Power Grid Statistics 2026
By Axis Intelligence Research
Co-author: Aidan Jad, EV & Energy Infrastructure | Last updated: September 17, 2026 | License: CC BY 4.0
Global electricity grid investment is set to reach around $550 billion in 2026, up nearly 20% in a year, according to the IEA World Energy Investment 2026 report. Yet Axis Intelligence Research calculates that delivering a U.S. grid connection now carries 2.39 times the friction of the pre-shortage baseline, measured by queue time, transformer lead time and equipment price.
Quick Answer: What Do the 2026 Power Grid Statistics Show?
Money is finally moving to the wires, but hardware and process are not keeping pace. The IEA puts 2026 grid spending near $550 billion. Over 2,500 GW of projects sit in grid queues worldwide, including 2,061 GW in the United States alone (Axis Intelligence Research sum of LBNL data). Large power transformers now average a 30-month lead time, and the U.S. transformer producer price index stood 91.4% above January 2020 in July 2026.
Key Findings
- According to Axis Intelligence Research, the U.S. Grid Friction Multiple (GFM) reads 2.39 as of September 2026, meaning grid delivery carries more than double the queue, lead-time and price friction of its historical baseline.
- According to Axis Intelligence Research, 2,061 GW of generation and storage sat in U.S. interconnection queues at the end of 2025, a Queue-to-Fleet Ratio (QFR) of 1.56 against the 1,323,895 MW installed fleet.
- According to Axis Intelligence Research analysis of BLS data, the U.S. producer price index for power and distribution transformers rose 91.4% between January 2020 and July 2026, including a 4.6% jump in a single month (June to July 2026).
- The IEA’s World Energy Investment 2026 estimates global grid spending at around $550 billion in 2026, equal to 82.7% of the $665 billion going to renewable power each year, per Axis Intelligence Research calculation.
- According to Axis Intelligence Research, IEA estimates imply grid-enhancing technologies and non-firm connections could unlock hosting capacity equivalent to 48% to 64% of the 2,500 GW global queue headline without waiting for new high-voltage lines.
How Much Is the World Investing in Power Grids in 2026?
The short version: more than ever, and for the first time the spending is leaning toward networks rather than generation.
The IEA’s World Energy Investment 2026 estimates total energy investment at $3.4 trillion in 2026. Electricity supply and infrastructure take $1.6 trillion of that, and electricity-related spending now accounts for nearly 60% of all global energy investment. Within the power sector, the IEA expects networks to absorb around $550 billion in 2026, up nearly 20% year on year, while battery storage for the power sector exceeds $100 billion.
Global Grid and Power Investment Snapshot (2026)
| Metric | Value | Scope | Source |
|---|---|---|---|
| Total global energy investment | $3.4 trillion | World, 2026e | IEA WEI 2026 |
| Electricity supply and infrastructure | $1.6 trillion | World, 2026e | IEA WEI 2026 |
| Electricity grid investment | ~$550 billion | World, 2026e | IEA WEI 2026 |
| Grid spending growth | nearly 20% | Year on year | IEA WEI 2026 |
| Battery storage (power sector) | over $100 billion | World, 2026e | IEA WEI 2026 |
| Renewable power investment | ~$665 billion per year | World | IEA WEI 2026 |
| Grid spend as share of renewables spend | 82.7% | Axis calculation (550 / 665) | Axis Intelligence Research |
The IEA attaches an important caveat to its own headline: part of the grid spending rise comes from tight supply chains raising the cost of transformers and cables. That line matters more than the dollar figure. A bigger capex number is not the same thing as more kilometers of conductor in the ground.
Why Two IEA Grid Investment Figures Disagree
Readers will find two IEA numbers in circulation. The Electricity 2026 grids chapter refers to “today’s USD 400 billion” in annual grid investment and says it must rise by approximately 50% by 2030. World Energy Investment 2026 puts the 2026 estimate at around $550 billion. Axis Intelligence Research does not merge or average them. The Electricity 2026 figure references the World Energy Outlook 2025 basis, while the $550 billion is the newer full-year 2026 estimate. For current-year spending, we use the WEI 2026 number; for the 2030 requirement framing, the Electricity 2026 baseline applies.
