Your electricity bill went up again. You probably noticed. What you probably did not notice is that the official explanation for why it went up is incomplete, and that the policy designed to fix the underlying problem has made it structurally unfixable by private capital. Those two facts are connected. This is how.
The Stakes
In November 2025, the average New York household paid 26.49 cents per kilowatt-hour for electricity. The national average that month was 17.78 cents. That gap is not a rounding error. At average New York consumption of 600 kilowatt-hours per month, it costs a typical household approximately $630 more per year than a household anywhere else in the country, calculated as 600 kWh times 12 months times the $0.087 per-kWh differential between New York and the national average. In 2020, New York's annual average residential rate was approximately 17.44 cents per kWh. By the end of 2024 it had reached 19.66 cents on an annual average basis, and monthly data through 2025 showed rates climbing toward 27 cents. New York's rate is not just high. The gap between New York and everywhere else is expanding.
Source: U.S. Energy Information Administration, New York Electricity Profile 2024; New York Electricity Profile 2020; U.S. Bureau of Labor Statistics, New York-Newark-Jersey City Average Energy Prices, December 2024
The standard explanation points at fossil fuels: aging infrastructure, expensive gas-fired generation, old peaker plants running in dense urban neighborhoods. That explanation is real. It is also incomplete in a way that matters enormously. The deeper driver is a structural reliability problem that New York's own climate law is making worse, not better, while simultaneously eliminating the market that would fix it. To understand the bill, you have to understand what reliability actually requires and what New York is systematically dismantling.
The Illusion
The number New York uses to measure its grid's progress is installed renewable capacity. Wind and solar now account for 11.5% of in-state generation, up from near zero a decade ago. Albany presents this as proof the Climate Leadership and Community Protection Act is working. The proposed three-year moratorium on new data centers is framed as protecting those gains.
Both claims measure something real. Neither measures the right thing.
The number that determines whether the lights stay on is dispatchable capacity: generation that a grid operator can summon on demand, within minutes, regardless of weather or time of day. A gas turbine, a nuclear reactor, a hydroelectric dam each delivers full rated output when called. Solar panels produce nothing after dark. Wind turbines produce nothing when the air is still. Those two conditions tend to occur at the same time: on the coldest nights of winter, when demand peaks and the cost of getting it wrong is highest.
New York's grid is losing dispatchable capacity faster than it is gaining any kind of capacity at all. Since 2019, the state has recorded 4,315 MW of generator deactivations and only 2,274 MW of additions and uprates. That is a net loss of more than 2,000 MW on a grid being asked to serve growing demand. For reference: 2,000 MW is roughly enough to power 1.6 million homes under normal conditions.
Source: Draft 2025 New York State Energy Plan, NYSERDA, Electricity Chapter
In its January 2023 State of the Grid presentation, NYISO, the nonprofit that operates New York's bulk electricity grid, stated the problem directly. Deactivating resources tend to be dispatchable and located downstate. Generator additions are largely renewable resources located upstate. New resources do not provide the same reliability services as exiting resources.
Source: NYISO, State of the Grid Webinar, January 2023
Every generator has a capacity factor: the percentage of time it actually produces power near its maximum rated output. A solar panel in New York, averaged across nights, cloudy days, and winter months, operates at roughly 15 to 20% of its nameplate capacity annually. An onshore wind turbine runs at 25 to 35%. A dispatchable thermal unit can be called to near-full output in minutes and held there indefinitely. When the grid needs 1,000 MW, it needs 1,000 MW, not a nameplate rating that delivers 150 to 200 MW on average. Retiring 1,000 MW of dispatchable capacity removes close to 1,000 MW of firm, reliable output. Adding 1,000 MW of solar nameplate capacity adds only 150 to 200 MW of equivalent reliability. That exchange, repeated across thousands of megawatts over several years, is the gap the installed capacity headline cannot show.
