In two weeks, the largest regulatory proceeding over data-center interconnection reaches its first real deadline. On November 16, 2026, five of the six regional grid operators — PJM, MISO, CAISO, ISO New England, and NYISO — must file responses to the Federal Energy Regulatory Commission’s June 18 show-cause orders on large loads. Southwest Power Pool follows on November 20, with answers due 30 days after each filing. What lands on FERC’s docket will not, by itself, connect a single data center. It will define, region by region, the application a large load must file, the study it must pass, the costs it must secure, and the service it can buy while its upgrades are still being built.
The six orders are often described as a single federal push on data centers. They are not: they are six separate section 206 cases, each directed at one operator and its transmission owners, each tailored to that region’s tariff, and each capable of producing a different answer. The common deadline is real. A common tariff is not on offer.
What a show-cause order is, and is not
A show-cause order is a demand for an answer, grounded in a preliminary finding — not a final one. In each of the six orders, reported at 195 FERC paragraphs 61,211 through 61,216, FERC preliminarily found that the operator’s tariff appears unjust, unreasonable, or unduly discriminatory because it lacks adequate provisions for large loads. The dockets are PJM (EL26-67), SPP (EL26-68), NYISO (EL26-69), MISO (EL26-70), CAISO (EL26-71), and ISO New England (EL26-72). Each operator must show cause why its tariff remains just and reasonable, or propose revisions remedying the problems FERC identified.
The original clock was 60 days, which would have put responses due August 17. Every operator instead sought, and received, a pause of about 90 days — an abeyance — to run its proposal through a stakeholder process and file it as a tariff revision under section 205 instead. These filings are therefore operator proposals, shaped by months of stakeholder negotiation, not FERC-imposed rules. FERC will then decide whether each satisfies the show-cause order; until it acts, the existing tariffs remain the operating reference. FERC also kept open the rulemaking docket begun by the Department of Energy’s October 2025 request, RM26-4.
Why FERC opened the cases
The orders grew out of a problem this publication’s interconnection explainer identified as the load-side twin of the generation queue: queues built for generators suddenly faced single sites demanding hundreds of megawatts, or gigawatts, often on timetables measured in months and at multiple sites at once. Operators had no reliable way to tell which requests were real, no consistent study process for loads of that size, and no settled rule for transmission upgrades if the load shrank, delayed, or disappeared after the wires were built.
FERC’s findings converged on the same diagnosis in each region: study processes slow, unclear, or designed for smaller customers; tariffs that did not protect other customers from cost shifting when upgrades for a large load ended up stranded; missing or ambiguous rules for co-located load and behind-the-meter generation; no service for a load willing to limit its own use for faster access; and no process for studying a generator together with the nearby load it was built to serve. The orders also required 30-day resource adequacy reports, filed in July.
The five questions every filing must answer
Each tailored order identifies five categories of reform. First, application and study: is there a defined large-load path, with timelines, deposits, and readiness screens separating real projects from speculative ones, and serious consideration of grid-enhancing technologies? FERC suggested study timelines of 60 to 90 days, a benchmark several regions adopted. Second, cost shifting and transparency: what prevents upgrade costs for one large load from landing on everyone else, and what can the public see? FERC’s Office of Public Participation put the gap bluntly: the tariffs did not publicly post which upgrades were built for which loads and what they cost. Third, co-location and behind-the-meter generation: what are the rates and terms when load sits behind the same point of interconnection as a generator, and how is its gross demand treated when its own generation runs? The netting question — whether a co-located load can subtract on-site generation from its demand — is one of the sharpest disputes, answered differently by region. Fourth, flexible service: what can a load buy if willing to be curtailed — interim service before upgrades are complete, non-firm service as a permanent product, or service conditioned on bringing new capacity? Fifth, proximate generation: can a nearby generator be studied with the load, so injection and withdrawal offset each other? FERC has said the proceedings are prospective, not intended to disrupt existing arrangements.