Aidan Jad’s read: A 20% jump in grid capex sounds like a turning point, and in dollars it is. But when the IEA itself flags that transformer and cable prices are inflating the total, the engineering question becomes how many MVA of capacity each dollar now buys. On U.S. pricing data covered below, the answer is noticeably fewer than in 2020.
Power Grid Investment by Region: China, the United States and Europe
China: State Grid’s 4 Trillion Yuan Plan
State Grid Corporation of China plans to invest up to 4 trillion yuan in fixed assets over the 15th Five-Year Plan period (2026 to 2030), a 40% increase on the previous plan, according to SASAC, China’s state asset regulator. That works out to an average of 800 billion yuan per year, per Axis Intelligence Research calculation. The same plan targets a 30% increase in ultra-high-voltage direct current (UHVDC) transfer capacity versus the end of the previous plan.
Execution is running hot. State Grid invested 75.7 billion yuan in January and February 2026, up 80.6% year on year, per SASAC. In the first half of 2026, the company reported more than 310 billion yuan in fixed-asset investment, up 12.6%, with 1.55 billion kW of wind and solar connected in its service areas, according to a company statement relayed by Xinhua.
United States: Utility Transmission Capex Accelerates
U.S. investor-owned electric companies spent $32.6 billion on transmission in 2024, up from $30.0 billion in 2023, and projected $39.9 billion for 2025, according to the Edison Electric Institute. That projected step is a 22.4% increase, per Axis Intelligence Research calculation. EEI members plan approximately $178 billion in transmission construction between 2025 and 2028, an average of $44.5 billion per year. Distribution spending is larger still: $60.2 billion in 2024, up from $56.7 billion.
Government data frames the longer arc. The U.S. Energy Information Administration reports that major utilities’ transmission spending nearly tripled from 2003 to 2023, reaching $27.7 billion, with capital investment up $2.7 billion (11%) in 2023 alone.
European Union: €1.2 Trillion of Grid Needs to 2040
The European Commission estimates EU grid investment needs to 2040 at €730 billion for distribution and €477 billion for transmission, as cited by ACER, the EU energy regulators’ agency. Distribution therefore carries 60.5% of the total need, per Axis Intelligence Research calculation, because most new solar, heat pumps, EV chargers and batteries connect at low and medium voltage.
Regional Grid Investment Comparison
| Region | Metric | Value | Period | Source |
|---|---|---|---|---|
| China | State Grid fixed-asset plan | up to 4,000 billion yuan | 2026 to 2030 | SASAC |
| China | State Grid average per year | 800 billion yuan | 2026 to 2030 | Axis Intelligence Research |
| China | State Grid H1 fixed-asset investment | over 310 billion yuan | H1 2026 | Xinhua (company statement) |
| United States | IOU transmission investment | $32.6 billion | 2024 | EEI |
| United States | IOU transmission investment (projected) | $39.9 billion | 2025 | EEI |
| United States | IOU planned transmission | ~$178 billion | 2025 to 2028 | EEI |
| European Union | Distribution grid need | €730 billion | to 2040 | ACER / European Commission |
| European Union | Transmission grid need | €477 billion | to 2040 | ACER / European Commission |
A Scope Note on U.S. Transmission Spending
EIA’s $27.7 billion (2023) and EEI’s $30.0 billion (2023) describe the same year but not the same thing. EIA uses FERC financial reports for major utilities in real 2023 dollars; EEI reports its investor-owned members in nominal dollars. Axis Intelligence Research publishes both side by side rather than reconciling them, because blending two methodologies would produce a number neither organization stands behind.
Aidan Jad’s read: China’s plan is denominated in fixed assets across a single national operator, so it moves like a program. The U.S. number is the sum of dozens of rate cases, each approved by a different commission. That structural difference, more than the headline totals, explains why UHVDC corridors get energized in China while interregional lines in the U.S. spend years in permitting.
How Many Projects Are Waiting in Grid Connection Queues?
