Grid-to-Chip P2 · Grid Access: connection rights, regulatory regimes and the transmission side

TL;DR · P2 · Grid access

• The real P2 asset is MW × node × date × probability of energisation. A megawatt is not an asset merely because someone has contracted it. The value depends on the node it can actually reach, the date it can reach it, and the probability that the upgrade path clears. Different markets wrap this in different legal instruments (GIA, transmission or distribution connection agreement, TSO/DSO connection, CTU/STU connectivity, state grid allocation, utility allocation). The economic object is the same.
• More than 2,500 GW of renewable, storage and large-load projects are currently stalled in grid-connection queues worldwide. The IEA's 2026 electricity outlook places the global aggregate at that scale, spanning supply, demand and storage in parallel. P2 in the Grid-to-Chip framework covers this global problem once "interconnection" gets read as "grid access" across regimes.
• Regime archetypes travel across borders better than a country-by-country tour does. Five archetypes cover the ground: ISO or RTO markets, vertically-integrated utilities, TSO/DSO connection regimes, state or grid-company allocation, hybrid or dedicated infrastructure. Each has distinct scarcity mechanisms and distinct vendor sets. The article uses these archetypes rather than a country-by-country walk.
• Every regime now carries three P2 sub-problems on the active docket. Generator interconnection puts a resource onto the grid. Large-load interconnection connects the AI campus as a load. Co-location and behind-the-meter interfaces cover generation plus load at a shared delivery point. The AI-buildout pushed all three onto regulators' active dockets across the major markets in parallel through 2024 to 2026.
• Two paths deliver grid access globally: building more of it, or extracting more from what already exists. Build-more-grid (transmission, substation, transformer, new HVDC) remains the base case but faces multi-year physical constraints. Extract-more (dynamic line rating, power-flow control, non-firm connections, storage, flexible load, grid-forming resources) is the accelerator path. IEA estimates 1,200 to 1,600 GW of advanced-stage projects could unlock through non-firm connections and grid-enhancing technologies. This shifts the weight investors should place on Tier 3 emerging platforms relative to the incumbent Tier 2 build-path vendors alone.
• The US case study runs deepest because the US produces the sharpest public evidence. FERC Order 2023, the June 2026 large-load integration action, the December 2025 PJM co-location directive and the June 2026 PJM Expedited Interconnection Track anchor a visible regulatory cycle. PJM Cycle 1 study population is approximately 220 GW; ERCOT reports approximately 474 GW of large loads seeking interconnection.
• The UK's TMO4+ connections reform provides the closest non-US analogue. Ofgem and NGESO have moved the transmission connections queue from first-come-first-served toward first-ready-first-connected. Ofgem reported approximately 315 data-centre projects representing about 73 GW of demand in the UK connection queue, with commitment fees and hard milestones proposed to filter speculative projects out of scarce connection capacity.
• Continental Europe reads P2 as systems integration for large loads. ENTSO-E has increasingly treated data centres as system-relevant electricity users, including their potential flexibility and grid-support role. EU Grid Action Plan (November 2023) sits above member-state TSO reforms at RTE, TenneT, 50Hertz, Amprion, Terna and Redeia. The binding question is how rapidly-growing large load fits into an already-constrained transmission system.
• APAC and MENA route through allocation-heavy regimes with sovereign infrastructure. Wood Mackenzie estimates roughly 32 GW of planned APAC data-centre capacity across more than 1,150 projects; developers are increasingly required to provide storage, accept curtailment risk and support grid stability. India's CEA advanced Green Energy Corridor III and ISTS waiver extensions. UAE and Saudi Arabia route P2 through utility allocation, with NEOM and UAE AI-cluster developments as hybrid or dedicated infrastructure.
• Investable vendor exposure at P2 spans three tiers, with global names in each. Tier 1 counterparties (utilities and transmission owners) span US, UK, Continental Europe, APAC and MENA. Tier 2 equipment concentrates at Hitachi Energy, GE Vernova Grid Solutions and Siemens Energy Grid Technologies, with Toshiba T&D, Mitsubishi Electric, Hyundai Electric, BHEL, CG Power, TBEA and XD Group participating regionally; Prysmian, Nexans, NKT, Sumitomo Electric and LS Cable & System on cables. Tier 3 emerging platforms cover grid-enhancing technologies, grid-forming inverters and DER aggregation.
• A handful of specific events would reshape P2 pricing over the next 24 months. PJM Expedited Interconnection Track first-cohort deliveries. Ofgem TMO4+ implementation outcomes. EU Grid Action Plan project delivery through 2027. India ISTS waiver renewal. Chinese vendor share dynamics in HV grid equipment globally. HV transformer supply-chain evolution. The IEA 2027 electricity outlook queue-depth update.

Reader shortcut: the six-question decision tree in Section 16 is a compact underwriting framework for a P2 exposure. New readers may find it useful to skim that section before returning to the full analysis.

Framework anchor

This piece is positioned at P2 in the Grid-to-Chip framework. Upstream: P1 · Electrons, where the contracted energy stream originates. Downstream: P3 · Campus power, where the interconnected energy meets the campus substation and steps down to the medium-voltage distribution stack. P1 and P2 are tightly coupled; P2 and P3 are tightly coupled through the delivery-point boundary.

01What P2 actually covers

P2 in the Grid-to-Chip framework is the grid-access layer. The economic unit at P2 is the contractual and physical right to connect power to or from the grid at a defined location, capacity and date. That right takes different legal instruments in different markets. For a generating resource in a US ISO or RTO market it takes the form of a Generator Interconnection Agreement (GIA). For a large load in Great Britain under NGESO it takes the form of a transmission or distribution connection agreement. For a Continental European connection point it takes the form of a TSO or DSO connection agreement with ENTSO-E coordination on the transmission side. In India it takes the form of CTU or STU connectivity. In China it takes the form of state or grid-company allocation. In the Middle East it takes the form of utility allocation with sovereign infrastructure planning behind it. The underwriting question does not change across regimes: can the required MW physically reach the project at the required location on the required date, through an economically allocated upgrade path.

P2 covers three related regulatory sub-problems that share the same delivery-point real estate. Generator interconnection (P2-A) puts a generating resource onto the grid. Large-load interconnection (P2-B) connects the AI campus itself as a load. Co-location or behind-the-meter interfaces (P2-C) sit generation and load at the same delivery point. The 2024 to 2026 AI-buildout pushed all three onto the regulator's active docket at the same time, and the pattern is visible across every major market; US-specific framing understates the spread.

P2-A

Generator interconnection

A generating resource injects into the grid under a Generator Interconnection Agreement (GIA) between the generator and the transmission provider. The queue, the cluster study, the network-upgrade allocation and the cost recovery all live here. FERC Order 2023 (July 2023) reformed the process in the FERC-jurisdictional regions.

P2-B

Large-load interconnection

The hyperscaler campus itself connects as a large load. The instrument is a load-service arrangement with the local utility, or an interconnection agreement specific to load facilities where the tariff structure supports one. This layer was managed as routine utility business until the AI-buildout demand spike pulled it onto the regulatory front. FERC launched an aggressive large-load integration action in June 2026 across all six FERC-jurisdictional regional grid operators.

P2-C

Co-location and behind-the-meter interfaces

A generator and a large load share the same delivery point (or the load is positioned directly behind the generator meter). Reliability, transmission cost allocation and tariff treatment all become live questions. FERC directed PJM in December 2025 to establish transparent rules for AI data centres and other large loads co-located with generation.

The three sub-problems occur at the same physical location and involve overlapping counterparties, but they are governed by different tariff sections, different regulatory filings, and often different commercial and legal teams inside each utility. Any P2 investment analysis that reads only one of the three misses the way the other two drive delivery risk and cost allocation on the specific project. The queue-quality material in Section 04 addresses P2-A specifically; Sections 05 and 07 cover the 2024 to 2026 regulatory events across all three sub-problems.

The underwriting matrix across the three sub-problems

The invariant P2 underwriting question decomposes into specific factors that behave differently across the three sub-problems. A P2 exposure with a specific mix of the three sub-problems (or a portfolio of them) prices on these axes.

The matrix holds across regimes. A generator interconnection in the US ISO/RTO regime, in the UK NGESO regime, in a Continental European TSO regime and in India's CTU regime has the same underwriting factors under different legal wrappers. A large-load interconnection has the same factors in its regime-specific form. Co-location interfaces have the same. The invariant P2 asset is MW × node × date × probability of energisation. Every P2 exposure prices somewhere on that four-dimensional surface; the matrix above says which cell.

Fig P2.A

The P1 to P2 to P3 handoff: where the three P2 sub-problems live

The contracted energy stream from P1 becomes a physical injection at P2-A, meets a load-service connection at P2-B, and (when generator and load share the delivery point) crosses the P2-C interface before reaching P3.

P1 · ElectronsContracted energy streamPPA, tolling, offtakeframework, utility tariff(nuclear, gas, renewables)P2 · Grid accessThree sub-problemsP2-A generator interconnectionGIA, queue, network upgradesP2-B large-load interconnectionLoad service, tariff, upgradesP2-C co-location / BTM interfaceShared delivery point rulesP3 · Campus powerHV to MV, distributionSite substation, MVtransformers, SST, MVswitchgear, ring-mainThree P2 sub-problems apply within every regulatory regimeEach sub-problem has its own tariff sections, regulatory filings and commercial counterparties

Simplified representation. Actual campus deployments can involve all three sub-problems simultaneously (a generation project injecting at a delivery point where the hyperscaler load also connects, potentially with co-location provisions). The FERC and state PUC filings for each sub-problem move on their own schedules.

The five axes of any P2 arrangement

A P2 arrangement, regardless of which sub-problem governs it, prices on axes.

Queue or study stage. Position in the cluster study, system-impact study, facility study or executed-agreement phase. An executed agreement with completed upgrades holds fundamentally lower delivery risk than an active study.

Network upgrade cost allocation. Who pays for the upgrades required to accommodate the injection or the load. Cost allocation between the interconnecting party and the broader transmission user base has been the most contested question in FERC transmission proceedings across the last decade.