PJM: the most watched filing, and the narrowest order
PJM’s filing is the most watched of the six, and its order is the narrowest. Because FERC had already acted on substantial parts of PJM’s large-load framework — including co-location, addressed separately and again on June 18 — the PJM order carries a shorter directive list than the template orders and is limited to large loads not co-located with generation. PJM’s stakeholder materials, posted October 29, say it will make several section 205 filings on or about November 16: one on large-load ride-through requirements, another on the bulk of the reforms, including new transmission services. Their precise number and structure remain subject to change.
The context includes an October ruling, in a separate docket, that draws a line the filing must respect. FERC approved PJM’s registry of large loads — defined there as cumulative peak demand of 50 megawatts or more at a single site — but rejected its August interim-service proposal, under which a new large load without matching capacity could connect early in exchange for being curtailed before other demand response. FERC’s objection was jurisdictional: the plan put obligations directly on new large loads, mostly retail customers, and effectively picked which retail customers would be curtailed first — choices that belong to the states.
MISO: a joint study, with the cost question still open
MISO’s answer is the furthest along in public, and the most incomplete in one respect. In an August 28 filing, it proposed interconnection reliability requirements for large loads — visibility, ramp-rate, and disturbance ride-through rules — defining a large load as 50 megawatts or more at a single substation, with a computational-load carve-out at 25 megawatts of IT-equipment demand, seeking a December 4 effective date. Then, on October 2, it proposed the 120-day Large Load Addition Resource Study, or LARS, studying a load above 200 megawatts together with new generation in the same local resource zone, in parallel, with both tied together for 15 years.
MISO has said cost-shifting protections are still to come in its November 16 response. That is the gap to watch. LARS answers the study question elegantly, but not who pays if the pair dissolves, the generation retires early, or the load fails to materialize. MISO’s filing must supply the cost recovery agreement, financial security, and transparency provisions FERC demanded, or explain why its existing tariff already does.
SPP: the head start, and the extra four days
SPP is the exception that clarifies the rule. It entered the proceeding further ahead than any other operator, and its deadline is four days later — November 20 — because its joint abeyance motion with its transmission owners asked for 95 days. SPP already had a High Impact Large Load process (HILL) and a companion generation assessment (HILLGA), accepted effective January 15, 2026, pairing a large load with shovel-ready generation and studying them together. In June, FERC accepted a further piece, Conditional High Impact Large Load Service (CHILLS), effective July 1, 2026, letting a very large load take non-firm, interruptible service for up to seven years while firm generation and upgrades are lined up.
SPP’s filing therefore has a different job: less invention, more completion. The questions are whether HILL, HILLGA, and CHILLS, with its pending peak-demand assessment proposal, satisfy the order’s remaining items — above all the pro forma cost recovery agreement and transparency provisions. CHILLS is the most generous early-access product in any region, and the most exposed: SPP can curtail a CHILLS customer when the system is constrained or in emergency conditions, with no tariff cap on how often or how long curtailment can last.
CAISO: board-approved, and first across the line
CAISO is the first operator to carry a proposal through board approval. On October 28, its Board of Governors approved a package it will file with FERC by November 16; president and chief executive Elliot Mainzer described the aim as better planning, cost-shift prevention, flexibility, and faster energization. The centerpiece is a flexible interim load interconnection: a large load — an end user with peak capacity of 50 megawatts or more — can connect before its network upgrades are complete, with flexibility requirements set by grid conditions until those upgrades are finished, and with the operator able to remotely disconnect the load when needed for reliability.
Transmission owners would be required to set tariff terms preventing cost shifting. Co-located generation serving only the facility would be exempt from an ISO study, while generation injecting into the grid must complete one, and co-located loads must be metered independently so they pay ancillary service charges. A permanent flexible option, and arrangements for electrically proximate generation, are deferred to 2027. The test for FERC will be whether the interim stage is bounded tightly enough that it does not become the permanent rule by default.
ISO New England: three services, one price trigger
New England’s proposal is the most fully specified service menu in the proceedings, still moving through stakeholder votes ahead of a November 16 filing, after committee votes on November 3 and a Participants Committee vote on November 12. It replaces a single service model with three — Firm, Interim, and Non-Firm Contract Demand Service — in any combination fully serving a large load at its point of delivery. A large load is a commercial or industrial customer at a single site with gross peak demand of 50 megawatts or more, interconnecting above 60 kilovolts; an existing load becomes one if its peak grows by 5 megawatts or more.