Global Grid Queue: Over 2,500 GW
The IEA reports that over 2,500 GW of renewable, large-load and storage projects are stalled in grid queues worldwide, and that grid connection queues have reached record levels. For the full queue breakdown by ISO, withdrawal economics and reform tracking, see our dedicated grid interconnection queue statistics. This section covers the queue only as it relates to the wider grid buildout.
U.S. Interconnection Queue: 2,061 GW at End-2025
At the end of 2025, roughly 8,200 projects were actively seeking interconnection in the United States, representing 1,312 GW of generation and approximately 749 GW of storage, according to Lawrence Berkeley National Laboratory’s Queued Up: 2026 Edition. Axis Intelligence Research sums that to 2,061 GW, a net decline of 229 GW from the end of 2024 total, when the 2025 Edition counted 1,400 GW of generation and about 890 GW of storage across roughly 10,300 projects.
U.S. Queue Mix by Technology (End-2025)
| Technology | Active queue (GW) | Change vs 2024 | Share of 2,061 GW | Source |
|---|---|---|---|---|
| Solar | 773 | -19% | 37.5% | LBNL 2026; share: Axis |
| Storage | 749 | -16% | 36.3% | LBNL 2026; share: Axis |
| Natural gas | 253 | +86% | 12.3% | LBNL 2026; share: Axis |
| Wind | 220 | -19% | 10.7% | LBNL 2026; share: Axis |
| Other generation | 66 | n/a | n/a | Axis (1,312 minus listed) |
Gas is the only major resource growing in the queue. It rose from 136 GW to 253 GW in a year, as the IEA separately notes that orders for new gas-fired plants surged to 130 GW globally in 2025, with U.S. data center demand a major factor. The data center link is covered in depth in our AI data center statistics and the running data center electricity demand tracker.
How Much of the Queue Gets Built?
Most of it does not. Only 13% of capacity that requested interconnection from 2000 to 2019 had reached commercial operation by the end of 2024; 77% had withdrawn and 10% was still active, per LBNL’s 2025 Edition. That is 5.9 megawatts withdrawn for every megawatt built, per Axis Intelligence Research calculation.
The wait has lengthened too. LBNL reports the median time from interconnection request to commercial operation was over 5 years for projects built in 2025, against under 2 years for projects built in 2000 to 2007.
A more encouraging signal: 549 GW already held a draft or executed interconnection agreement at the end of 2025, up from 408 GW a year earlier. That is 26.6% of the active queue, per Axis Intelligence Research, and it is the portion most likely to show up as real steel. Of the 549 GW, LBNL lists 256 GW of solar, 161 GW of storage, 76 GW of wind and 45 GW of gas.
Aidan Jad’s read: The headline queue shrank, and some coverage will call that progress. It is mostly attrition. The number worth watching is the 549 GW with signed or draft agreements, because that is the pipeline that will collide with transformer and breaker lead times over the next three years.
Axis Grid Friction Multiple (GFM): How Much Harder Is It to Connect to the Grid?
Spending totals say how much money is moving. They do not say how hard it has become to turn that money into a live connection. Axis Intelligence Research built the Grid Friction Multiple (GFM) to measure exactly that for the United States.
GFM = geometric mean of three multiples, each comparing today’s condition with a documented historical baseline:
| Component | Today | Baseline | Multiple | Source |
|---|---|---|---|---|
| Queue time (request to operation) | over 5 years (2025 builds) | under 2 years (2000 to 2007 builds) | 2.50 (conservative floor) | LBNL 2026 and 2025 editions |
| Large power transformer lead time | 30 months (H1 2025 average) | 5 to 16 months (2010), midpoint 10.5 | 2.86 | NLR, citing Wood Mackenzie and DOE |
| Transformer producer prices | PPI 381.783 (July 2026) | PPI 199.5 (January 2020) | 1.91 | BLS via FRED |
| Grid Friction Multiple | 2.39 | Axis Intelligence Research |
Formula: GFM = (2.50 x 2.857 x 1.914)^(1/3) = 2.39. Each component carries equal weight, and the geometric mean is used so that no single component can dominate the reading.