Cure milestones. The specific project readiness milestones (site control, air permits, offtake commitment) that must be satisfied to hold the queue position and progress through study phases. Post FERC Order 2023, cure milestones and commercial-readiness deposits are the primary mechanism to filter speculative queue entries.

Withdrawal penalties and deposits. Commercial readiness deposits under Order 2023 change the economics of speculative queue positions. Deposit-and-penalty structures apply differently in load-service arrangements at P2-B.

Transmission-provider counterparty. An ISO/RTO in the organised markets differs from a vertically-integrated utility in a bilateral region on tariff structure, study cadence and dispute-resolution process. In non-FERC jurisdictions (ERCOT, non-US markets) the counterparty framework changes entirely.

02Why P2 has become the global bottleneck

The International Energy Agency's 2026 electricity outlook places more than 2,500 GW of renewables, storage and large-load projects in grid connection queues worldwide. That aggregate is roughly the same order of magnitude as installed generation capacity in the entire US electricity system. The queue crisis spans global markets; US-specific framing understates the spread, and grid capacity now is positioned as the shared bottleneck across supply (generation queues), demand (large-load queues) and storage (interconnection requests) in parallel.

Composition of the queue varies by region. US ISO queues are heavy on solar plus storage with growing gas and large-load requests. UK NGESO queue has substantial offshore-wind, solar and battery storage plus a fast-growing data-centre load component. Continental European queues concentrate on renewables (offshore wind, solar) with increasing data-centre and industrial-electrification demand. India's queue skews toward renewables plus storage with growing captive and open-access load structures. China's queue runs through provincial and state grid company processes and is less publicly transparent. What is common across all of them: interconnection queue expansion has outpaced transmission capacity expansion, and neither has kept pace with the AI-buildout load growth.

The regulatory response has taken different forms in different regimes but the substantive content converges. Every major market is running some combination of (a) queue reform to filter speculative entries, (b) new frameworks for large-load participation and (c) explicit treatment of co-located generation and load. FERC Order 2023 in the US, Ofgem plus NGESO TMO4+ in the UK, the EU Grid Action Plan, ENTSO-E's data-centre integration frameworks, India's CEA Green Energy Corridor III and ISTS reforms, Australia's AEMO Integrated System Plan refresh: same P2 problem, different legal wrapper.

Fig P2.0

Global connection-queue estimates by region and category, 2026

Approximate queue figures across major markets. Definitions vary by region and by generation vs load category; the chart is a visual composite; a directly comparable table would require harmonised regime definitions. IEA aggregate estimate provides the global scale.

Sources: IEA Electricity 2026 report cites the >2,500 GW global aggregate; PJM Inside Lines for the PJM Cycle 1 2026 figure; ERCOT Senate testimony July 2026 for the ERCOT large-load figure; CAISO year-in-review 2025 for Cluster 15; Ofgem 2026 connection-queue disclosures for UK; Lawrence Berkeley National Laboratory for the US-wide gen+storage aggregate at end-2025; Wood Mackenzie APAC data-centre 2026 estimate for APAC data centres. Continental Europe TSO estimate, India renewables plus load estimate and UAE + Saudi dedicated-infrastructure estimate are adikumar.co composites synthesised from ENTSO-E TYNDP 2024, national TSO disclosures, India CEA reports and regional utility disclosures respectively; treat as directional analyst estimates, not audited figures. Reporting frames differ significantly across sources; the chart is a visual composite intended to convey scale, not a directly comparable dataset.

03The P2 regulatory regime archetypes

Global P2 organises cleanly into regulatory regime archetypes; fifteen country-by-country stories would obscure the structural pattern. Each archetype has a distinct legal wrapper for the underlying grid-access right, a distinct vendor set on the equipment side and a distinct scarcity mechanism. Regional characteristics vary within each archetype.

Regime A

ISO or RTO market

Non-profit market operators run interconnection studies and manage the queue for a multi-utility footprint. Wholesale market clearing is positioned above the interconnection layer. US six FERC-jurisdictional regions (PJM, MISO, SPP, CAISO, NYISO, ISO-NE) run under FERC Order 2023 process reforms. ERCOT operates outside FERC jurisdiction under the Texas Public Utility Commission. Parts of Australia (AEMO for the NEM) and Canada (Ontario IESO, Alberta AESO) run similar structures with local variations. Investable exposure runs through the transmission-owner utilities that participate as ISO members; direct ISO-entity participation stays the alternative channel.

Regime B

Vertically-integrated utility

A single utility handles generation, transmission, distribution and load connection through one entity, typically under state or national regulator oversight. US Southeast (Duke Energy, Southern Company, TVA, Entergy in its non-MISO footprint), parts of the US West outside CAISO, Middle East (DEWA in Dubai, EWEC and TAQA in Abu Dhabi, Saudi Electricity Company), parts of Latin America. Interconnection process runs inside the utility with regulator prudency review after the fact. Vendor competition at the equipment level is similar to the ISO regimes.

Regime C

TSO / DSO connection regime

Separate transmission and distribution operators, each with distinct connection agreements. UK (NGESO for national transmission, DNOs and iDNOs for distribution). Continental Europe (national TSOs plus DSOs at member-state level, ENTSO-E coordination on cross-border and long-term planning). Japan (ten regional utilities operating with OCCTO cross-region coordination). Investable exposure spans utility-parent equity, independent transmission ventures where they exist and the equipment supply chain.

Regime D

State or grid-company allocation

Grid connection through provincial or national grid-company decisions with integrated utility structure. China (State Grid Corporation of China across the north-east and centre-north, China Southern Power Grid across the south). Vietnam (EVN plus subsidiaries). Parts of Southeast Asia. Investment access is limited for global investors; the exposure runs through equipment suppliers (Chinese vendors that export globally, plus international vendors that participate in Chinese projects through joint ventures); grid-company participation stays the alternative channel.

The China-specific implication. China's P2 scarcity is less visible as a public queue and more embedded in provincial planning, grid-company allocation and project siting. Absence of a transparent queue does not indicate absence of scarcity; scarcity gets allocated administratively; a public process would be the alternative mechanism. The IEA's global >2,500 GW estimate includes a Chinese component that is reported in different terms from PJM Cycle 1 or Ofgem TMO4+ but represents the same underlying dynamic. Investable exposure to this regime is fundamentally different from Regime A or C: opaque process, integrated planning, limited direct equity access, but genuine scarcity that manifests in project-siting outcomes; queue reports stay the alternative evidence source.

Regime E

Hybrid or dedicated infrastructure

Sovereign or hyperscaler-anchored campuses where generation, transmission and load are planned together. Saudi NEOM (dedicated infrastructure programme with integrated generation plus transmission). UAE AI-cluster developments (specific infrastructure planning around AI-anchored campuses). Certain hyperscaler-anchored greenfield developments elsewhere. Vertical integration by design; regulatory legacy would be the alternative driver. Investable exposure runs through the specific project developers, EPC contractors and equipment suppliers involved.

The five archetypes overlap at the boundaries. Vertically-integrated utilities in the US Southeast host generation projects that connect similarly to an ISO regime; UK NGESO co-ordinates with DNOs in ways that resemble a US ISO/utility split; the UAE hybrid infrastructure has a utility-allocation component sitting behind it. The archetypes are analytical categories that structure the discussion; mutually exclusive classification would overstate the boundaries.

The rest of this article treats the US as the deepest investable case study, because the US produces unusually granular public queue data and hosts a highly investable transmission ecosystem. Case-study material on the UK, EU and APAC follows in Section 14. The vendor set in Section 06 covers all regimes with named international vendors.

04US case study: generator-interconnection quality spectrum

P2-A generator interconnection is unequal by project stage. A hyperscaler underwriting a P1 PPA at a specific delivery point cares less about whether interconnection exists in the abstract and more about where in the queue-to-COD pipeline the specific project is positioned.

Five quality archetypes for a generator project

1. Executed GIA with completed network upgrades. The gold standard. Injection capacity is available, the delivery point is defined, no material physical work remains. Legacy positions from queue cycles prior to the current AI-buildout demand surge dominate this bucket.
2. Executed GIA with pending network upgrades. The project has cleared its interconnection studies and signed its GIA, but transmission or substation upgrades are still under construction. Delivery timing depends on the upgrade scope and the transmission provider's construction cadence.
3. Post-facility-study, pre-GIA. All studies complete; commercial terms of the GIA are in negotiation. Typical residual issues at this stage are cost-allocation disputes and cure-milestone timing.
4. Active cluster study. The project is in one of the ISO's cluster-study windows. Typical time from cluster entry to executed GIA in the current cycle runs into multiple years, subject to specific project characteristics and ISO queue depth.
5. Pending queue entry or withdrawn. Not yet in an active study cycle, or previously withdrawn and awaiting re-entry. Delivery timing is a multi-year proposition even in optimistic scenarios.

Fig P2.1

US case study: generator-interconnection queue-to-COD pipeline by queue stage

Approximate ranges from current stage to commercial operation date. Actual timelines vary significantly by ISO, project size, network upgrade requirements and site-specific factors. Post FERC Order 2023 the process has cleaner deadline discipline.

Source: adikumar.co synthesis of ISO/RTO queue reports, FERC interconnection proceedings and public utility disclosures. Reflects Q3 2026 conditions in the US organised markets. Vertically-integrated utility regions can hold different (sometimes shorter) timelines depending on utility-side capacity and regulatory posture.

Value gradient across the spectrum is steep. An executed GIA at a low-congestion node commands a substantially different valuation from matched capacity in active cluster study, and the gradient has steepened as queue depths have grown. Available executed-GIA-plus-cleared-upgrades positions are finite and being spent down; hyperscalers with legacy queue positions from pre-2022 have a delivery advantage that is difficult to replicate through new applications.

Fig P2.2

Interconnection queue depth: reported figures by market (US + global sample), 2026

Reported queue figures by market as of mid-2026. Definitions differ across ISOs and between generator queues and large-load requests; comparisons across markets stay illustrative; like-for-like treatment would require harmonised definitions.