Since 2019, New York's winter capacity margin has shrunk by 2,443 MW. By summer 2024, NYISO reported reliability margins had declined more than 1,000 MW in just the prior two years. What remains is old. A full 25% of the state's total thermal generating capacity has been in operation for more than 50 years. In New York City specifically, the average age of thermal generators is 65 years. When those aging machines are forced to cycle on and off repeatedly to fill gaps left by intermittent generation, they operate at their least efficient point and burn more fuel per unit of output than they would at steady baseload. That inefficiency flows into the wholesale electricity price. The wholesale price flows into your bill. That is where the $630 annual premium begins.
Source: NYISO, Winter 2023-2024 Reliability Outlook; NYISO, Summer 2024 Reliability Press Release; NYISO, 2025-2034 Comprehensive Reliability Plan, November 2025
The Compound
The reliability problem has four compounding pressures. Each is documented. Together they describe why this is not a transition problem waiting to resolve itself on schedule.
The import cushion is gone.
New York quietly lost a reliability buffer that appeared in no renewable capacity headline: clean, controllable hydroelectric power from Quebec. Between 2016 and 2022, Canadian electricity supplied 11% of everything New York's grid delivered. That share fell to 5% in 2023, 3% in 2024, and through August 2025 it stood at approximately 2%. By August 2025, New York was a net exporter to Canada on all but three days of the month, sending power north rather than receiving it. Three consecutive years of Canadian drought have constrained Hydro-Quebec's output. Roughly one in nine kilowatt-hours that New York consumed in an average year was clean, dispatchable, and imported. That supply has largely disappeared. The grid fills the gap with whatever aging in-state dispatchable generation remains, at higher cost per megawatt-hour, passed through to every ratepayer.
Source: U.S. Energy Information Administration, "U.S. Northeast Is Relying Less on Electricity Imports from Canada," September 2025
The policy is mandating new load while retiring supply.
The CLCPA's electrification requirements are not external load arriving despite the policy. They are load the policy itself is requiring the grid to absorb. Starting January 1, 2026, the All-Electric Buildings Act prohibits fossil fuel heating and appliances in most new construction under seven stories statewide. That restriction extends to all new construction by 2029. Starting in 2030, homeowners may only replace broken furnaces or water heaters with electric alternatives. Gas appliance sales end statewide in 2035, as do sales of new gasoline vehicles. NYISO's Power Trends 2025 quantified the combined effect: by 2030, demand could increase by an additional 1,600 MW to nearly 4,000 MW from building electrification and new large loads alone. That is not a projection of organic growth. It is the grid operator's forecast of what the state's own statutes will require it to absorb.
Source: New York State Fire Prevention and Building Code Council, All-Electric Buildings Act, approved July 25, 2025; CLCPA Scoping Plan; NYISO, Power Trends 2025
The structural consequence is more severe than the megawatt figure suggests. New York's grid currently peaks in summer, during hot afternoons. NYISO's Power Trends 2025 projects that electrification of space heating and transportation will shift New York to a winter-peaking system by the 2040s. Winter is when natural gas supply constraints are most severe, when solar generation is at its annual minimum, and when peak demand will increasingly coincide with the lowest output hours for every weather-dependent resource on the grid.
Source: NYISO, Power Trends 2025; Regional Plan Association, "The State of the Grid in New York," December 2025
The data center moratorium addresses the smaller part of the problem.
At the end of 2024, NYISO's interconnection queue, which is the waiting list of projects seeking permission to connect to the grid, held roughly 4,000 MW of large-load projects. By January 2026 that had grown to approximately 12,000 MW, most seeking service before 2031. Senate Bill S9144, introduced February 2026, would apply to new facilities over 20 MW within state borders. But the electrification load is already written into state law and accumulates whether or not a single new data center is permitted. Beyond that, the moratorium's geographic logic does not hold at the grid level. New York, New Jersey, and the broader Northeast operate as a single interconnected regional system. NYISO connects to ISO New England, which manages the six New England states, and to PJM, which manages the grid for New Jersey, Pennsylvania, and 11 additional states. A data center denied a permit in Westchester is built in northern New Jersey and draws from the same infrastructure. NYISO's Kevin Lanahan, vice president of external affairs, stated it directly: increased energy consumption anywhere in the system can impact the price and the reliability for city residents because the grids are inseparable.