How to read it: a GFM of 1.0 would mean queue time, equipment lead time and equipment price all sit at their baseline. At 2.39, the average friction across the three is well over double. Because LBNL reports the queue figures as bounds (“over 5 years”, “under 2 years”), the queue component is a floor, which makes 2.39 a conservative reading. Baselines differ by component because each source publishes a different historical reference; they are disclosed in the table above and in the CSV.
What would move the GFM down: FERC Order 2023 cluster studies shortening queue duration, new domestic transformer capacity announced by Siemens Energy (Charlotte, expected 2027) and Hitachi Energy (South Boston, from 2028) per NLR, or a sustained fall in the transformer PPI. What would push it up: further tariff pressure on grain-oriented electrical steel and copper, or another step change like the 4.6% one-month PPI jump recorded in July 2026.
Axis Intelligence Research will publish GFM readings at each new LBNL queue edition and each material PPI move, with the same formula and disclosed baselines.
Transformer Shortage Statistics: Lead Times, Prices and Import Dependence
How Long Does It Take to Get a Large Power Transformer?
Procurement for a large power transformer (LPT) now averages 30 months, up from 5 to 16 months in 2010, according to the National Laboratory of the Rockies (U.S. DOE) supply chain report, citing Wood Mackenzie. Extra-high-voltage units can take up to 60 months, per DOE. Components are just as tight: high-voltage bushings take up to 130 weeks, and on-load tap changers take 80 to 90 weeks, according to industry interviews in the same report.
Set that against the IEA’s build-time comparison. A data center takes 1 to 3 years to build. A 30-month transformer wait, or 2.5 years, lands inside that window, so the substation equipment can become the critical path for the whole campus. New grid infrastructure takes 5 to 15 years to plan, permit and complete, per the IEA.
Transformer Price Statistics
| Measure | Value | Date | Source |
|---|---|---|---|
| U.S. transformer PPI (Dec 1999 = 100) | 199.5 | Jan 2020 | BLS via FRED |
| U.S. transformer PPI | 351.192 | Jul 2025 | BLS via FRED |
| U.S. transformer PPI | 364.861 | Jun 2026 | BLS via FRED |
| U.S. transformer PPI | 381.783 | Jul 2026 | BLS via FRED |
| Change since Jan 2020 | +91.4% | Jul 2026 | Axis Intelligence Research |
| Change year on year | +8.7% | Jul 2026 | Axis Intelligence Research |
| LPT price vs 2014 to 2021 level | ~80% higher | 2025 | NLR |
| LPT cost per MVA | $15,000 to $21,000 | 2024 | NLR, citing USITC |
| Cost of a 100 MVA unit | $1.5 to $2.1 million | 2024 | Axis calculation from NLR range |
| Maximum single-unit LPT cost | up to $10 million | 2024 | NLR, citing USITC |
The PPI series is not smooth. It sat flat at 362.638 for five straight months (November 2025 to March 2026), after reaching 365.135 in October 2025, before stepping up to 364.861 and then to 381.783 in July 2026. Utilities still budgeting on 2025 quotes are pricing equipment below the current index.
Why the U.S. Imports Most Large Transformers
Roughly 80% of large power transformers consumed in the United States are imported, and import reliance for high-grade grain-oriented electrical steel (GOES) is about 90%, per NLR. Each MVA of transformer capacity needs an estimated 200 to 400 kg of GOES. Between 2011 and 2023, U.S. LPT consumption grew 28% while domestic manufacturing capacity grew only 11%, and domestic plant utilization climbed from 40% to 78%.
The installed base is aging at the same time. The U.S. had an estimated 4,900 to 6,799 LPTs in 2022, adding more than 700 per year while about 170 reach their design lifetime annually, per NLR.
Aidan Jad’s read: The transformer problem is not one shortage. It is three stacked constraints: a steel grade the U.S. barely makes, bushings and tap changers from a handful of qualified suppliers, and winding labor that takes years to train. New factory announcements address the third partially and the first not at all. That is why the lead-time component of the GFM is the one we expect to fall last.