Sources: PJM Cycle 1 2026 study capacity per PJM Inside Lines; ERCOT large-load figures per ERCOT Senate testimony July 2026; CAISO Cluster 15 initial vs proceeding-to-study per CAISO 2025 year-in-review; MISO and SPP figures per ISO queue reports; Lawrence Berkeley National Laboratory US-wide aggregate at end-2025. Different reporting frames apply across sources; the chart is a visual composite, not a directly comparable table.

05US case study: the 2023 to 2026 regulatory resets

FERC has run several major regulatory actions across the 2023 to 2026 window that restructure P2 across all three sub-problems.

FERC Order 2023 (July 2023). Generator-interconnection process reform for the FERC-jurisdictional regions. Cluster studies replace serial first-come-first-served processing. Commercial-readiness deposits and withdrawal penalties filter speculative queue entries. Study-start deadlines create process discipline. Implementation phased through 2024 to 2026 across the six FERC-jurisdictional ISO/RTO regions.

FERC Order 1920 and 1920-A (May 2024 and subsequent). Long-term regional transmission planning and cost allocation. Requires transmission providers to identify long-term needs on a 20-year horizon and to allocate the resulting costs across beneficiaries. Order 1920-A refined specific implementation provisions.

FERC Order 1977 (adopted 2024). Backstop transmission siting authority for interstate transmission projects designated as being in the national interest, following the Infrastructure Investment and Jobs Act framework. Provides a partial mitigation of state-level siting delays for the specific corridor category.

FERC Order in December 2025 directing PJM on co-located loads. Established requirements for transparent rules covering AI data centres and other large loads co-located with generation. Applied specifically to PJM as the market operator with the highest concentration of co-located load activity.

PJM Expedited Interconnection Track (approved June 2026). FERC approved a PJM proposal targeting a Generator Interconnection Agreement in about 10 months and commercial operation within three years for qualifying projects that meet enhanced readiness criteria.

FERC large-load integration action (June 2026). FERC launched an aggressive targeted action directing all six FERC-jurisdictional regional grid operators to address tariff structures for large loads including AI data centres.

ERCOT operates outside FERC jurisdiction as the seventh major US ISO/RTO. Its process reforms are governed by the Public Utility Commission of Texas; FERC authority stays the alternative regulator. ERCOT reported roughly 474 GW of large loads seeking interconnection as of mid-2026, on top of a generation queue also in the hundreds of gigawatts.

What the resets change for a P2 investor

The generator-interconnection reforms under Order 2023 have started shifting the composition of the queue toward higher-completion-probability projects. PJM reports that under the reformed process it no longer holds a backlog of new service requests; approximately 220 GW entered its 2026 Cycle 1 for study.

The Order 1920 planning framework starts producing measurable regional transmission additions on longer horizons (10 to 20 years); near-term impact on hyperscaler-linked project delivery is limited. Order 1977 backstop siting has yet to be tested on any high-profile hyperscaler-linked transmission project through mid-2026.

The large-load integration directive from June 2026 is the newer front. Every ISO is now on the clock to produce tariff structures that address AI data centre load explicitly. Winners at Tier 1 (utility and ISO/RTO counterparties) are the ones whose tariff proposals get approved with terms that support the incumbent utility's cost recovery on hyperscaler-linked upgrades.

06The vendor set: three tiers

Tier 1

P2 contract counterparties

Entities that execute Generator Interconnection Agreements, load-service arrangements or co-location interface agreements, and that control the underlying transmission and distribution facilities. Reprices with ratebase growth, transmission-owner returns and regulatory outcomes on cost allocation.

ISO / RTO transmission providerPJM, MISO, SPP, CAISO, NYISO, ISO-NE (six FERC-jurisdictional) plus ERCOT (Texas, non-FERC)

Position. Non-profit market operators that run interconnection studies and manage the queue in the US organised markets. Not investable directly but the counterparty on every GIA in the organised regions. Study quality, cadence and dispute-resolution posture vary substantially across the seven US ISO/RTOs. FERC Order 2023 applies to the six FERC-jurisdictional regions; ERCOT operates under a separate Texas PUC framework.

What matters for a P2 investor. Aggregate queue clearance rate. Cluster-cycle timing. Implementation of FERC Order 2023 (six FERC-jurisdictional) or equivalent PUCT-driven reform (ERCOT). Interconnection cost-allocation posture at the specific ISO.

Constraints. Non-profit and not directly investable; the equity exposure runs through the transmission-owner members and through the generation projects that clear the queue.

Vertically-integrated utilityDuke Energy, Southern Company, Entergy, Xcel Energy, TVA

Position. Utilities in vertically integrated regions (Southeast, TVA footprint, parts of the West) operate their own transmission planning and interconnection processes outside the ISO/RTO structure. Direct GIA counterparty for AI-campus loads sited in these regions. Also handle large-load interconnection through their own load-service arrangements.

What reprices the equity. Ratebase growth from AI-campus interconnection capex. Regulatory approval on cost-allocation structures at state PUCs. Cost-recovery mechanisms for the specific interconnection upgrades.

Constraints. State PUC posture. Prudency risk on ratebase additions. Cost-allocation disputes between hyperscaler load and residential ratepayers.

Independent transmission ownerITC Holdings (Fortis), AEP transmission, Eversource transmission, LS Power, Berkshire Hathaway Energy transmission

Position. Transmission-only asset owners with regulated returns under FERC formula rate structures. Portfolio of transmission facilities that host generator interconnections. Directly benefit from network upgrade capex tied to hyperscaler-linked generation and large-load interconnection projects.

What reprices the equity. Approved ratebase additions. Successful bids in FERC Order 1000 competitive transmission projects. Order 1920 long-term planning additions. Renewal of formula rate structures.

Constraints. FERC returns regime. Interregional transmission cost-allocation disputes. State-level siting risks for new lines.

Utility parent equity with large-load exposureDominion Energy, American Electric Power, PPL, Ameren, Evergy

Position. Utility parent equity captures the ratebase growth from transmission and distribution capex tied to AI-campus interconnection. AEP and Dominion Energy have been particularly visible on hyperscaler-linked capex through 2024 to 2026, driven by the concentration of data-centre load in northern Virginia and in the AEP footprint.

What reprices the equity. Approved capex programmes. Data-centre load growth in the specific utility footprint. Successful state PUC prudency reviews on hyperscaler-linked upgrades. Outcomes on Virginia SCC and equivalent proceedings.

Constraints. State-level cost-allocation friction. Interregional cost-allocation friction on FERC-jurisdictional facilities.

Large-load interconnection service providerRegional utility large-load service groups (Dominion Virginia Power, Duke Carolinas, Entergy Louisiana, Georgia Power, Xcel Colorado)

Position. Not a separate corporate entity in most cases but a distinct service group inside the utility that handles large-load customer connections. In 2024 to 2026 this function moved from routine utility business to strategic priority for the specific utilities hosting concentrated hyperscaler load.

What reprices the parent equity. Approved large-load tariff structures. Rate cases advancing dedicated hyperscaler tariff schedules. Cost-recovery certainty on large-load upgrades.

International Tier 1 counterparties

UK · listed integrated + transmissionNational Grid PLC, SSE plc

Position. National Grid PLC owns and operates the England and Wales transmission network (NGESO separated as independent public entity, but National Grid retains transmission-owner ratebase); US operations at National Grid USA (New York, Massachusetts) add regulated US exposure. SSE plc holds substantial UK transmission via SSEN Transmission (northern Scotland) plus renewables developer and distribution operator businesses.

What reprices the equity. Approved TMO4+ connection reform outcomes on UK data-centre load; approved transmission-upgrade capex at Ofgem; US ratebase-growth at National Grid USA subsidiaries; SSEN Transmission ratebase growth in Scotland plus SSE renewables pipeline.

Continental Europe · listed TSOsTerna (Italy), Redeia (Spain), Elia Group (Belgium + Germany's 50Hertz)

Position. Three listed European TSOs with regulated returns and multi-year capex programmes. Terna: Italian national TSO with cross-border interconnections to France, Switzerland, Austria, Slovenia, Malta, Greece, Tunisia, Montenegro. Redeia (Red Eléctrica plus telecom): Spanish TSO with cross-border links to France, Portugal, Morocco. Elia Group: Belgian TSO plus majority ownership of 50Hertz (eastern Germany), exposed to North Sea offshore-wind connections and Belgian plus eastern-Germany data-centre load.

What reprices the equities. Approved capex at national regulators (ARERA, CNMC, CREG, BNetzA). Data-centre load growth in Milan, Madrid, Berlin/Brandenburg. ENTSO-E cross-border project delivery.

Continental Europe · state-controlled TSOs (indirect access)RTE (France), TenneT (DE/NL)

Position. Two largest European TSOs by scope. RTE operates the French transmission network under EDF group historical ownership. TenneT operates both the Dutch national network and the eastern-plus-southern German network. Both state-controlled and not directly listed; investor access runs through debt (regulated utility bonds) and adjacent listed entities (EDF, potential future TenneT stake divestment).

India · listed transmission platformsPower Grid Corporation of India (PGCIL), Adani Energy Solutions (AESL), Sterlite Power, Indigrid (InvIT), Tata Power

Position. India offers a genuinely rare mix: state-controlled central transmission utility (PGCIL, listed with regulated returns under CERC) plus listed private transmission platforms competing on CTU BOOT tenders. AESL is the largest listed private platform with growing intra-state distribution. Sterlite Power holds substantial private BOOT portfolio. Indigrid is a listed Infrastructure Investment Trust holding operational transmission assets. Tata Power operates diversified generation plus transmission plus distribution across multiple states.

What reprices the equities. Green Energy Corridor III delivery; ISTS waiver renewal outcomes; new BOOT project wins; data-centre load in Mumbai, Chennai, Hyderabad, Bengaluru; state-level distribution privatisation outcomes for Tata Power.