Source: NYISO, 2025-2034 Comprehensive Reliability Plan; Spectrum News, February 2026; The City, March 2026
The 2040 deadline eliminated the market that would solve it.
Building a new dispatchable generating asset requires a capital investment with a useful life of 30 to 40 years. An investor committing that capital today is underwriting a revenue stream extending to approximately 2055 to 2065. The CLCPA mandates zero emissions from the electricity sector by 2040. That mandate carries legal force. The state demonstrated its intent in October 2021, when the Department of Environmental Conservation denied air permits for two proposed gas-fired generation projects, the NRG Astoria repowering in Queens and the Danskammer plant in the Hudson Valley, on the grounds that new fossil generation was incompatible with CLCPA targets. A private developer cannot finance a 35-year asset in a market that has legally committed to eliminating that asset's revenue stream within 15 years. No rational investor builds a power plant whose income the law has already scheduled for termination. The Independent Power Producers of New York, the trade group representing the state's merchant generators, identified the consequence directly: the central unresolved question for the sector was what dispatchable technology could comply with the 2040 requirement, and without that answer, private capital has no path to commit.
Source: City and State New York, March 2022; S&P Global Commodity Insights, citing IPPNY director of market policy, February 2021
The technology that would fill the gap is what NYISO calls Dispatchable Emissions-Free Resources: generation callable on demand like a gas plant, but producing zero emissions like a solar panel. Long-duration battery storage, green hydrogen combustion, and advanced nuclear are the leading candidates. None exists at commercial scale. NYISO's 2023 Comprehensive Reliability Plan estimated that between 27 GW and 45 GW of this zero-emissions dispatchable capacity would be needed by 2040. New York's entire current generating fleet, all fuel types combined, is roughly 40 GW. NYISO's Power Trends 2025 stated it plainly: without investment in new dispatchable resources, grid operators are increasingly relying on an aging fleet of fossil generation.
Source: NYISO, 2023-2032 Comprehensive Reliability Plan; NYISO, Power Trends 2025
The 2040 law was written to eliminate an emissions problem. It simultaneously eliminated the investment case for building the replacement, mandated new electric loads that will stress the grid further, and created a winter-peaking demand profile that is structurally hardest to serve with renewable generation. The $630 annual premium a typical New York household pays over the national average is, in part, the cost of running a reliability system the market has no reason to modernize and the law has inadvertently made it impossible to replace.
|
NYISO's CEO Richard Dewey stated publicly in 2024: "I have grave concerns about the winters of 2030, 2031 and 2032." |
Source: Smarter NY Energy, citing NYISO CEO Richard Dewey, 2024
That statement did not come from a fossil fuel advocate. It came from the operator of the grid.
Three Things the Data Gets Wrong
"Renewable capacity growth means the grid is getting more reliable."
Reliability is determined by the worst hours of the worst days: still winter nights, polar vortex events, sustained heat waves with no wind. What matters on those hours is dispatchable capacity, the generation that shows up regardless of weather. New York's dispatchable margin has shrunk by more than 2,000 MW since 2019. As building and vehicle electrification shifts New York to a winter-peaking system, those worst hours will arrive with increasing frequency on the days the renewable capacity number can help the least.
"Banning data centers in New York protects the grid."
It addresses the discretionary portion of New York's load problem. The electrification of heating and transportation, mandated by the same CLCPA that motivates the data center concern, adds 1,600 to 4,000 MW of new load by 2030 regardless of whether a single new data center is permitted. Data centers denied permits here are built in adjacent states on the same interconnected regional grid, so the load materializes on shared infrastructure either way. The moratorium is not the wrong idea. It is aimed at the smaller part of a two-part problem.
"High electricity prices are the cost of the transition and will fall once renewables are built out."