Can Existing Power Lines Carry More Electricity? Grid-Enhancing Technology Statistics
The fastest new capacity may already be hanging on existing towers. The IEA estimates regulatory measures plus technology could free enough hosting capacity to connect 1,200 to 1,600 GW of advanced-stage projects now stuck in queues. Non-firm connection agreements could enable 750 to 900 GW, and grid-enhancing technologies 450 to 700 GW. Axis Intelligence Research notes that the 1,200 to 1,600 GW range equals 48% to 64% of the 2,500 GW global queue headline, although the IEA’s numerator refers to advanced-stage projects only.
Grid Upgrade Options: Capacity Gain vs Lead Time
| Technology | Capacity increase | Lead time | Source |
|---|---|---|---|
| Dynamic line rating | 20% to 30% | 1 to 2 years | IEA Electricity 2026 |
| Dynamic transformer rating | 5% to 15% | 1 to 2 years | IEA Electricity 2026 |
| Topology optimisation | 5% to 15% | 1 to 2 years | IEA Electricity 2026 |
| Advanced power flow control | 10% to 20% | 2 to 3 years | IEA Electricity 2026 |
| Storage as a transmission asset | 30% to 40% | 2 to 3 years | IEA Electricity 2026 |
| Reconductoring | 50% to 100% | 3 to 4 years | IEA Electricity 2026 |
| Voltage uprating | 100% to 200% | 4 to 7 years | IEA Electricity 2026 |
| New high-voltage lines | n/a | 7+ years | IEA Electricity 2026 |
The IEA cautions that these gains are not additive, since several technologies relieve the same thermal or voltage constraint. Storage deployed as a transmission asset connects directly to the buildout covered in our data center battery storage statistics.
Aidan Jad’s read: Dynamic line rating gets the headlines, but the row that should interest planners is reconductoring: up to double the capacity on an existing right-of-way in 3 to 4 years, with no new corridor to permit. Where community opposition is already blocking new builds, as our data center community opposition statistics show, that trade-off becomes decisive.
What Is Driving Power Grid Demand in 2026?
Three loads are pulling on the network at once. The first is data centers: the IEA estimates total energy sector investment for data center buildout exceeded $100 billion in 2025, including grid upgrades. Energy use behind that figure is tracked in our AI data center energy consumption statistics, and capacity scenarios in the AI data center forecast to 2030.
The second is new generation needing export capacity, led by solar; see our solar energy statistics. The third is firm capacity returning to the queue, from gas turbines to reactors covered in our nuclear energy statistics.
Methodology
Collection. Axis Intelligence Research fetched and read every source cited on this page on September 17, 2026: IEA World Energy Investment 2026 (full report PDF), IEA Electricity 2026 (grids chapter), LBNL Queued Up 2026 and 2025 editions, the National Laboratory of the Rockies report NLR/TP-5700-96742 (May 2026), the BLS producer price index series WPU117409 via FRED (monthly data through July 2026), EIA Today in Energy (November 2024), EEI Industry Data, ACER (September 2025) and SASAC (January and March 2026). Chinese first-half 2026 investment is taken from Xinhua’s report of a State Grid statement and is flagged as secondary in the dataset.
Axis calculations. Every derived figure is shown with its formula in the CSV method_note column:
- Queue total: 1,312 GW generation + 749 GW storage = 2,061 GW.
- Queue-to-Fleet Ratio (QFR): 2,061 GW (end-2025) / 1,323.895 GW installed fleet (end-2024, EEI) = 1.56. As-of dates differ by one year, since EEI had not published an end-2025 fleet figure on its page.
- Grid Friction Multiple (GFM): geometric mean of 5/2, 30/10.5 and 381.783/199.5 = 2.39.
- Technology shares: each LBNL technology total divided by 2,061 GW.
- PPI changes: July 2026 value divided by the reference month, minus one.
- Growth rates and averages: EEI and SASAC totals divided as shown in the CSV.
Declined calculations. Two tempting figures are published as retracted rows instead of numbers. We did not back out a 2025 global grid spend from the IEA’s “nearly 20%” growth, because that would invent precision. We also did not deflate global IEA spending with the U.S.-only transformer PPI, since the scopes are incompatible.