MENA · state-controlled listed utilitiesDEWA (Dubai), TAQA (Abu Dhabi), Saudi Electricity Company (SEC)

Position. Three state-controlled listed utilities providing direct MENA P2 exposure. DEWA: vertically-integrated Dubai utility (DFM-listed); direct exposure to Dubai data-centre load and AI-cluster infrastructure. TAQA: Abu Dhabi diversified generation plus transmission plus water platform (ADX-listed); direct UAE AI-cluster participation plus international generation. SEC: vertically-integrated Kingdom-wide utility (Tadawul-listed); direct participation in Vision 2030 renewables integration and NEOM infrastructure.

What reprices the equities. Data-centre load growth in Dubai; UAE AI-cluster programme wins; NEOM infrastructure delivery signals; approved capex programmes at DEWA, ECRA and adjacent regulators.

APAC · Japan and Korea utilitiesTEPCO Power Grid, Chubu Electric Power Grid, OCCTO coordination (Japan); KEPCO (Korea)

Position. Japan's ten regional utilities separated transmission subsidiaries under reforms; TEPCO Power Grid and Chubu Electric Power Grid are the two largest. OCCTO coordinates cross-region planning. Parent equities (TEPCO Holdings, Chubu Electric) provide indirect exposure. KEPCO is South Korea's vertically-integrated national utility (KRX-listed, state-controlled). Data-centre load growth in Tokyo, Osaka, Seoul metropolitan areas is the primary driver.

Australia · state TNSPsTransGrid (NSW), AusNet Services (VIC, ASX-listed), Powerlink (QLD), ElectraNet (SA), TasNetworks (TAS)

Position. State-level transmission network service providers across the National Electricity Market. AusNet Services is publicly listed on ASX. TransGrid, Powerlink, ElectraNet, TasNetworks held by infrastructure-fund consortia (direct equity access varies).

What matters for P2 investors. AEMO Integrated System Plan project awards. Renewable Energy Zone programme delivery. Data-centre load growth in Sydney, Melbourne, Perth.

Tier 2

P2 enabling suppliers

HV switchgear OEMs, transformer manufacturers, HVDC converter vendors and transmission cable makers whose equipment feeds the network upgrades that Tier 1 delivers. Reprices with the order book, with lead-time compression or extension and with pricing power in constrained sub-segments.

HV and EHV transformer OEMHitachi Energy, GE Vernova Grid Solutions, Siemens Energy Grid Technologies, Wilson Transformer, SGB-SMIT, Prolec-GE, Virginia Transformer

Position. The HV and EHV transformer segment is one of the most supply-constrained parts of the transmission stack. Lead times are in the multi-year range for large orders at the top-of-line vendors, with the longest-lead units reported into the 48-60+ month range depending on specification and vendor. The constraint has several component causes (specialty electrical steel supply, large-forging capacity, insulation systems, skilled labour) that each require their own capex cycles to address.

What reprices the equity. Announced capacity additions. Order-book depth on multi-year contracts. Pricing power in the constrained sub-segment. Service and refurbishment attach rates on installed base.

HV switchgear and protection OEMHitachi Energy, GE Vernova Grid, Siemens Energy Grid, ABB, Toshiba T&D, Schneider Electric

Position. HV circuit breakers, gas-insulated switchgear (GIS), disconnectors, protection relays. Vendor set overlaps with the HV transformer vendor set on the largest names. Lead times are shorter than for HV transformers but have extended materially through 2024 to 2026. SF6 regulatory phase-out affects switchgear specification and vendor product portfolios.

What reprices the equity. Order book. Pricing power on specific sub-segments (GIS in particular). Grid-modernisation and digital-substation attach rates.

HVDC converter station OEMHitachi Energy, Siemens Energy, GE Vernova

Position. High-voltage direct current transmission for long-distance and cross-market projects plus offshore-wind connections. The Western HVDC converter-station market is highly concentrated among three vendors with global capability at scale; additional Chinese and other regional players exist but face market-access constraints in specific jurisdictions. Lead times for HVDC converter stations are comparable to or longer than HV transformer lead times.

What reprices the equity. Announced HVDC project wins. Offshore-wind project pipeline. Interregional transmission project awards.

Transmission cable manufacturerPrysmian, Nexans, NKT, Sumitomo Electric, Southwire, LS Cable & System

Position. Manufacturers of the HV and EHV cables used in underground and submarine transmission projects. Offshore wind and cross-market interconnections have driven demand growth through the current cycle. Submarine cable specialisation is a subset with tighter capacity.

What reprices the equity. Contracted project pipeline. Cable-plant capacity expansion. Copper price exposure.

Tier 3

P2 emerging platforms

Technology vendors and platform developers whose commercial reach expands as the interconnection problem intensifies. Grid-enhancing technologies (GETs), grid-forming inverters, DER aggregators, virtual power plant operators. Reprices on regulatory recognition, on utility procurement decisions and on standardisation events.

Grid-enhancing technology (GET)LineVision, Smart Wires, Heimdall Power (dynamic line rating and power-flow control)

Position. Technologies that increase the effective capacity of existing transmission lines without new construction. Dynamic line rating (DLR), power-flow control (advanced power flow controllers, or APFCs), topology optimisation. Growing regulatory recognition from FERC and state PUCs as accelerators of interconnection resolution.

What reprices the equity. Utility procurement of GET deployments. FERC orders that recognise GET as an alternative to network upgrades. Cost-per-MW installed relative to conventional transmission upgrades.

Grid-forming inverterSMA Solar, GE Vernova (renewables), Power Electronics, Sungrow, Enphase Energy (residential-scale variant)

Position. Inverters capable of providing grid-forming behaviour (voltage and frequency support without a grid reference) are increasingly important as inverter-based resources (IBR) dominate new generation. NERC and FERC standards for IBR performance have been in flux through 2024 to 2026; grid-forming capability is a distinguishing technical property.

DER aggregator / VPP operatorSunrun, Tesla Energy (VPP), Voltus, CPower, Enel X

Position. Aggregators of distributed energy resources providing grid services and capacity to wholesale markets. FERC Order 2222 required ISOs to accommodate DER aggregations in the wholesale markets; implementation is ongoing across ISOs.

07The 2024 to 2026 comps and reform milestones

Fig P2.3

US case study: generator-interconnection study cycle before and after FERC Order 2023

Directional comparison of historical serial-study process against cluster-study process under Order 2023. Actual timelines vary by ISO, project readiness and study complexity. Post-reform figures reflect target cycle times; specific ISO implementations are still bedding in.

Source: adikumar.co synthesis of ISO interconnection procedures pre- and post-Order 2023, and PJM Expedited Interconnection Track approved June 2026. Directional; specific cycle timings vary by ISO implementation and by project category.

Micro-case: Amazon to Talen Susquehanna path

Deal · November 2024 (ISA rejection) plus June 2025 (FOM PPA)Amazon (AWS) × Talen Energy · Susquehanna nuclearEvent 1November 2024: FERC (2-1 vote) rejected PJM's proposed amendment to the Susquehanna ISA that would have supported the Amazon behind-the-meter co-location arrangement. FERC's stated basis was that PJM had not demonstrated the need for the proposed non-conforming provisions. The order exposed unresolved questions on reliability, transmission service and cost allocation for large loads co-located with generation.Event 2June 2025: A distinct commercial arrangement (a front-of-meter PPA structure for approximately 1.92 GW) followed. Amazon and Talen continued working together at Susquehanna under a different commercial form. Different legal instruments (a FERC-filed ISA amendment vs a retail PPA); press coverage commonly describes the second as a restructuring of the relationship.Cross-tier readThe FERC posture on co-located BTM configurations is now a first-order variable in any nuclear-anchored BTM proposal. FERC subsequently directed PJM (December 2025) to establish transparent rules for AI data centres and other large loads co-located with generation.Vendor reratingTransmission-owner equity (PJM member utilities) benefited from the shift toward FOM configurations, which drive more visible network-upgrade capex through the utility ratebase. Talen equity rerated on the June 2025 PPA announcement.

Micro-case: Northern Virginia data-centre load and PJM transmission planning

Multi-year · through 2025 to 2026PJM / Dominion Energy · Northern Virginia data-centre loadPositionNorthern Virginia hosts the largest single concentration of data-centre load globally. PJM and Dominion Energy have jointly assessed and proposed transmission-upgrade programmes to accommodate continued load growth in the region.Regulatory postureState PUC (Virginia SCC) and FERC-jurisdictional cost-allocation questions are both in play. Specific cost-allocation between data-centre customers and residential ratepayers has been contested at Virginia SCC.Vendor exposureDominion Energy is the utility with the largest direct exposure. AEP, LS Power and other transmission-owner participants in PJM have secondary exposure. HV transformer OEMs and HV switchgear OEMs supply the physical upgrades.

Micro-case: PJM Expedited Interconnection Track

Regulatory milestone · June 2026FERC approves PJM Expedited Interconnection TrackStructureFast-track interconnection process targeting Generator Interconnection Agreement in about 10 months and commercial operation within three years for qualifying projects meeting enhanced commercial-readiness criteria (site control, offtake, cost commitment).ImplicationsMoves the visible bottleneck away from queue existence for well-readied projects. Highlights the gap between the fast-track subset and the broader queue population that continues on longer cycles.Vendor exposureBenefits transmission owners with well-scoped project pipelines. Highlights equipment and permitting as the next binding constraints as the process bottleneck relaxes.

Micro-case: UK Ofgem plus NGESO TMO4+ connection reform

Multi-year · 2024 to 2026Ofgem + NGESO · TMO4+ connections queue reform (Regime C)StructureOfgem and NGESO moved the transmission connections queue from first-come-first-served to first-ready-first-connected under the TMO4+ framework. Commitment fees and hard milestones filter speculative projects. Data-centre load treated explicitly within the reform scope alongside generation.ScaleOfgem reported approximately 315 data-centre projects representing about 73 GW of demand in the UK connection queue as the reform bit; the reform is intended to advance the ready projects and filter the speculative ones out of scarce connection capacity.Vendor exposureDirect exposure at National Grid PLC (transmission-owner ratebase growth on approved network upgrades), SSE (SSEN Transmission ratebase in Scotland), DNOs on the distribution side. Equipment supply chain overlaps with the US market on HV transformer and HVDC vendor concentration.Cross-readUK provides the closest available non-US analogue to the US FERC Order 2023 process reform combined with the FERC June 2026 large-load integration action. Similar reformist rhetoric, similar underlying scarcity dynamics, different legal instrument.

