tariff applicability* check by August 1 (after PSC promulgates the implementing tariff over the summer) — does this site's anticipated peak demand trigger the 50 MW threshold, has any colocation arrangement at the same site been disclosed, and what's the corresponding cost-of-service exposure? Florida data-center pipeline is real and growing — the JEA / Jacksonville cluster, the Lakeland campus inventory near Polk County, and Miami-area colocation are all potentially in scope. (3) The 12-month NDA authorization is structurally interesting in two directions. Pro-applicant: it gives developers the political cover Texas Ch. 313 used to provide for negotiated abatement deals — a county can negotiate substantively without exposure to immediate public comment cycles. Anti-applicant: it lengthens the 'true filing date' uncertainty for any site under NDA, since the applicant's regulatory clock doesn't start ticking publicly until the NDA expires or is voluntarily released. Cliff's submission-readiness product should add an 'NDA-status' input to the Florida site model. (4) DeSantis's 'watered down' framing matters for forecasting — the original SB 484 was more restrictive (broader local authority, lower MW threshold under one draft, and stricter PSC mandates). The signed version's compromises were largely industry-driven (Florida Internet & Television Association, Data Center Coalition, big-tech direct lobbying), which means a 2027 follow-on session targeting the gaps — particularly the NDA shield and the 50 MW threshold — is now a high-probability event. Florida statutory regime is unstable in the direction of *more* restriction, not less, and Cliff's customers should plan for a 2027 SB 484 amendment cycle.