Firm service requires the load to designate new capacity resources of its own, and large loads are removed from the region’s Installed Capacity Requirement. Interim service is temporary and non-firm — a bridge while upgrades are finished or designated capacity comes online — then converts to firm. Non-firm service is permanent, skips the bring-your-own-capacity requirement, and is studied around the load’s flexibility so fewer upgrades are needed. While on interim or non-firm service, a load can be re-dispatched when its local price exceeds a proposed $250 per megawatt-hour. Applications would carry a $50,000 fee, a $150,000 study deposit, and a readiness deposit of $10,000 per megawatt, with studies meant to finish in 90 days. The proposal targets June 1, 2027 for applications, with market-rule implementation June 1, 2028.
NYISO: adequacy first, and a new participation model
New York’s filing starts from a different problem statement. NYISO’s tariff already had load interconnection procedures and a distinctive financial-reservation model, but FERC preliminarily found the study process insufficiently sophisticated, the application process in need of enhancement, and the netting of co-located generation against co-located load potentially unjust and unreasonable. Its schedule targets a section 205 filing on November 16.
The proposal has three elements worth watching. First, a Large Load Integration Procedures track with readiness deposits and queue-withdrawal rules to filter speculative requests. Second, withdrawal services — firm, non-firm, and zero withdrawal — separating the right to take power from the right to sell capacity, with non-firm loads required to curtail when directed and without compensation. Third, a co-located load aggregation model treating load and generation separately rather than netting them, with a 15-year term proposed for resources partnering with loads seeking firm withdrawal service. New York is also where adequacy is most explicit: the ISO has told FERC its studies show a risk of shortfalls as older units retire and demand grows.
What it means for timelines, and for existing customers
For developers, the honest summary is that the filings compress the study phase, not the construction phase. A 90-day study in New England, a 120-day joint study in MISO, a monthly review cycle for CAISO load upgrades — all give an answer sooner. None pours concrete faster; the upgrades identified still take years to build. What the new services change is the waiting: interim and non-firm products let a load energize, at least in part, before the full upgrade set is complete — in SPP’s case for up to seven years — in exchange for accepting curtailment. The question for any site is no longer only how long the queue is, but what service it can buy while it waits, how often it can be interrupted, and what it must bring to convert interim service into firm service.
For existing customers, the filings determine whether the next large load’s costs stay with that load. The mechanism FERC keeps returning to is the cost recovery agreement — a standard, pro forma agreement among the operator, the transmission owner, and the transmission customer taking service on behalf of the large load, backed by financial security, obliging the load to cover the costs incurred to serve it, including network upgrades, even if the project is canceled or operates below its anticipated demand. Without one, an upgrade built for a load that never arrives sits in the rate base and is paid for by everyone else. Transparency is the companion protection: a posted record of which upgrades were built for which loads, at what cost and assigned to whom, gives state commissions the evidence to allocate costs among retail classes. One limit should be stated plainly: even the best agreement is only as strong as its enforcement.
What to watch after the filings land
Three things will determine whether the November filings become working tariffs or another round of paper. The first is the answer period, 30 days after each filing, testing the weakest joints — MISO’s cost provisions, PJM’s jurisdictional line, the $250 trigger in New England, interim-service duration in California and SPP, and the netting rules in New York. The second is FERC’s disposition: it can accept a filing, accept it subject to condition, set it for hearing, or prescribe a remedy itself. The PJM experience this autumn — a registry accepted, a curtailment plan rejected on jurisdictional grounds, a backstop procurement suspended for five months — is a reminder that FERC does not move on the applicant’s schedule. The third is the difference between a study and a connection: even the best filing will take months to become an effective tariff, and its upgrades years to build. The filings will be judged not by how quickly they promise power, but by whether the loads they admit are real, the costs they assign stay assigned, and existing customers are no worse off.