Micro-case: EU Grid Action Plan and ENTSO-E TYNDP 2024

Multi-year · November 2023 through 2026EU Commission + ENTSO-E · Grid Action Plan and Ten-Year Network Development Plan (Regime C)StructureEU Commission published the Grid Action Plan in November 2023, followed by member-state TSO reforms running through 2024 to 2026. ENTSO-E TYNDP 2024 sets the ten-year framework for cross-border transmission investment. Data centres explicitly treated as systemically relevant electricity users with flexibility and grid-support participation as first-class considerations.ScaleENTSO-E aggregate transmission-investment pipeline runs into the hundreds of billions of euros across the 2024 to 2034 horizon. Specific project delivery cadence varies by member state, with cross-border interconnection capacity a first-order variable.Vendor exposureDirect exposure at national TSOs (RTE, TenneT, 50Hertz, Amprion, Terna, Redeia, Statnett, Elia, Fingrid). Equipment supply chain concentrates at Hitachi Energy, Siemens Energy Grid Technologies, GE Vernova, Prysmian and Nexans on the largest projects.Cross-readFraming differs from the US: less about generator queue reform, more about large-load integration into an already-constrained transmission system. Regulatory tempo is set at the EU level, execution runs through member-state TSOs.

Micro-case: India Green Energy Corridor III and ISTS reform

Multi-year · 2024 through 2026 and ongoingIndia CEA + CTU + private transmission platforms (Regime B / C hybrid)StructureIndia's Central Electricity Authority advanced Green Energy Corridor III (multi-billion-dollar transmission programme to accommodate renewables build-out) plus continued Inter-State Transmission System (ISTS) waiver extensions supporting renewables and hybrid load configurations. Private transmission BOOT tenders under CTU frameworks continue to advance.ScaleGreen Energy Corridor III is a multi-billion-dollar programme with a 2024 to 2030 delivery horizon. Data-centre load concentration in Mumbai, Chennai, Hyderabad and Bengaluru drives specific state-level tariff developments alongside the national transmission programme.Vendor exposurePower Grid Corporation of India (PGCIL, listed state-controlled) holds the largest programme exposure. Adani Energy Solutions, Sterlite Power, Indigrid (InvIT) and Tata Power on the private transmission side. BHEL, CG Power, Kalpataru Projects International on the equipment and EPC side.Cross-readIndia's regime blends Regime B (state-controlled central utility) with Regime C-like private transmission BOOT tenders on specific projects. The result is a hybrid setup that gives investors direct listed equity access to the transmission-development story in ways not available in most Regime B or D markets.

Micro-case: NEOM AI-cluster infrastructure planning

Multi-year · ongoing through 2030Saudi Arabia NEOM + UAE AI clusters · Regime E hybrid or dedicated infrastructureStructureNEOM in Saudi Arabia and specific UAE AI-cluster developments plan generation, transmission and load together as integrated sovereign infrastructure programmes. Regime E in the P2 taxonomy: hybrid or dedicated infrastructure. Multi-billion-dollar programmes with 2025 to 2032 delivery horizons.Vendor exposureSaudi Electricity Company (SEC), Saudi Power Procurement Company on the Saudi side. DEWA, EWEC, TAQA on the UAE side. Equipment vendor set includes international leaders (Hitachi Energy, Siemens Energy Grid Technologies, ABB) with growing Chinese vendor participation on specific export contracts and Korean (Hyundai Electric) participation on Korean-consortium projects.Cross-readRegime E delivers the highest control over cross-stage integration at P1, P2 and P3 simultaneously because the same sovereign entity coordinates the full stack. Trade-off is limited direct equity access for global investors; exposure runs through debt instruments, project-specific SPVs and adjacent listed entities.

08Sequencing across queue positions and geographies

Concentrating hyperscaler interconnection commitments at a single node stakes a full campus on one set of regulatory and physical outcomes. Running a portfolio across ISOs and readiness stages diversifies the aggregate delivery risk in ways a single-node exposure cannot.

Observable hyperscaler behaviour through the 2024 to 2026 window has shifted toward portfolio construction. Positions run in parallel across multiple ISOs, GIA-optionality nodes hold staged deposits, and interconnection risk gets diversified across regions with different regulatory postures and different transmission-owner counterparties. Concentrated one-shot commitments have become the exception; the base case has shifted.

Three sequencing dimensions apply at P2. The geographic dimension covers which ISO or which vertically-integrated utility footprint. The development-stage dimension spans queue-cluster, post-facility-study and executed-GIA positions in the same portfolio. The buyer-side flexibility dimension separates anchor-contract-locked projects from anchor-contract-optional ones. Hyperscalers with portfolios spanning all three dimensions hold lower exposure to any single interconnection event. On the vendor side, transmission owners and GET vendors with multi-region operations capture the disproportionate value from customers who are running that portfolio strategy.

09Physical bottlenecks that bind delivery globally

HV and EHV transformer lead times

Lead times for large HV and EHV transformers at the top-of-line vendors are in the multi-year range for new orders. Widely reported figures place the longest-lead units into the 48-60+ month range for specific classes and vendor combinations; individual project procurement varies with specification, urgency and vendor relationship. Constraint components include specialty electrical steel supply, large-forging capacity, insulation systems, skilled labour and testing capacity.

Fig P2.4

HV / EHV transformer lead-time trajectory, 2020 to 2028

Approximate lead times for large HV and EHV transformer orders at the top-of-line vendors. Directional; specific project lead times vary with specification and vendor relationship.

Source: adikumar.co synthesis from Hitachi Energy, GE Vernova and Siemens Energy Grid Technologies earnings commentary and industry reports 2020 to 2026. Values are directional; specific orders vary widely by kVA rating, cooling configuration and delivery urgency.

HVDC converter station capacity and vendor concentration

The Western HVDC converter-station market concentrates among Hitachi Energy, Siemens Energy and GE Vernova as the vendors with global delivery capability at scale. Chinese vendors and other regional players exist but face market-access constraints in specific jurisdictions. HVDC lead times are comparable to or longer than HV transformer lead times; vendor concentration means that a delivery slot loss at one vendor is difficult to recover at another.

Fig P2.5

HVDC converter station contracted capacity, Western vendor market, 2022 to 2027

Approximate HVDC converter station orders (or contracted commitments) by primary vendor. Global market; specific vendor shares depend on how transmission-owner announcements are attributed.

Source: adikumar.co synthesis from Hitachi Energy, Siemens Energy and GE Vernova disclosures 2022 to 2026. Directional; HVDC project attribution methodology varies across sources.

Transmission-line permitting across regimes

Permitting cadence varies substantially across regime archetypes. A cross-region P2 investor underwrites each specific regime; a universal profile would overstate the harmonisation.

US. NEPA review plus state siting plus private ROW plus federal-lands ROW (BLM, USFS). Historical major-line timelines exceed a decade. Order 1977 backstop siting applies to national-interest corridors only; scale delivery unproven as of mid-2026.

UK. Nationally Significant Infrastructure Projects (NSIP) regime under the Planning Act 2008; Planning Inspectorate adjudicates, Secretary of State consents via Development Consent Order (DCO). Multi-year pre-application-through-decision cycle; legal challenges extend further. TMO4+ addressed the queue; the underlying planning regime stays separate.

Continental Europe. National permitting frameworks at member-state level, with EU-level acceleration under Renewable Energy Directive III (RED III) plus Grid Action Plan (2023). RED III introduced fast-track procedures for projects deemed of "overriding public interest." Delivery cadence varies between France (relatively centralised), Germany (federal plus Länder complications), Italy (municipal opposition), Spain (fast throughput on renewables) and the Nordics. TenneT's North Sea offshore-wind programme is the most-visible current test.

India. Ministry of Environment, Forest and Climate Change (MoEFCC) environmental clearance plus state-level ROW under the Indian Electricity Act. Forest (Conservation) Act 1980 clearance constrains forested corridors. CTU coordination and CEA planning sit above the local layer. BOOT tenders typically clear faster than PGCIL-executed projects.

China. Integrated with State Grid Corporation of China planning through provincial and municipal approvals under NDRC oversight; land and ROW through state channels. Delivery cadence on large HV and HVDC corridors runs substantially faster than Western regimes, reflecting the integrated planning-plus-permitting structure characteristic of Regime D. Cross-border projects add bilateral-treaty timelines.

Middle East. UAE (DEWA, EWEC, TRANSCO) and Saudi Arabia (SEC) route permitting through utility-integrated planning with sovereign approval. Timelines run substantially faster than Western regimes on projects aligned with national plans (Vision 2030, UAE AI-cluster programmes). NEOM operates as an integrated programme with dedicated permitting authority.

Fig P2.B

Transmission-line permitting: parallel processes and typical cycle stages

Federal and state permitting steps for a new HV transmission line and their typical sequence. Each step has independent duration risk. Order 1977 backstop siting authority applies to specific federally-designated national-interest corridors only.

Transmission-line permitting: parallel and sequential processesFederal NEPA reviewEIS / EA processFederal-lands ROWBLM, USFS approvalsState siting (or Order 1977 backstop)State PUC / national-interest corridorPrivate ROWLandowner negotiationConstruction start → energisationOnly after every parallel process clearsEACH PROCESS RUNS IN PARALLEL WITH INDEPENDENT DURATION RISK

Source: adikumar.co synthesis. Simplified representation; actual permitting sequences vary widely by project. A fast NEPA review does not accelerate the project if state siting or private ROW is stuck. Order 1977 backstop siting authority applies to a specific corridor category only.