Scope. Queue data covers transmission-level requests only, not distribution-connected or behind-the-meter projects. IEA 2026 figures are estimates. EU figures are European Commission need estimates, not committed budgets. Transformer lead times combine consultant survey data and industry interviews as reported by NLR.
About This Dataset
Title: Power Grid Statistics 2026: Investment, Transmission, Connection Queues and Transformer Supply
File: power-grid-statistics-2026.csv (162 rows, one observation per row, with source URL and retrieval date on every row)
Coverage: World, United States, European Union, China; observations from 2010 to 2026, with projections to 2040
Creator and publisher: Axis Intelligence Research
License: CC BY 4.0
Cite as: Axis Intelligence Research, Power Grid Statistics 2026, 2026, axis-intelligence.com/power-grid-statistics/
Cite This Research
APA: Axis Intelligence Research. (2026, September 17). Power grid statistics 2026: Investment, transmission, connection queues and transformer supply. Axis Intelligence. https://axis-intelligence.com/power-grid-statistics/
MLA: Axis Intelligence Research. “Power Grid Statistics 2026: Investment, Transmission, Connection Queues and Transformer Supply.” Axis Intelligence, 17 Sept. 2026, axis-intelligence.com/power-grid-statistics/.
Chicago: Axis Intelligence Research. “Power Grid Statistics 2026: Investment, Transmission, Connection Queues and Transformer Supply.” Axis Intelligence, September 17, 2026. https://axis-intelligence.com/power-grid-statistics/.
Power Grid FAQ: Questions Utilities, Developers and Analysts Are Asking
Why can a data center be built faster than the grid connection it needs?
The IEA puts data center construction at 1 to 3 years, while new grid infrastructure takes 5 to 15 years to plan, permit and build. Large power transformers alone average a 30-month lead time, according to NLR, so substation equipment often becomes the campus critical path.
How much of the U.S. interconnection queue actually gets built?
Only 13% of capacity that entered U.S. queues from 2000 to 2019 reached commercial operation by the end of 2024, per LBNL; 77% withdrew. Axis Intelligence Research calculates that is 5.9 megawatts withdrawn for every megawatt built.
Is natural gas replacing solar in the U.S. interconnection queue?
Not in volume, but in direction. Gas in U.S. queues rose 86% to 253 GW in 2025 while solar fell 19% to 773 GW, per LBNL. Solar still holds 37.5% of the 2,061 GW queue against 12.3% for gas, per Axis Intelligence Research.
Why are transformer prices still rising in 2026?
The BLS producer price index for power and distribution transformers reached 381.783 in July 2026, 91.4% above January 2020 and 4.6% above June 2026, per Axis Intelligence Research. NLR attributes the pressure to imported electrical steel, copper and a small pool of qualified component suppliers.
Can reconductoring or dynamic line rating replace new transmission lines?
Partly. The IEA estimates reconductoring adds 50% to 100% capacity in 3 to 4 years and dynamic line rating adds 20% to 30% in 1 to 2 years, against 7+ years for new high-voltage lines. Gains are not additive where constraints overlap.
Which country spends the most on its power grid?
China’s State Grid plans up to 4 trillion yuan of fixed-asset investment from 2026 to 2030, per SASAC. U.S. investor-owned utilities plan about $178 billion of transmission construction from 2025 to 2028, per EEI. The EU needs €1.2 trillion across distribution and transmission by 2040.
What is the Axis Grid Friction Multiple (GFM)?
The Grid Friction Multiple is an Axis Intelligence Research metric: the geometric mean of U.S. queue-time, transformer lead-time and transformer price multiples versus historical baselines. It reads 2.39 as of September 2026, meaning grid delivery friction is more than double its baseline.
Does $550 billion in grid spending mean the bottleneck is easing?
Not yet. The IEA says part of the rise reflects higher transformer and cable costs, not only more physical build. Axis Intelligence Research treats rising spend alongside a rising Grid Friction Multiple as a sign that each dollar buys less grid than it did in 2020.