New York's rates are 49% above the national average and the gap is widening. Adding renewable megawatts does not resolve the forces driving prices up: aging dispatchable generation cycling inefficiently to fill reliability gaps, lost Canadian hydro imports, mandatory electrification load without confirmed dispatchable replacement, and no private investment case for building modern thermal assets. Prices fall when zero-emissions dispatchable generation arrives at scale and displaces the aging fallback fleet. That generation does not yet exist commercially, the 2040 mandate removed the incentive to build its predecessor, and the electrification mandates are adding load to the same stressed system on a fixed statutory timeline.
The Tools
NYISO publishes three documents that together tell the real story of New York's grid. The quarterly Short-Term Assessment of Reliability tracks reliability margins region by region, updated every three months. The annual Power Trends report shows the trajectory of dispatchable capacity additions versus retirements and electrification load forecasts each year. The biennial Comprehensive Reliability Plan models reliability margins under multiple demand and retirement scenarios out ten years.
Four numbers to track:
|
Reliability margin for New York City under expected weather conditions. The early warning indicator for shortfalls. |
|
Ratio of dispatchable deactivations to total capacity additions. Shows whether the supply gap is widening or closing. |
|
Large-load interconnection queue size versus new generation entering service. Tracks whether supply is keeping pace with mandated demand growth. |
|
Winter versus summer peak demand trajectory. Shows whether electrification is shifting the grid toward its most vulnerable demand profile faster than zero-emissions dispatchable resources are arriving. |
All three reports are free, primary-sourced, and published by the entity with legal responsibility for keeping the lights on.
nyiso.com/power-trends links to the STAR and Comprehensive Reliability Plan from the same page.
The Bottom Line
New York's grid faces a structural problem that the installed renewable capacity number cannot measure and the current policy framework cannot resolve through private investment. Dispatchable capacity is retiring faster than any replacement is being built. The law that mandates zero emissions by 2040 also eliminated the financial case for building the transitional assets that would get there reliably. The same legislation is mandating new electric loads through building and vehicle electrification that will shift peak demand to winter nights: the hours when solar produces nothing, wind underperforms, and aging gas infrastructure is most constrained. The $630 annual premium New York households pay over the national average is not the cost of a transition in progress. It is the cost of a transition whose supply-side investment market closed before the replacement technology arrived.
NYISO's own planning documents project reliability deficiencies as severe as 10,000 MW under conservative scenarios by 2034. The technology needed to avoid those deficiencies does not currently exist at commercial scale. The timeline to develop it likely exceeds the policy mandate that would require it.
|
That leaves one question worth sitting with: if the dispatchable replacement cannot be built by private capital and does not yet exist at commercial scale, what is the actual plan for the winters of 2030, 2031, and 2032? |
Sources
NYISO, 2025-2034 Comprehensive Reliability Plan (November 2025) · NYISO, 2023-2032 Comprehensive Reliability Plan · NYISO, Short-Term Assessment of Reliability Q3 2025 (October 2025) · NYISO, State of the Grid Webinar (January 2023) · NYISO, Winter 2023-2024 Reliability Outlook · NYISO, Summer 2024 Reliability Press Release · NYISO, Power Trends 2025 · Draft 2025 New York State Energy Plan, NYSERDA, Electricity Chapter · New York State Fire Prevention and Building Code Council, All-Electric Buildings Act, approved July 25, 2025 · CLCPA Scoping Plan · Regional Plan Association, The State of the Grid in New York, December 2025 · U.S. Energy Information Administration, New York Electricity Profile 2024 · U.S. Energy Information Administration, New York Electricity Profile 2020 · U.S. Energy Information Administration, U.S. Northeast Is Relying Less on Electricity Imports from Canada, September 2025 · U.S. Bureau of Labor Statistics, New York-Newark-Jersey City Average Energy Prices, December 2024 · S&P Global Commodity Insights, February 2021 · City and State New York, March 2022 · The City NYC, March 2026 · Spectrum News, February 2026 · Smarter NY Energy, citing NYISO CEO Richard Dewey, 2024