10Two competing global solutions: build more grid, or extract more from existing grid

Build-vs-extract is one of the two or three central P2 investment contests, and it is the fastest-moving of them. Global P2 scarcity has two competing solution paths, both live at once. Build-more-grid is the base case but faces multi-year physical constraints. Extract-more-from-existing-grid is the accelerator path. The IEA's 2026 estimate of 1,200 to 1,600 GW of unlockable advanced-stage projects breaks down into roughly 750 to 900 GW addressable via flexible or non-firm connection frameworks and 450 to 700 GW via grid-enhancing technologies (dynamic line rating, advanced power-flow control, topology optimisation). Which path a jurisdiction favours restructures which vendors capture the P2 rent pool.

The bottleneck framework in Section 09 treats HV transformer lead times, HVDC vendor concentration and transmission-line permitting as binding constraints on the pace of grid delivery. Those constraints are real, and none of them resolves within a single capex cycle at any individual OEM. The question globally is what to do about them, and the answer runs along two tracks that operate in parallel across every regime archetype.

The build-more-grid path

Add HV and EHV transmission lines, expand substation footprints, procure MV and LV transformers, build new HVDC converter stations for long-distance transfers and for offshore-wind interconnection. Vendor beneficiaries concentrate at Tier 2: Hitachi Energy, GE Vernova Grid Solutions, Siemens Energy Grid Technologies, Toshiba T&D, Mitsubishi Electric, Hyundai Electric, BHEL, CG Power, Prysmian, Nexans, NKT, Sumitomo Electric, LS Cable & System. Tier 1 counterparties (utilities and transmission owners) capture ratebase growth as the network upgrades pass into regulated recovery. This path is the base case in every regime; it is also capacity-constrained on the equipment side and permitting-constrained on the siting side.

The extract-more-from-existing-grid path

Grid-enhancing technologies (GETs) increase the effective capacity of existing transmission lines without new construction. Dynamic line rating (DLR) uses real-time weather and conductor temperature data to raise transmission line ratings above conservative static values. Power-flow control (advanced power flow controllers, APFCs) reroutes power across meshed networks to unlock underused capacity. Topology optimisation uses software to reconfigure network switching in ways that reduce congestion. Non-firm connections give generators or loads grid access with acceptance of curtailment risk, allowing the same physical infrastructure to serve more MW than firm connections alone would allow. Storage plus flexible load smooths peaks such that peak-hour network reinforcement can be deferred. Grid-forming resources (inverters, BESS in grid-forming mode) provide voltage and frequency support and reduce the need for conventional synchronous condenser reinforcement.

Vendor beneficiaries concentrate at Tier 3: LineVision, Smart Wires, Heimdall Power on GETs; SMA and Sungrow on grid-forming inverters; Sunrun, Voltus, CPower, Enel X, Octopus Energy on DER aggregation and virtual power plants; Tesla Megapack, Fluence, Powin, Sungrow on grid-forming BESS with hall-level and utility-scale deployment. Regulatory adoption of the extract path is uneven across regimes: FERC has recognised GETs as network-upgrade alternatives in specific proceedings, Ofgem has advanced non-firm connection frameworks under TMO4+, ENTSO-E has framed data-centre flexibility as first-class grid support, but implementation is running behind the reformist rhetoric across most jurisdictions.

The investment implication

A P2 investor underwriting only the build path captures the incumbent transmission-owner and equipment-supplier ratebase-growth story but misses the accelerator path. An investor underwriting only the extract path captures the emerging-platform re-rating but misses the base-case demand for physical grid reinforcement. The IEA's 1,200 to 1,600 GW estimated unlock through non-firm connections and GETs is a large enough number to make Tier 3 emerging platforms individually investable in their own right, and to shift the weight investors should place on them relative to Tier 2 incumbents alone. The two paths do not substitute for each other in the aggregate; they run in parallel, each capturing a share of the total P2 problem.

11Investment contests at P2

Five contests will define P2 winners globally through the end of the decade. Each cuts across regimes; specific regime instances become examples within a common underlying question.

Contest 1. Grid-access reform quality across regimes

Whether each regime's reformed connection process clears deliverable projects to energisation. FERC Order 2023 implementation cadence at the six FERC-jurisdictional ISO/RTOs (with PJM's June 2026 Expedited Interconnection Track as a leading signal, MISO, SPP via its 2026 CPP, CAISO, NYISO and ISO-NE on distinct timelines). Ofgem plus NGESO TMO4+ implementation cadence in Great Britain (315 data-centre projects, 73 GW). EU Grid Action Plan project-delivery cadence at national TSOs (RTE, TenneT, 50Hertz, Amprion, Terna, Redeia, Elia, Statnett, Fingrid). India's Green Energy Corridor III and ISTS reform. Common thread: reforms that filter speculative queue entries without triggering waves of withdrawal that leave transmission planning short of firm demand signals.

Contest 2. Large-load tariff design and cost allocation

Which specific large-load tariff structures become the reference templates. In the US: state PUC proceedings at Virginia SCC (Dominion), Georgia PSC (Georgia Power), Duke's home states, Xcel's Colorado footprint, Texas PUCT (outside FERC), plus the FERC June 2026 large-load integration action across the six FERC-jurisdictional operators and FERC Order 1920 cost allocation contested at state PUCs across MISO, SPP and PJM. In the UK: Ofgem's evolving data-centre-specific tariff and cost-allocation frameworks. In Continental Europe: national tariff frameworks aligning with the Grid Action Plan. Winners at Tier 1 (utility and ISO/RTO counterparties) are the ones whose tariff proposals get approved with terms that support cost recovery on hyperscaler-linked upgrades.

Contest 3. Physical grid capacity: HV transformers, HVDC and vendor concentration

Whether top-of-line HV transformer vendors add capacity in a manner that compresses lead times or whether the constraint persists as a durable feature for the rest of the decade. Announced capacity additions at Hitachi Energy, GE Vernova Grid Solutions and Siemens Energy Grid Technologies are the primary signals. HVDC converter-station capacity in the Western market concentrates among the same three vendors. Chinese vendors (TBEA, XD Group, Wanshida) hold substantial global export presence particularly in Belt-and-Road markets, Middle East, Africa and Southeast Asia; trade-policy considerations constrain their share in specific Western jurisdictions. The competitive pressure on the Western top-of-line vendors in export markets from Chinese entrants is a secondary contest within this.

Contest 4. Transmission-line permitting reform across regimes

Whether permitting reforms in each regime compress the actual project-development timeline. US: Order 1977 backstop-siting authority on national-interest corridors plus any federal permitting legislation through 2027. UK: NSIP regime under the Planning Act and Development Consent Order cadence. Continental Europe: Renewable Energy Directive III fast-track procedures under the "overriding public interest" designation. India: MoEFCC clearance timelines plus Forest Act clearance for forested corridors. China: SGCC-integrated planning-plus-permitting continues to deliver faster than Western regimes on HV/HVDC corridors aligned with national plans. Winners are the regimes whose permitting timelines compress; the transmission-owner and equipment-vendor exposure follows the delivery cadence; reformist rhetoric adjusts around it.

Contest 5. Build-grid versus extract-grid economics

The share of aggregate P2 demand that gets addressed by the build-more-grid path versus the extract-more-from-existing-grid path (see Section 10). The IEA's 1,200 to 1,600 GW estimated unlock through non-firm connections and grid-enhancing technologies is a large enough number to make Tier 3 emerging platforms individually investable in their own right. Contest outcome depends on regulatory recognition of GETs and non-firm connections in each regime (FERC has partially recognised, Ofgem has advanced non-firm frameworks under TMO4+, ENTSO-E has framed data-centre flexibility as system support), on utility procurement of GET deployments, and on hyperscaler willingness to accept curtailment risk in exchange for faster energisation. Vendor winners: LineVision, Smart Wires, Heimdall Power (GETs); SMA, Sungrow (grid-forming inverters); Sunrun, Voltus, CPower, Enel X, Octopus Energy (DER aggregation and VPP); Tesla Megapack, Fluence, Powin (grid-forming BESS).

12How P2 links to P1 upstream and P3 downstream

The stages are in series and only work together. A P1 PPA delivers economic value once the counterparty can get the megawatt-hours to load; that path runs through the P2 arrangement. Injecting power into the transmission system through P2 accomplishes little until the campus at P3 has a substation ready to receive it. Underwriting a P2 exposure without also underwriting the P1 offtake anchor and the P3 receiving infrastructure leaves out the constraints that drive delivery risk on the specific project.

Two concrete implications determine the coupling works in practice. A hyperscaler that arrives at the ISO with a well-scoped campus substation design and a planned MV distribution build-out at P3 is a more credible interconnection counterparty than one without. That influences cluster-study prioritisation, transmission-owner cost-recovery certainty and the tone of the FERC-jurisdictional filing process. The delivery-point selection at P2 also constrains the P3 substation design; voltage levels and short-circuit ratings at the injection point determine what receiving equipment gets specified.

Vendor implications flow both ways across the two boundaries. On the Tier 1 side, customers arriving with pre-scoped P3 receiving capability (from internal engineering, prior utility partnership or reference-project experience) get better queue treatment. On the Tier 2 side, HV switchgear and transformer specifications flow from P3 requirements back into P2 network upgrade design; the vendor chosen at P3 often gets first look at P2 upgrades on the same project. A P2 investor who treats interconnection as a self-contained asset class misses these cross-stage effects.

13Historical anchor: how the queue got this deep

The current interconnection queue is the compound of three separate accelerants that arrived through the 2020 to 2024 window.

The Inflation Reduction Act of August 2022 accelerated a strong renewables-development pipeline. Storage additions grew in parallel. Interconnection queues expanded across every ISO in response, with wind, solar and battery storage projects dominating new queue entries.

AI accelerator deployment through 2022 to 2024 drove step-change load growth in specific regions (northern Virginia, central Ohio, west Texas, central Iowa, parts of Arizona). Load-side interconnection requests grew in parallel with generation-side requests, compounding pressure on cluster studies.

Coal retirements continued through the 2020 to 2024 window. Retiring generation frees interconnection capacity in the near term but requires transmission-planning revisions to accommodate the shift, adding study complexity.

Fig P2.6

US case study: the three accelerants of aggregate ISO queue growth by category, 2018 to 2026

Approximate share of active US ISO queue capacity by generation type. Illustrative composition; specific ISO-level shares vary. Lawrence Berkeley National Laboratory placed the US-wide generation-plus-storage queue at over 2,000 GW at end 2025.

