The 45-day abeyance window inside FERC's six show-cause orders becomes the next hard deadline — PJM is the only RTO with a credible path to use it (because of the Dec 2025 co-location order already on a July 31 effective date), the moratorium streak hits day nine with La Crosse County WI 26-2 for an 18-month pause, and the streak gets its first counter-signal in nine days with Clark County NV unanimously approving Switch's LAS 19.

Why it matters

Two implications for the corpus and for site-readiness timing. (1) The interesting filing date is not Aug 17 anymore — it is Aug 2. The first RTO to file a credible abeyance request sets the template every other RTO can either copy or distinguish itself from. The corpus should be monitoring all six RTO docket pages every business day starting now, not waiting for the 60-day mark. (2) For sites in pre-approval right now, the question 'what tariff regime will my interconnection cost-allocation be litigated under' has a binary fork: if the local RTO files for abeyance, the answer is whatever Section 205 tariff they propose; if not, the answer is whatever FERC orders under 206 after a contested proceeding. Those two paths produce different cost-allocation outcomes on different timelines and developers/hyperscalers should be tracking them per project.

Why it matters

Two near-term consequences. (1) PJM is the most-watched docket of all six for the next six weeks — and probably the only one where the substantive tariff debate happens through a Section 205 process rather than a Section 206 process. That changes who participates and who has the burden of proof, which means the regulatory filings that will end up shaping data-center cost allocation in PJM look more like industry-driven tariff design than FERC-driven enforcement. The corpus has to track those filings differently than it would the other five. (2) The July 31 effective date PJM requested back in February is now the de facto reference deadline for the federal cost-allocation answer in the largest data center market in the country — which is two and a half weeks before the August 17 default show-cause deadline. A developer trying to forecast cost-allocation exposure for a Northern Virginia / PJM site should plan to know the answer by August 1, not August 17.

Why it matters

Two corpus implications. (1) The 18-month + study-committee bundle is a different artifact from a standalone 180-day pause and should be tagged differently in the docket database — these are the moratoriums most likely to convert to a permanent zoning ordinance rather than expiring. Per-jurisdiction fields should include 'study committee chartered = y/n,' 'study committee deliverable date,' and 'auto-expire or affirmative-renewal-required.' (2) The vote margin (26-2, ~93%) is the kind of structural data the corpus needs to capture for predictive purposes. Yesterday's note flagged hearing-attendance-as-percent-of-population; vote margin among the governing body is the other half of that signal. A bundle of jurisdictions with 90%+ vote margins is materially less likely to reverse course than a bundle with 55-45 margins, and that should be queryable directly — 'show me every active moratorium with a vote margin under 60%, ranked by expiry date' is a strategic surface for developers looking for sites where reversal is plausible.

Why it matters

Two pattern implications for what 'survives' the moratorium wave. (1) The Clark County approval pre-removed water by design before the vote was held. That is the most concrete data point in the corpus for what design choices make a project moratorium-resistant — and it suggests cooling architecture (dry vs evaporative vs hybrid) should be a per-project field in the database, not buried in narrative summaries. A search like 'show me every proposed data center in a jurisdiction with an active or recently-debated water-concerns moratorium, ranked by cooling architecture' would surface exactly the sites that are most likely to be either approved or contested. (2) One counter-signal in nine days is not a trend reversal; it is a data point showing the opposition is real but project-specific, not categorical. The corpus should not treat every jurisdiction in 'data-center-debate' status as a blanket pause — some, like Clark County, will continue to approve projects that pre-resolve the politically salient opposition issues. Tagging jurisdictions as 'active pause / active scrutiny / open' with a default of 'open' rather than 'pause' is the calibration the data implies.

Why it matters

The strategic frame change after June 18. Before yesterday's FERC order, NC SB 730 was a state-level bill on a parallel track to whatever FERC was going to do. After June 18, NC SB 730 is the template for the work FERC explicitly said should happen at the state level rather than the federal level. That recasts it from a state-by-state legislative story into the visible front-edge of the post-FERC state regulatory build-out, and it raises the value of tracking similar bills across every other state, especially the Duke-served and TVA-served states (NC, SC, GA, TN, FL) that are not in an RTO and therefore did not get a Section 206 show-cause clock. Those states are the ones where the cost-allocation question is structurally a state-only question. The corpus should be running a daily diff against the legislative trackers in every state that is either outside RTO jurisdiction (Duke, TVA, Southern Company, vertically-integrated Western states) or partially inside one (NC, KY, IL) and surfacing every introduced bill that touches data center cost-share, large-load tariffs, or local-incentive prohibitions. That diff is what the next two months of state-level news will be made of.