Source: adikumar.co synthesis from ISO/RTO queue reports 2018 to 2026 and Lawrence Berkeley National Laboratory queue-tracking data. Queue composition varies by ISO. Storage often appears in queues co-located with solar or standalone.

14Queue to deliverability: where the bottleneck moved

Through the 2024 cycle, the visible P2 bottleneck sat in the queue itself. Under the current cycle, it has migrated to what happens after the queue clears: study completion, network-upgrade construction, equipment availability and permitting. Counterparties who have built deliverability capability capture the current P2 rent pool; queue clearance alone captures a shrinking share.

15International case studies

The regime archetypes in Section 03 give the structure of the global picture. Three case studies below fill in the specific 2024 to 2026 evidence that supports the framework: the UK connection queue reform (Regime C, TSO/DSO), the Continental European treatment of data centres as systemically relevant users (Regime C), and the APAC and MENA allocation-heavy regimes (Regimes D and E, with Regime B on the utility side). Each case study is a comparator to the US material; comprehensive country coverage stays outside the scope.

United Kingdom · TSO / DSO connection regimeOfgem plus NGESO connection queue reform (TMO4+)

NGESO manages the transmission connections queue; DNOs and iDNOs manage distribution connections. Under TMO4+, Ofgem and NGESO have moved the connections queue from first-come-first-served toward first-ready-first-connected. Ofgem reported approximately 315 data-centre projects representing about 73 GW of demand in the UK connection queue, with commitment fees and hard milestones proposed to filter speculative projects out of scarce connection capacity. The equipment bottlenecks apply as they do in the US market: HV transformer lead times, HVDC converter-station capacity for offshore-wind and cross-border interconnection, transmission-siting under the Planning Act. The reform is the closest available non-US analogue to Order 2023 (the process) and to the FERC June 2026 large-load integration action (the load-side dimension). Investable exposure runs through National Grid PLC, SSE, adjacent independent transmission ventures and the equipment supply chain that overlaps materially with the US market.

Continental Europe · TSO / DSO connection regimeENTSO-E data-centre integration; national TSO reforms

ENTSO-E has advanced frameworks that treat data centres as systemically relevant electricity users, with flexibility and grid-support participation as first-class considerations; passive-load treatment applied earlier. EU-level policy has provided direction (Grid Action Plan November 2023, REPowerEU implementation) with member-state TSO reforms running in parallel: RTE (France), TenneT (Germany plus Netherlands), 50Hertz, Amprion, Terna (Italy), Redeia (Spain), Statnett (Norway), Elia (Belgium), Fingrid (Finland). The binding constraint on data-centre P2 in Continental Europe frames as "how does rapidly growing large load fit into an already-constrained transmission system" (the earlier question was "can the generator get a GIA." Cross-border interconnection capacity is a first-order variable in a way the US regime does not share. Vendor set concentrates on Hitachi Energy, Siemens Energy Grid Technologies, GE Vernova Grid Solutions, Prysmian, Nexans and NKT on the largest projects; regional specialists (SGB-SMIT in Germany, others) participate in specific segments.

APAC and MENA · allocation-heavy regimesWood Mackenzie estimates ~32 GW planned APAC data-centre capacity across ~1,150 projects

Wood Mackenzie estimates roughly 32 GW of planned APAC data-centre capacity across more than 1,150 projects. Developers are increasingly required to provide storage, accept curtailment risk and support grid stability. India's CEA advanced Green Energy Corridor III and ISTS waiver extensions. Power Grid Corporation of India holds the interstate transmission role, with Adani Energy Solutions, Sterlite Power, Indigrid and Tata Power on the private transmission side and BHEL and CG Power on the equipment side. Japan runs through ten regional utilities (TEPCO, KEPCO Kansai, Chubu, others) with OCCTO cross-region coordination. Korea runs through KEPCO. Australia works through AEMO plus state TNSPs (TransGrid, AusNet, Powerlink, ElectraNet, TasNetworks). Singapore's SP Group manages a specific data-centre-tariff framework after the earlier moratorium. In MENA, UAE (DEWA in Dubai, EWEC plus TRANSCO in Abu Dhabi, TAQA as diversified generation and utility) and Saudi Arabia (Saudi Electricity Company, Saudi Power Procurement Company) route P2 through utility allocation. NEOM and UAE AI-cluster developments follow the Regime E hybrid or dedicated infrastructure pattern with integrated generation, transmission and load planning.

16Scorecards, tier by tier

Tier 1 scorecard: contract counterparties

Dimensions: process quality, ratebase growth exposure, regulatory posture (positive), cost-allocation clarity, delivery-execution track record. Weightings shown; the composite is analyst-judgement based and intended to structure the underwriting discussion; a statistically derived ranking would require different inputs.

CounterpartyProcess quality (0.15)Ratebase growth (0.30)Regulatory posture (0.15)Cost-allocation clarity (0.20)Delivery track record (0.20)CompDominion Energy (D)7106677.5American Electric Power (AEP)897787.9Duke Energy888888.0Southern Company878887.7Entergy777777.0ITC Holdings (Fortis)988888.2Xcel Energy777777.0TVA767786.9

Composite spread is narrow. Dimension-level scores hold the underwriting signal; the aggregate is a rough screen only. Dominion's ratebase growth exposure to northern Virginia data-centre load holds a high weight on that dimension but is offset by cost-allocation uncertainty at the Virginia SCC.

Tier 1 scorecard: international contract counterparties

Same dimensions as the US table; investability varies (some names are state-controlled and not directly listed; investment access runs through debt, adjacent listed entities or state-directed IPP programmes; public equity stays the alternative wrapper). "Directly investable" flag noted where relevant.

CounterpartyProcess quality (0.15)Ratebase growth (0.30)Regulatory posture (0.15)Cost-allocation clarity (0.20)Delivery track record (0.20)CompNational Grid PLC (UK, listed)897787.9SSE plc (UK, listed)787787.5RTE (France, state-controlled)887787.7TenneT (DE/NL, state-controlled)797777.6Terna (Italy, listed)888888.0Redeia (Spain, listed)778887.5Elia (Belgium, listed)878877.5Power Grid Corp. of India (listed)797787.7Adani Energy Solutions (India, listed)796677.2Sterlite Power (India, private)786676.9Indigrid (India, listed InvIT)787787.4TransGrid (Australia, private)787777.2AusNet Services (Australia, private)777777.0TEPCO Power Grid (Japan, subsidiary)777777.0KEPCO (Korea, state-controlled listed)776676.6DEWA (Dubai, state-controlled listed)888888.0TAQA (UAE, state-controlled listed)788787.6Saudi Electricity Company (state-controlled listed)787777.2

Composite spread is narrow across the international set as well. Terna and DEWA are at the top of the composite on regulatory posture and cost-allocation clarity; Adani Energy Solutions and Power Grid Corporation of India hold the highest ratebase-growth exposure to India's grid and load buildout; TenneT holds the highest ratebase-growth exposure among the European TSOs owing to its concurrent North Sea offshore-wind and Dutch industrial-electrification programmes but limited direct equity access. Scores are analyst-judgement based; investability varies with the specific listing structure at each name.

Tier 2 scorecard: enabling suppliers

Dimensions and weights differ from Tier 1 because bottleneck position and order-book depth matter more than process quality. Bottleneck position holds the highest weight (0.35) as it drives pricing power in the current cycle.

VendorOrder-book depth (0.20)Bottleneck position (0.35)Service attach (0.15)Customer diversity (0.15)Capacity expansion (0.15)CompHitachi Energy10108868.8GE Vernova Grid Solutions998868.2Siemens Energy Grid Technologies998868.2Prysmian886767.3Nexans776766.7NKT776666.6Wilson Transformer / SGB-SMIT786677.1Prolec-GE / Virginia Transformer776776.9

Composite spread wider under bottleneck-weighted framework. Hitachi Energy separates from the field on the combination of order-book depth and bottleneck position. The 0.35 weight on bottleneck position reflects the current-cycle reality that pricing power in HV transformer and HVDC segments dominates other dimensions.

17A decision tree for a P2 investor

Question 1. Is the target a Tier 1 counterparty, a Tier 2 supplier or a Tier 3 emerging platform?

The three tiers have different underwriting disciplines. Do not apply a Tier 1 lens to a Tier 2 name or vice versa.

Question 2. Which of the three P2 sub-problems does the target's exposure actually touch?

P2-A generator interconnection exposure differs from P2-B large-load exposure and P2-C co-location exposure. Regulatory posture and vendor selection all vary.

Question 3. Which ISO or non-ISO region is the target primarily exposed to?

Each ISO has distinct queue depth, reform quality and cost-allocation posture. Regional concentration is a significant risk factor. Non-US exposure introduces further regime-specific variables.

Question 4. What is the queue quality of the underlying projects driving the target's economics?

Executed-GIA-plus-cleared-upgrades economics differ fundamentally from active-cluster-study economics. Post FERC Order 2023 the PJM Expedited Track adds a further tier on the higher-readiness side.

Question 5. What is the target's exposure to the three physical bottlenecks?

HV transformer lead-time exposure, HVDC vendor concentration exposure and transmission-line permitting exposure each drive different underwriting outcomes.

Question 6. What is the target's P1 anchor and P3 receiving capability?

Interconnection without a P1 offtake counterparty is speculative. Interconnection with a hyperscaler-anchored PPA and a well-scoped P3 receiving stack is a very different underwriting.

18Falsification triggers

Falsification trigger 1

PJM Expedited Interconnection Track slippage or partial rollback. The June 2026 approval targets GIA in 10 months and operation in three years for qualifying projects. Slippage would signal that process reform alone cannot deliver against the deliverability constraint.

Falsification trigger 2

HV transformer supply-chain deterioration. If the specialty electrical steel supply, large-forging capacity or skilled-labour constraint deepens; easing would be the alternative trajectory, HV transformer lead times extend and the network-upgrade timelines slip in step. Vendor-side impact concentrates at Hitachi Energy, GE Vernova Grid Solutions and Siemens Energy.

Falsification trigger 3

Transmission-line permitting reform failure at Order 1977 scale. If backstop siting under Order 1977 fails to advance national-interest corridor projects through 2027, the transmission-siting constraint persists for the decade regardless of the planning reforms under Order 1920.

Falsification trigger 4

State PUC adverse posture shift on large-load cost allocation. If a major state PUC hosting concentrated hyperscaler load (Virginia SCC on Dominion; Georgia PSC on Georgia Power; others) shifts to require substantial hyperscaler-borne cost recovery, the utility-tariff structures underpinning several 2025 to 2026 hyperscaler interconnection arrangements need renegotiation. Live dockets at Virginia SCC (Dominion large-load rate cases through 2026 to 2027) are the most-visible watch item.

Falsification trigger 5

HVDC vendor capacity shortfall. The concentrated Western HVDC converter-station market is a narrow bottleneck. A capacity shortfall at any one of Hitachi Energy, Siemens Energy or GE Vernova affects the entire long-distance and cross-market transmission pipeline. Chinese vendor market entry in specific jurisdictions could partially relieve the concentration but faces trade-policy and reliability-standards friction.

Falsification trigger 6

FERC Order 1920 cost-allocation stalemate. If the long-term transmission planning framework fails to produce actionable cost-allocation decisions on approved facilities, the transmission-owner ratebase-growth thesis compresses. Contested proceedings at MISO, SPP and PJM footprints are the visible signals.

Falsification trigger 7

Upstream P1 demand softening. A material softening of hyperscaler firm-power commitments would reduce demand for new interconnection capacity. This is a P1 event that reprices P2 exposure.

19The 24-month watch calendar

Q4 2026PJM Cycle 1 first study results. First substantive post-reform study outputs from the 220 GW Cycle 1 population.Q4 2026PJM Expedited Track first-cohort GIA execution. The 10-month target from the June 2026 approval starts hitting deliverables.2026-2027State PUC dockets on large-load tariffs. Virginia SCC (Dominion), Georgia PSC (Georgia Power), Ohio PUCO, Colorado PUC (Xcel) all have live proceedings.2026-2027FERC large-load integration action responses across the six FERC-jurisdictional ISOs. Each ISO's tariff-reform filings in response to the June 2026 directive.2027Federal permitting reform legislation. Any bipartisan federal permitting-reform bill would restructure P2 dynamics; status is a rolling watch item.2027Order 1977 backstop-siting first project decisions. Test cases on federally-designated national-interest corridors.2027HV transformer capacity announcements. Additions at Hitachi Energy, GE Vernova Grid Solutions and Siemens Energy Grid Technologies. Vendor capacity guidance leads lead-time evolution.2027-2028MISO LRTP tranche approvals. Cost-allocation and cost-recovery decisions at state PUCs across the MISO footprint.2027-2028SPP CPP implementation outcomes. First integrated planning-and-interconnection outputs from the Consolidated Planning Process.2028PJM cluster cycle results. First substantive post-reform cluster completion signals under Order 2023.2028HVDC project award cadence. New HVDC awards through 2028 clarify the concentrated-vendor market share.2028 (rolling)NEPA transmission-siting decisions. Specific project-level NEPA outcomes provide the on-the-ground signal of federal permitting posture.Q4 2026 - 2027Ofgem TMO4+ implementation and first data-centre commitment-fee filings. The connection reform applies through the first waves of confirmed queue positions in Great Britain. Deferred, withdrawn or accelerated data-centre projects will show through Ofgem disclosures.2027ENTSO-E TYNDP 2026 approval and implementation start. Continental Europe cross-border transmission planning progresses into project delivery. Member-state TSO project pipelines follow.2027India ISTS waiver renewal decisions. Continued transmission-access reform for renewables and hybrid load in India. Power Grid Corporation of India, Adani Energy Solutions, Sterlite Power, Indigrid all directly exposed.2027Australia AEMO Integrated System Plan mid-term refresh. Renewable Energy Zone (REZ) programme cadence and cost-allocation updates.Q2 2027IEA Electricity 2027 outlook publication. Updated global queue-depth aggregate; refreshed estimate of GET / non-firm connection unlock potential.2027-2028NEOM and UAE AI-cluster development milestones. Regime E hybrid-infrastructure delivery signals from the Gulf.2027-2028Chinese HV grid equipment vendor global share dynamics. TBEA, XD Group and adjacent Chinese vendors gaining or losing share in international markets under evolving trade-policy conditions.2028TenneT DE/NL HVDC subsea programme delivery. Multi-GW HVDC deliveries hit installation milestones on the North Sea programme.2028UK offshore wind AR7 CfD round results. Government CfD allocation round outcomes drive the offshore-wind pipeline and its interconnection demand.

Are the queue-depth headline numbers as bad as they look?

The headline numbers overstate deliverable capacity. Historically, only a fraction of queue entries reach commercial operation. FERC Order 2023 reforms shift the composition of the queue toward higher-completion-probability projects through commercial-readiness deposits and cure-milestone requirements. PJM reports that under the reformed process it no longer holds a backlog of new service requests; approximately 220 GW entered its 2026 Cycle 1 for study.

What is FERC Order 2023 in one paragraph?

A July 2023 FERC order requiring US transmission providers in the FERC-jurisdictional regions to shift from first-come-first-served serial interconnection studies to cluster studies, with commercial-readiness deposits, withdrawal penalties and study-start deadlines. Applied to six regional grid operators (PJM, MISO, SPP, CAISO, NYISO, ISO-NE). ERCOT operates under a separate Texas PUC framework.

What is FERC Order 1920 and how does it differ from Order 2023?

Order 1920 (May 2024) covers long-term regional transmission planning on a 20-year horizon and the allocation of resulting costs across beneficiaries. Order 1920-A refined implementation. Order 2023 (July 2023) covers the generator interconnection process. Both are relevant to P2 but address different parts of the transmission-planning stack. Cost allocation under Order 1920 is a specific contested item at state PUCs across MISO, SPP and PJM.

What is FERC Order 1977?

Order 1977 (adopted 2024) establishes backstop transmission siting procedures for interstate transmission projects designated in the national interest under the Infrastructure Investment and Jobs Act framework. Partial mitigation of state-level siting delays for the specific corridor category.

What is the difference between an ISO and a vertically-integrated utility for P2 purposes?

An ISO/RTO (PJM, MISO, SPP, CAISO, NYISO, ISO-NE, ERCOT) runs interconnection studies and market operations for a multi-utility footprint. A vertically-integrated utility (Duke, Southern, TVA, Entergy in the non-MISO footprint) runs its own transmission planning and interconnection process. Regulatory posture, dispute-resolution and cost-allocation all differ between the two structures.

Why did FERC reject the November 2024 Amazon-Talen amendment?

FERC ruled (2-1) that PJM had not demonstrated the need for the proposed non-conforming provisions of the amended Interconnection Service Agreement. The proceeding raised unresolved questions on reliability, transmission service and cost allocation for large loads co-located with generation. The June 2025 Amazon-Talen structure took a front-of-meter form; the earlier BTM approach stays the comparison case; different legal instrument, different regulatory posture.

Why are HV transformer lead times so long?

The constraint has several distinct component causes: specialty electrical steel supply, large-forging capacity, insulation systems, skilled labour and testing capacity. None is addressable within a single capex cycle. Announced capacity additions at the top-of-line vendors are the leading indicator of eventual lead-time evolution. Widely reported figures place the longest-lead units into the 48-60+ month range for specific classes and vendor combinations.

What is HVDC and why does it matter for AI-buildout interconnection?

High-voltage direct current transmission for long-distance and cross-market transmission plus offshore-wind connections. The Western HVDC converter-station market concentrates among Hitachi Energy, Siemens Energy and GE Vernova. Important for AI buildout when long-distance interregional transfers are needed to bring generation to load centres.

What are grid-enhancing technologies (GETs)?

Technologies that increase the effective capacity of existing transmission lines without new construction: dynamic line rating, power-flow control, topology optimisation. Growing recognition from FERC and state PUCs as accelerators of interconnection resolution.

How does the FERC June 2026 large-load integration action affect P2?

FERC directed all six FERC-jurisdictional regional grid operators (PJM, MISO, SPP, CAISO, NYISO, ISO-NE) to address tariff structures for large loads including AI data centres. Each ISO is now on the clock to produce tariff filings that address large-load interconnection, cost allocation and reliability specifically. Outcomes drive the ratebase-recovery certainty and the cost-borne-by-hyperscaler proportion for large-load-heavy utilities.

Is the northern Virginia data-centre concentration replicable elsewhere?

Partially. Other regions (central Ohio, west Texas, central Iowa, parts of Arizona) show related but not identical patterns. Northern Virginia's specific combination of legacy fibre infrastructure, land availability and utility-friendly posture is unusually well-suited to hyperscale data-centre concentration.

What is the P2 short thesis?

A defensible P2 short thesis targets specific transmission-owner names where cost-allocation posture is being challenged in live proceedings and where the ratebase growth built into the equity stays unsupported by the current regulatory trajectory at the state PUC or FERC level. Dominion Energy in the Virginia SCC docket cycle is the most-cited example. Another defensible short targets HV switchgear or transformer names whose announced capacity expansion timelines create margin risk once installed capacity outstrips the current lead-time regime; the timing here is 2028 to 2030; near-term framing would overstate the pace. Neither should be read as a general short against transmission-owner exposure, which benefits from durable ratebase and demand tailwinds through the decade.

Series completion: the Grid-to-Chip framework completes at P5 · Rack to gate · the last conversion cascade from rack input to the transistor gate. Every megawatt contracted, delivered and stepped down through P1–P4 ultimately makes it to compute via the P5 cascade.

See also: the fourth companion, Load-side flexibility · How firm AI load becomes a flexible grid asset, covers the software-defined competitor to physical storage with structural edge on the connection-acceleration value stream. Introduces the reclassification premium framework · the economic value created when firm AI capacity acquires a defined flexibility right.